No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Intelligent tutoring systems: a systematic review of characteristics, applications, and evaluation methods
Abstract
With the rapid growth of technology, computer learning has become increasingly integrated with artificial intelligence techniques in order to develop more personalized educational systems. These systems are known as Intelligent Tutoring systems (ITSs). This paper focused on the variant characteristics of ITSs developed across different educational fields. The original studies from 2007 to 2017 were extracted from the PubMed, ProQuest, Scopus, Google scholar, Embase, Cochrane, and Web of Science databases. Finally, 53 papers were included in the study based on inclusion criteria. The educational fields in the ITSs were mainly computer sciences (37.73%). Action-condition rule-based reasoning, data mining, and Bayesian network with 33.96%, 22.64%, and 20.75% frequency respectively, were the most frequent artificial intelligent techniques applied in the ITSs. These techniques enable ITSs to deliver adaptive guidance and instruction, evaluate learners, define and update the learner’s model, and classify or cluster learners. Specifically, the performance of the system, learner’s performance, and experiences were used for evaluation of ITSs. Most ITSs were designed for web user interfaces. Although these systems could facilitate reasoning in the learning process, these systems have rarely been applied in experimental courses including problem-solving, decision-making in physics, chemistry, and clinical fields. Due to the important role of a cell phone in facilitating personalized learning and given the low rate of using mobile-based ITSs, this study has recommended the development and evaluation of mobile-based ITSs.
... In their comprehensive review, Desmarais and Baker [80] recognised and presented BNs as the most general approach to modelling learner skills, highlighting their versatility and effectiveness in educational contexts. Mousavinasab et al. [197] systematically reviewed 53 papers about IASs applications from 2007 to 2017, exploring the characteristics, applications, and evaluation methods, and found that a significant proportion of the reviewed papers employed BN techniques, highlighting their widespread adoption and success in modelling learner knowledge. More recent works continue to support the use of BNs. ...
... Our implementation exploits the noisy gates mechanism to simplify parameter elicitation, making the system more efficient while preserving accuracy in assessing AT skills. Unlike conventional methods that assign a single score per student-task, our approach uses posterior probabilities to construct a comprehensive learner model that provides a more detailed understanding of students' competence profiles, highlighting their proficiency across various skill levels [197,249,340,341]. ...
... Addressing these differences is essential for ensuring that every student has the opportunity to succeed and develop their AT skills. This perspective aligns with research suggesting that tailored, personalised and adaptive educational approaches that address individual needs and characteristics can enhance learning experiences [80,133,191,197,278,314]. By adapting educational practices to address diverse learning preferences and abilities, educators can create more inclusive and supportive environments that foster success for all students. ...
The rapid digitalisation of contemporary society has profoundly impacted various facets of our lives, including healthcare, communication, business, and education. The ability to engage with new technologies and solve problems has become crucial, making CT skills, such as pattern recognition, decomposition, and algorithm design, essential competencies. In response, Switzerland is conducting research and initiatives to integrate CT into its educational system. This study aims to develop a comprehensive framework for large-scale assessment of CT skills, particularly focusing on AT, the ability to design algorithms. To achieve this, we first developed a competence model capturing the situated and developmental nature of CT, guiding the design of activities tailored to cognitive abilities, age, and context. This framework clarifies how activity characteristics influence CT development and how to assess these competencies. Additionally, we developed an activity for large-scale assessment of AT skills, offered in two variants: one based on non-digital artefacts (unplugged) and manual expert assessment, and the other based on digital artefacts (virtual) and automatic assessment. To provide a more comprehensive evaluation of students' competencies, we developed an IAS based on BNs with noisy gates, which offers real-time probabilistic assessment for each skill rather than a single overall score. The results indicate that the proposed instrument can measure AT competencies across different age groups and educational contexts in Switzerland, demonstrating its applicability for large-scale use. AT competencies exhibit a progressive development, with no overall gender differences, though variations are observed at the school level, significantly influenced by the artefact-based environment and its context, underscoring the importance of creating accessible and adaptable assessment tools.
... To tailor instructional strategies, activities, and material to each learner, focus is placed on the development and integration of intelligent tutoring systems [11,12]. Intelligent tutoring systems can identify, track, and monitor learners' cognitive, behavioral, and psychological states [13,14]. ...
... The integration of intelligent tutoring systems within augmented reality and virtual reality environments shows great potential to improve the educational domain [42]. Although there have been several review and meta-analysis studies that have explored the individual use of augmented reality [43][44][45][46][47], virtual reality [28,[48][49][50], and intelligent tutoring systems [11,15,25,51] in education, to the best of our knowledge, there has not been any systematic literature review study that has focused on the integration of intelligent tutoring systems in extended reality environments. As these technologies are rapidly advancing and since their combination presents several benefits and new opportunities, it is important to examine their convergence, as recent studies have shown [31]. ...
... Therefore, studies have put emphasis on examining how recent technologies, such as the metaverse and extended reality technologies, are being adopted and used in education [93]. Moreover, studies have highlighted the educational benefits that can be yielded through the adoption of intelligent tutoring systems [11,15,25,51], augmented reality [43][44][45][46][47], and virtual reality [28,[48][49][50] in teaching and learning activities in both traditional and vocational education [94]. The integration of intelligent tutoring systems into augmented reality and virtual reality environments can result in the development of social and affective entities that can support the education process, promote collaborative learning and self-directed learning, and enhance students' learning outcomes [31]. ...
Given the advancements in artificial intelligence and extended reality technologies, this study aims to examine the integration of intelligent tutoring systems into augmented reality and virtual reality environments through a systematic literature review. Following the PRISMA framework, 32 related theoretical, showcase, and case studies published during the period of 2015–2024 are examined. Based on the results, this combination of technologies emerged as an effective educational means that can support both students and teachers, promote lifelong learning, and support face-to-face, blended, and online learning across educational levels and in the workplace. These systems offered immersive, realistic, and interactive learning environments and personalized learning experiences. Additionally, they could identify, monitor, and analyze students’ characteristics, performance, preferences, and motivational, cognitive, and psychological states. These systems could also adapt the learning content, resources, activities, and assessment according to students’ needs and make suitable recommendations. Their ability to offer tailored and real-time feedback, guidance, analytics, and evaluation was highlighted. Additionally, it was revealed that these systems offer meaningful learning experiences and enhance cognitive, affective, psychomotor, and embodied learning through self-directed learning, collaborative learning, personalized learning, and experiential learning approaches. Regarding learning benefits, students who learnt using this combination demonstrated increased engagement, motivation, confidence, immersion, and enjoyment. The students also reported better learning outcomes and academic performance, enhanced knowledge and skills, and improved information understanding and recall. This study also presents the main topics and areas examined, goes over the existing challenges, and suggests future research directions. Finally, the study emphasizes the importance of capitalizing on both human intelligence and machine intelligence to support students, meet their needs, and provide them with quality education and lifelong learning opportunities.
... Advances in adaptive e-learning systems, such as intelligent tutoring systems (ITS), can help address these issues. ITS use artificial intelligence to adapt instruction based on learner characteristics, such as prior knowledge, problem-solving strategies, and emotional states, to optimize individual learning gains [18] [19]. These systems build a rich dataset of learner interactions that supports learning analytics, enabling instructors to make data-driven decisions [20]. ...
... As software engineering instructors working with an illdefined domain, we identified a need for a tool to support us and our ever-growing student pool during the teaching and learning process. We derive requirements that the tool should fulfill based on our experience, our previous research [22], and studies exploring instructors' needs for learning analytics in classrooms [19] [21]. We define the requirements in Table II and explain how they contribute to the stakeholders' goals. ...
... To design a viable solution for our context, we examined the literature on adaptive e-learning systems and quickly converged on intelligent tutoring systems. An ITS is a type of learning technology that models principles of efficient instruction (i.e., instructor model) to adapt educational content (i.e., domain model) to the learner's specificities (i.e., learner model), such as prior knowledge [19]. The ITS's interface model handles the interaction between a learner and the ITS and defines learning instruments that present content and handle user interaction [19]. ...
Software engineers are tasked with writing functionally correct code of high quality. Maintainability is a crucial code quality attribute that determines the ease of analyzing, modifying, reusing, and testing a software component. This quality attribute significantly affects the software's lifetime cost, contributing to developer productivity and other quality attributes. Consequently, academia and industry emphasize the need to train software engineers to build maintainable software code. Unfortunately, code maintainability is an ill-defined domain and is challenging to teach and learn. This problem is aggravated by a rising number of software engineering students and a lack of capable instructors. Existing instructors rely on scalable one-sizefits-all teaching methods that are ineffective. Advances in elearning technologies can alleviate these issues. Our primary contribution is the design of a novel assessment item type, the maintainability challenge. It integrates into the standard intelligent tutoring system (ITS) architecture to develop skills for analyzing and refactoring high-level code maintainability issues. Our secondary contributions include the code maintainability knowledge component model and an implementation of an ITS that supports the maintainability challenge for the C# programming language. We designed, developed, and evaluated the ITS over two years of working with undergraduate students using a mixed-method approach anchored in design science. The empirical evaluations culminated with a field study with 59 undergraduate students. We report on the evaluation results that showcase the utility of our contributions. Our contributions support software engineering instructors in developing the code maintainability skills of their students at scale.
... The research results indicated that although completed fish protection may result in loss of fishing areas, the combination of cost layer analysis could balance conservation and fishing impacts, and the combination of regional and national analysis could help determine the optimal reserve allocation [16].Tedim F et al. proposed a research method combining online questionnaires and Delphi surveys to fill the knowledge gap on the role of spatial planning in reducing wildfire hazards. The results of the research showed that Portugal had major weaknesses in the application of spatial planning legislation and policies for the management of forest fires, in particular the constraints encountered in the incorporation of national hazard maps into overall planning.It reflected the mismatch between the high inter-annual variability of fire disasters and the long-term definition of urban spatial planning [17].A. S. Akopov et al. proposed a new multi-agent genetic algorithm Magamo to solve multi-objective optimization problems. The algorithm is based on synchronous agent dynamic interactions and has the characteristic of intelligent agent decomposition of size space to form population evolution. ...
The rational planning of territorial space is related to the speed of economic development and the protection of ecology. To realize effective spatial layout planning, there is an urgent need for an more advanced integrated territorial space layout method. For additional spatial layout design, the study suggests a multi-objective genetic algorithm based on spatial data prediction and coupled with land simulation modeling in the context of big data and machine learning development. The innovation of the research lies in the optimization of the multi-objective genetic algorithm by improving the crossover and mutation process of genetic operators, enhancing the adaptability of the algorithm in dynamic environments, and improving the prediction accuracy of the spatial quantity structure of the national territory. The outcomes revealed that this approach was applied to the prediction of populations of different sizes 400 and 600 with an average accuracy of 67%, which was 17% higher than that of the traditional genetic algorithm. The difference between the predicted and true values of future spatial population by the proposed algorithm was less than 2%, which was 56.5% lower than the other two prediction algorithms on average. The proposed land simulation model reached the highest accuracy at 100 iterations with an average fitness of 2.4×1011, which was 13% and 27% higher than the other two traditional neural network algorithms, respectively. In the two simple functions of f1 and f2, the highest convergence accuracy reached 10-30 and 10-10, respectively. In the two more complicated functions of f3 and f4, the optimal solutions were approximated in the ranges of [102,1012] and [10-2,106] without significant fluctuations. Therefore, the proposed algorithm can effectively predict the number of territorial space in macro and micro simulation, and has high feasibility and accuracy. This provides a reliable basis for the government to carry out land resource planning, and promotes the sustainable development of ecology and economy.
... Invariant across these various educational tutoring contexts is the use of adaptive teaching strategies tailored to current states of student understanding and empathizing with their socio-emotional states by providing encouragement and motivation (Du Boulay & Luckin, 2016). In AIED systems, delivering such just-in-time adaptations with high confidence so as not to unintentionally harm student learning is a critical design principle (Mousavinasab et al., 2021). ...
I nudge reevaluation of the idea that artificial intelligence for education (AIED) is merely about using artificial intelligence (AI) tools to automate understanding and responding to learning processes. Instead, I advocate for a human-centered approach to AIED that emphasizes the importance of personal connections, relationship-building, and scaffolding that goes beyond simplifying tasks to push students in their critical thinking. This approach calls for curating multimodal data from ecologically valid learning settings to train AIED systems, and maintaining flexibility in expectations around rational learner behavior when analyzing data from such systems. Given that the definition of good AIED is often discipline-specific and influenced by the underlying pedagogical models of student learning, the article calls on learning sciences researchers to integrate their complementary yet often competing theoretical lenses in rigorously studying AI-supported learning phenomena at scale.
... According to their work, ITS can successfully deliver a personalized educational experience with high educational standards. Building on this foundation, Mousavinasab et al. (2018) systematically reviewed ITS characteristics, applications, and evaluation methods, identifying key success factors in their implementation within higher education settings. ...
Purpose
This study aims to analyze the most frequently discussed topics in the scientific discourse on artificial intelligence (AI) in higher education using Natural Language Processing (NLP) techniques.
Design/methodology/approach
This paper analyzes 52 peer-reviewed articles published between 2017 and 2024, utilizing NLP techniques to identify prevalent unigrams, bigrams and trigrams related to AI in higher education.
Findings
The analysis identifies an emerging concern with utilizing AI tools to enhance educational processes, with “Higher education,” “artificial intelligence” and “generative AI” becoming ubiquitous terms in use. LLM and ChatGPT represent types of technology that evoke potential for personalized learning and enhanced practice in instruction.
Research limitations/implications
In review studies, samples with a post-secondary educational background usually restrict generalizability to school environments. Future studies can examine the long-term consequences of AI technology in extended academic environments, longitudinal studies and educational environments.
Practical implications
The frequency patterns from our analysis offer essential insights for educators and administrators regarding curriculum development and teaching practices. The high occurrence of terms like “artificial intelligence” (1,193 times) and “higher education” (824 times) highlights the need for incorporating AI literacy into curricula. This integration should include guidelines for responsible AI use and training programs for faculty. The frequent mentions of “teaching learning” (226 times) and “AI education” (319 times) highlight important implications for teaching practices. Educational institutions must establish frameworks that blend traditional methods with AI-enhanced strategies, including assessment plans that consider AI tools while upholding academic integrity. Additionally, institutions should prioritize investment in AI infrastructure and support systems.
Social implications
Our findings highlight important societal implications beyond education. The frequency analysis reveals concerns about educational equity, including disparities in access to AI-enhanced education, digital literacy gaps and economic barriers to adopting AI tools. Addressing these issues is vital to prevent the worsening of social inequalities. Additionally, our results emphasize the need for workforce development. Educational institutions should focus on equipping students with the AI competencies that employers demand and bridging the gap between academic training and industry needs. The policy implications of our findings are equally significant. Our analysis suggests the need for educational policies that address AI integration while establishing clear guidelines for ethical AI use in academic settings. These policies should include standards for AI tool evaluation and implementation to guide institutions' adoption decisions. The economic impact of these developments is also noteworthy, as our results indicate the potential for enhanced workforce preparedness through AI-integrated education, improved educational efficiency through automation and new opportunities for educational technology development.
Originality/value
This study contributes to the field by providing an overview of prominent trends in AI within higher education, discussing the practical application, future research opportunities, and challenges associated with the responsible and effective use of AI in education.
... The technologies mentioned in the Table 1 were developed in US and used in education in general and in teaching English Language specifically. Mousavinasab et al. (2018) studied the effectiveness of intelligent tutoring systems helps to provide real time feedback and hint; it is also effective in assisting learners to understand things, solve problems and retentions of knowledge According to recent studies like Chen et al. (2023) and Wang et al. (2022), students had favorable opinions about utilizing AI tools to learn languages. AI technologies also assist students in becoming self-sufficient learners, according to Srinivasan (2022). ...
This study examines the awareness and integration of Artificial Intelligence (AI) technology among English as a Foreign Language (EFL) teachers at Ambo University, highlighting its significant impact on English language instruction and general education through tasks such as grading, lesson planning, and progress tracking. Utilizing a descriptive survey research design, the research involved 20 English teachers, employing questionnaires and interviews for data collection. Quantitative data were analyzed using SPSS version 23, while qualitative data were assessed narratively. Findings revealed that the teachers were largely unaware of many advanced AI tools, such as AlphaCode, YouChat, and DALL-E, with some familiarity with ChatGPT and GrammarlyGo. Despite their interest in learning how to incorporate AI into their teaching, teachers expressed a lack of confidence and reported insufficient access to training and resources needed for effective integration. Moreover, they unanimously agreed on the potential of AI technology to enhance the quality of English language instruction. To foster an interactive and student-centered teaching approach, it is essential for higher education institutions in Ethiopia, particularly Ambo University, to establish platforms for AI integration and provide relevant training for faculty.
... The second important method for solving personalized and adaptive learning problems is formed through learning with the help of artificial intelligence (Khotimah & Priyanti, 2023). Intelligent tutoring systems have been developed for about fifteen years and have hundreds of specific applications, approaches and methods that have proven practical (Mousavinasab et al., 2018). ...
Utilizing advanced learning technologies, such as data analysis and artificial intelligence, teachers can identify student learning patterns, anticipate possible difficulties, and provide specific additional support. For example, by analyzing students' engagement with online learning platforms, teachers can tailor interventions to address individual learning needs, leading to more effective learning outcomes. Moreover, personalized learning in a digital environment goes beyond the delivery of content; it involves fostering 21st century skills such as critical thinking, communication, collaboration, innovation, and problem-solving. Research has shown that integrating technology into project-based learning activities can significantly enhance students' ability to develop these skills. By optimizing the potential of personalized learning approaches in a digital environment, educators can ensure that every student has an equal opportunity to develop the skills necessary to thrive in an ever-changing world.
At its core, AI in education aims to personalize learning, streamline administrative tasks, and foster more efficient, engaging educational interactions. Technology's ability to process vast amounts of data, recognize patterns, and generate insights can significantly improve both student outcomes and teacher effectiveness. By analyzing students' interactions, progress, and assessments, AI can identify strengths and weaknesses, adapting lessons and resources to better suit each learner. For instance, intelligent tutoring systems use algorithms to provide customized feedback and support, enhancing students' grasp of subjects and addressing gaps in their understanding more effectively than generic instruction.AI also contributes significantly to automating administrative tasks, which can alleviate some of the burdens on educators. Furthermore, AI fosters greater engagement and interactivity in the classroom. Educational tools powered by AI, such as virtual assistants and chatbots immediate help and resources to students, facilitating a more interactive learning experience.
Artificial intelligence continues to advance in many spheres, including education. The integration of Artificial intelligence into educational practices has evidently transformed teaching and learning processes over the decades. In this context, this chapter seeks to explore the dynamic landscape of AI in educational practices, focusing on current innovative applications, effective pedagogical strategies, and the challenges educators face. By reviewing recent studies published within the last six years, this chapter aims to provide a comprehensive understanding of how AI can be effectively integrated into educational settings.
Education is utilizing generative artificial intelligence more and more, especially with models like GPT, BERT, and other neural networks. Through the automatic generation of information, individualized learning resources, and responses to student inquiries, these technologies are transforming traditional teaching methods, course design, and interactions with learners. Despite the fact that these tools have the potential to drastically change the nature of education, it is crucial to comprehend their existing uses, advantages, and drawbacks. The purpose of this chapter is to present a thorough analysis of current scholarly research on the application of generative AI in education across all academic levels.
The chapter focuses on the examination of the strategic role of artificial intelligence (AI) and simulation tools in the context of higher education, and most importantly, how these two elements can substantially contribute towards enhancing effective learning and engagement. The chapter has identified the increasing role of AI in contemporary societies, with its impact in education (AIEd) to have gained an increasing role as a result of rapid growth of technological tools, and the need for higher institutions to incorporate such technological developments into their educational practices. This requires universities to identify the emerging opportunities and focus on the development of their technological infrastructure both tangible (equipment) and intangible (human). AI contributes towards the development of smart personalised education, therefore enabling both educators and students to fully develop their unique skills, leading towards improved educational and learning results for all parties involved in the process.
The rapid advancements in modern technologies have opened new possibilities for enhancing educational experiences for students with Special Educational Needs and Disabilities (SEND). This paper conducts a systematic review of 139 studies on the integration of AI, VR, and LLMs in Special Education. Using a deductive thematic analysis framework, it identifies key themes and challenges to synthesize the current state of knowledge and propose future research directions. The findings underscore the transformative potential of AI and Immersive Technologies in fostering personalized learning, improving social engagement, and advancing cognitive development among SEND students. Additionally, current SEN methodologies and practices are defined, teachers'attitudes toward inclusion and technology adoption, and the prevailing technological tools utilized, based on various syndromes and disorders. Challenges such as ethical considerations, accessibility barriers, and resource limitations are also discussed. Moreover, this study explores tailored technological tools specifically designed to meet the unique needs of these students. The paper concludes with limitations, recommendations for cross-sector collaboration and inclusive policies to ensure these technologies are effectively utilized to prioritize and enhance learning experiences for SEND students, as well as future directions.
Artificial intelligence (AI), in this data-driven digital world, is revolutionizing modern life with far-reaching implications for individuals, teams, organizations, and society. Using comments from 126 undergraduate students in South Florida, this theoretical paper highlights concepts and concerns regarding AI challenges related to cheating, plagiarizing, and biased information. The worries about the impact of AI are analogous to what the internet was three decades ago. People were using the internet as it was being developed, fine-tuned, and improved; it felt like walking over a long and tall bridge as it was being built, and the same is true for the growth of AI. Drawing parallels with the internet’s transformative impact over the past three decades, this paper emphasizes that AI is poised to drive similar positive changes, fostering increased productivity, transparency, accountability, and ethics, but at a much faster pace. In the meantime, due to the availability of data and digital content, the virtual world increased misinformation, disinformation, bias, and prejudiced speech, which AI can easily exacerbate. While AI adoption may cause process-related disruptions, its integration into everyone’s daily life is inevitable. As a natural extension of the information superhighway, AI will usher in a new wave of innovation, ultimately and perpetually transforming the fabric of our personal and professional lives. Drawing on literature and recent trends forecasted by experts, this theoretical manuscript provides an overview of AI uses, its history, challenges, and ethical implications for us all. The conceptual paper ends with recommendations for educators, managers, entrepreneurs, and human resources professionals to create awareness regarding the benefits of this new endemic technology, to ease people’s anxiety, and to reduce or mitigate hallucinations so AI tools can be used to enhance everyone’s effectiveness and efficiency.
Current educational trends leverage artificial intelligence (AI) to provide high-quality teaching and enhance students’ learning competitiveness. This study aimed to evaluate the acceptance of artificial intelligence generated content (AIGC) for assisted learning and design creation among art and design students. Based on an extended technology acceptance model (ETAM), this study explored how external variables influence perceived usefulness (PU) and perceived ease of use (PEOU), which in turn affect attitude towards use (ATT) and behavioral intention (BI). Data were collected from 382 students via a questionnaire survey and analyzed using a structural equation model. The results confirmed 12 out of the 14 hypotheses. Among them, facility condition (FC), output quality (OQ), task-technology fit (TTF), and hedonic motivation (HM) positively influenced PU and PEOU, whereas AI anxiety (AIA) negatively affected PU and PEOU. ATT had a significant positive effect on BI. This study provides theoretical support and practical insights for promoting AIGC applications, advancing sustainable education, and optimizing user engagement.
In this chapter, we look into the historical aspects and current trends of adaptation and personalization in human-centric AI. First, we distinguish the terms adaptation and personalization. Then we proceed on providing historical context on how adaptation evolved and personalization became a-kind-of adaptation. We further present the current trends, especially those based on human-centric aspects, which we refer to as theory-driven personalization. Finally, we critically assess the threats to adaptive and personalized human-centric AI and provide an outlook into the future.
In this paper, we use the improved Apriori analysis method to construct an information management system for evaluating the teaching quality of teachers in higher vocational colleges and universities. Firstly, data such as students’ evaluations and teachers’ personal information are preprocessed, and then the improved Apriori algorithm is used to mine the data. Finally, for the multiple meaningful strong association rules mined out by the algorithm, the constitutive relationships between teachers’ age and title, age and education are found out to realize the improvement of teaching quality. The empirical analysis shows that the teachers with the title of associate professor are in the age of 50 years and above, and all of them have a bachelor’s degree. Most of the teachers in graduate school are above 35 years old. There is a strong correlation rule relationship between teachers’ education, teachers’ attitude and teachers’ age. Therefore, it is important to bring in more highly educated and highly titled teachers so as to improve the teaching level of higher vocational colleges and universities. The survey analysis shows that about 60% of the lecturers and evaluating teachers are supportive of this paper’s evaluation system to improve teaching quality.
With increasing popularity of artificial intelligence
(AI) in the education industry, intelligent tutoring system (ITS)
powered by AI have been widely adopted to optimize the
learning experience. However, the relationship between
students’ engagement level of Generative AI (GenAI) and their
academic performance is still under exploration. Also, current
popular GenAI products like ChatGPT suffer from the
hallucination problem, which includes factuality, faithfulness,
and maliciousness issues in the generated answer. This paper
presents GPTutor, an ITS leveraging GenAI to support
students' learning processes. GPTutor integrates a Retrieval-
Augmented Generation (RAG) pipeline to deliver actual and
contextually rich answers aligned to student questions and
intended learning outcomes (ILO). A pilot evaluation involving
undergraduate and postgraduate students assessed the system’s
association with user experience, engagement, and academic
performance. Results demonstrated that students generally
recognize the effectiveness of GPTutor. Some students also
provided insightful feedback on the benefits of GPTutor in
improving learning efficiency and some limitations to be
addressed. Notably, students with higher engagement levels
showed significantly better academic performance on the final
exam. This study proposed GPTutor to provide an interactive
and knowledge-grounded learning experience and showed the
strong association between students’ engagement in GPTutor
and academic performance.
This chapter examines the entry of AI into classroom management through Intelligent Tutoring Systems. ITSs enhance customized learning through adaptive learning as it customizes the educational content based on an individuals requirement thus enhancing the interactive learning experience for the learner and making it more productive. AI can be utilized by the ITS to monitor student behavior and performance data, identifying knowledge gaps and providing personalized intervention recommendations. The chapter explores ethical questions related to educational AI, such as data privacy and understood algorithms, aiming to develop learners' independence and teach teachers when to observe and intervene purposefully in educational AI. The study presents case studies and empirical evidence highlighting the transformative power of AI-enabled ITS in improving learning outcomes, equity, and inclusive styles, arguing that strategic adoption of AI technologies can lead to smarter, responsive classes.
This work in progress paper investigates the use of AI chatbots by undergraduate computer science students learning English as a second language in Japan. The study employs a mixed-methods approach, combining qualitative surveys and interviews with quantitative clustering analysis. The objectives are to identify the types of AI chatbots used, determine their usage patterns, and explore the benefits and challenges associated with their use in language learning. The qualitative data collection (surveys, n=96) has been completed, while interviews and quantitative analysis are ongoing. The study aims to identify distinct clusters of AI chatbot users and their characteristics, highlight challenges, and contribute to the growing knowledge on AI application in STEM students studying a second language. Future research should explore long-term effects, optimal balance between AI-assisted and human-led instruction, and guidelines for integrating AI chatbots in language learning curricula.
With the widespread adoption of the internet, intelligent tutoring systems have experienced rapid advancements in education. Knowledge tracing (KT) aims to predict students’ future performance by modeling their behavioral sequences. While existing models have made notable progress in some areas, challenges remain in fully uncovering the higher-order relationships among exercises, concepts, and students. To address these limitations and effectively capture the intricate dependencies between learning behaviors and exercises, this paper introduces a novel approach: Dual-Channel Knowledge Tracing with Self-Supervised Contrastive and Directed Interaction Learning (SCDI). This method enhances predictive accuracy by leveraging graph-structured embeddings of interaction data, offering a more comprehensive analysis of the learning process. The SCDI model consists of two distinct channels. The first channel constructs a heterogeneous graph with student, exercise, and concept nodes, using a Meta-path Aggregated Graph Neural Network (MAGNN) to extract high-order information from exercise and concept nodes. It also incorporates a self-supervised graph contrastive learning mechanism to improve the quality of node embeddings. The second channel builds a directed graph from interaction sequences, using Graph Attention Convolution (GATC) to capture dynamic relationships and model how interactions evolve over time. Features from both channels are integrated and passed to a downstream KT model to predict students’ future performance. Experimental evaluations on four datasets demonstrate that SCDI significantly enhances the accuracy of downstream KT models, establishing its effectiveness and robustness.
The purpose of this chapter is to provide a brief historical overview of the technological evolution of artificial intelligence (AI) (in layman’s terms) and its use in strategic communication to offer context for the rest of the book.
Focusing on the identification of important themes, new trends, and essential elements that define the subject, this report offers a thorough bibliometric analysis of the research terrain surrounding Smart Learning Environments (SLEs). Three main clusters were found by means of a thorough investigation of keyword co-occurrence and network visualization: learner-centric technologies and methods, artificial intelligence and adaptive learning systems, and assessment and learning results. Driven by developments in mobile learning, virtual reality, artificial intelligence, and learning analytics, the study exposes a growing focus on tailored, immersive, and data-driven educational experiences. The study also looks at the elements affecting SLE adoption: technology infrastructure, instructional efficacy, personal attitudes, and organizational support. The results underline the dynamic and changing character of SLE research and provide understanding of how these settings may be efficiently used into teaching strategies to improve outcomes for learning.
Ensuring inclusion and equity in the rapidly improving educational landscape is a major challenge. Integrating artificial intelligence (AI) offers transformational potential to address this gap through personalized and personalized learning experiences This study explores the interface between AI technology and higher education in 2010, and highlights how AI can enhance inclusion and equity in the education system. Cultural high school, cultural differences, continuous learning such as traditional materials, plus academic achievement inclusion and includes individualized approaches to learning. The paper examines existing AI-driven initiatives and assesses their effectiveness in preventing educational gaps and promoting diversity. Examines the ethical implications of implementing AI, addresses concerns about data privacy, algorithmic biases, and reinforcement of existing asymmetries Detailed literature and case studies build Emphasizes successful AI applications and provides best practice ethical frameworks. The research approach includes active data preprocessing, machine learning, and ethical considerations for AI-driven interventions aimed at advancing student achievement and educational equity. The findings suggest that AI can significantly improve inclusion by providing personalized learning strategies and warning systems for at-risk students, although ethics must be treated with caution solutions to the challenges of. The paper offers policy recommendations for educational institutions to create an enabling environment for the ethical and positive use of AI, with the aim of creating a future where higher education is a beacon of opportunity for all students, including those from in regardless of education system.
The integration of AI, especially Generative AI (GenAI), is producing a significant change in the ever-changing higher education scene. This chapter investigates how GenAI is transforming teaching, learning, and academic literacy. Academic literacy facilitators must now negotiate a complex landscape that includes conventional materials, digital resources, and AI-enhanced texts. They train scholars in GenAI tools and pioneer creative teaching methodologies. This chapter provides GenAI ontology to help guide you through this revolutionary journey. It prepares facilitators and students to utilize GenAI successfully by promoting specialized teaching techniques and individualized literacy evaluations. In conclusion, this chapter discusses GenAI's potential to innovate, improve access, and boost intellectual prowess in higher education.
With the rapid advancement of information technology, the Intelligent Teaching System (ITS) has emerged as a pivotal tool in mathematics education. This paper aims to evaluate the effectiveness of ITS by exploring its impact on personalized learning, increased student interaction and participation, intelligent assessment and feedback, teacher support, and resource optimization. Through a comprehensive analysis, the study examines the specific effects of ITS on student learning outcomes, satisfaction, and overall teaching efficiency. Focusing on key aspects such as adaptive learning pathways, real-time feedback, and enhanced engagement, this paper highlights how ITS can revolutionize traditional teaching approaches, thereby improving both teaching quality and student performance.
https://orcid.org/0000-0003-1342-5924
The research aims to address the importance of the smart teaching system to guide and assist university students. By improving their learning experience. This system helps solve several problems related to the limited resources of universities and students, language barriers, and cultural differences. Uses technology to provide individual feedback, adaptive learning methods, and data analytics to improve student performance. Key features of the Smart Tutoring system include an easy-to-use interface, adaptive learning methods, and data analytics to monitor student achievement, identify areas for improvement, and help deliver personalized interventions. The system increases student engagement and motivation, improving the learning experience and promoting active participation in education.
This chapter highlights the contributions of EPATHLO, an innovative AI-based educational adventure game, in fostering 21st-century skills. The game utilizes advanced AI techniques to assess students’ knowledge dynamically, adapting educational content and scenarios to achieve dual objectives: improving knowledge acquisition and maintaining engagement. By involving students in interdisciplinary tasks through non-playable characters (NPCs), EPATHLO creates a prosocial gaming environment that cultivates critical soft skills essential for modern careers, including communication, problem-solving, teamwork, time management, and adaptability. The chapter concludes by discussing the challenges and future directions for intelligent educational games, highlighting the potential for integrating emerging technologies to enhance learning outcomes.
Educators can utilize authoring tools, namely software designed to create and update educational content without requiring technical programming expertise. These tools also enable educators to track their students’ learning progress. However, most authoring tools accompanying Intelligent Tutoring Systems (ITSs) do not allow interference with the AI mechanisms operating in the background. This chapter introduces an innovative authoring tool that not only empowers educators to monitor students’ progress and create courses but also allows them to intervene with the AI mechanism and modify its functions. The ITS in this work utilizes a fuzzy mechanism, which processes imprecise or uncertain information to enhance adaptability and personalize the learning experience. The tutor can introduce additional fuzzy variables to assess the student’s level of knowledge, such as question frequency, percentage of incorrect responses, and the time taken to answer a question beyond the expected duration. This additional customization feature is expected to enable the tutor to tailor the intelligent mechanisms and draw even more precise conclusions about the student’s academic performance or the material to be provided. As a result, students can benefit from a progressively personalized learning experience and improve their learning abilities.
This study is a systematic and bibliometric analysis of research on artificial intelligence, ChatGPT, and large language models, focusing on the top fifty publications from high-impact linguistics journals indexed in Web of Science and Scopus. This study aims to identify and analyze the historical significance of ChatGPT’s release and its broader influence on linguistic research as a field. Through scientific mapping we examine research publications from before and after the launch of ChatGPT, analyzing how it presents in choice of topics and methodologies. The study presents a thematic review of articles published up to March 2024, categorizing them into major themes such as editorials, survey studies, translation research, and assessment methodologies. Through the use of Bibliometrix software, the research analyzes the geographical networks of articles, authorship collaborations, citations, and recurring keywords. The findings reveal the growing influence of artificial intelligence, particularly ChatGPT, on various subfields of linguistics, and demonstrate how AI technologies are being integrated into areas such as translation, language assessment, and educational practices. By mapping the trajectory of AI’s integration into linguistics, the study offers insights into future research directions, emphasizing the need to continuously assess AI’s impact on educational and linguistic practices.
In recent years, STEM education, which encompasses science, technology, engineering, and mathematics, has experienced significant and dynamic progress. These advancements are characterised by the proliferation of scientific knowledge and the development of cutting-edge educational resources that rely on artificial intelligence technology, among other factors. A promising avenue for advancing formal education is presented by intelligent tutoring systems, which offer intelligent instruction and feedback, thereby facilitating a more personalized and practical learning experience. This study explored emerging trends and the feasibility of integrating intelligent tutoring systems in STEM education. A systematic literature review was carried out following PRISMA guidelines, with a total of 24 studies included, selected based on predefined inclusion criteria aligned with the research objective. The analysis reveals a growing interest in intelligent tutoring systems within STEM education between 2019 and 2024. Furthermore, the majority of research conducted thus far has focused on the K-12 education system and higher education institutions. This research initially examined the impact of Intelligent Tutoring Systems on enhancing student motivation and overall academic performance in STEM education courses. The findings substantiate the assertion that integrating artificial intelligence into intelligent tutoring systems positively impacts student motivation and achievement in STEM education. Additionally, students' prior knowledge of STEM subjects enhances their engagement and motivation when using intelligent tutoring systems. Integrating intelligent tutoring systems into STEM education has significantly improved student motivation and academic achievement.
As research on artificial intelligence (AI) in education continues to expand, many scholars predict significant changes in the roles of teachers, schools, and educational leaders. This study seeks to examine potential outcomes of integrating AI into education and its implications for the future of educational institutions. Using a phenomenological approach, a qualitative research method, the study explores the perspectives of individuals from diverse sectors. The findings indicate that AI's introduction in education will bring innovative tools and advantages for schools and teachers, while also presenting certain challenges.
Evaluations on the role of artificial intelligence (AI) in education emphasize its potential contributions to student-centered learning and personalized education. However, while studies have begun to explore the expected contributions of these relatively new AI applications, comparative differences—specifically performance assessments—between AI usage and direct human effort are not yet sufficiently developed. Although there are limited studies aimed at determining learning styles through the use of AI, their consistency with actual results is not thoroughly examined. This study aims to assess the individual differences of accounting students at a vocational and technical high education school using the Kolb Learning Style Inventory (KLSI) and to evaluate the performance (consistency) of AI applications (ChatGPT, Gemini, and Copilot) against actual implementations. To this end, responses from 11 vocational and technical high school accounting students, whose learning styles were previously determined using KLSI, were utilized. Three different AI tools were instructed to determine the learning styles of these students using the same commands. In this way, the effectiveness of AI tools in identifying and assessing individual differences among students was examined both independently and comparatively. According to the findings, ChatGPT showed the highest performance, with only one incorrect assessment, while the other AIs made three incorrect assessments. Notably, the observation that ChatGPT incorrectly identified did not overlap with the incorrect observations of the others. In contrast, two of the three incorrect assessments by Gemini and Copilot pertained to the same two observations. Based on all the findings, this study, which provides an initial evaluation of the performance of AI in meeting the expected contributions and, specifically, in using KLSI, suggests that while AI can facilitate the identification and evaluation of individual differences in teaching, the possibility of errors should not be overlooked. Essentially, the study, with its empirical evidence, highlights that AIs still need to continue learning themselves and that relying solely on AI in zero-tolerance-required tasks, such as identifying students' individual characteristics, could be risky.
Recent years have seen growing interest in open-ended interactive educational tools such as games. One of the most crucial aspects of developing games lies in modeling and predicting individual behavior, the study of computational models of players in games. Although model-based approaches have been considered standard for this purpose, their application is often extremely difficult due to the huge space of actions that can be created by educational games. For this reason, data-driven approaches have shown promise, in part because they are not completely reliant on expert knowledge. This study seeks to systematically review the existing research on the use of data-driven approaches in player modeling of educational games. The primary objectives of this study are to identify, classify, and bring together the relevant approaches. We have carefully surveyed a 10-year sample (2008–2017) of research studies conducted on data-driven approaches in player modeling of educational games, and thereby found 67 significant research works. However, our criteria for inclusion reduced the sample to 21 studies that addressed four primary research questions, and so we analyzed and classified the questions, methods, and findings of these published works, which we evaluated and from which we drew conclusions based on non-statistical methods. We found that there are three primary avenues along which data-driven approaches have been studied in educational games research: first, the objective of data-driven approaches in player modeling of educational games, namely behavior modeling, goal recognition, and procedural content generation; second, approaches employed in such modeling; finally, current challenges of using data-driven approaches in player modeling of educational games, namely game data, temporal forecasting in player models, statistical techniques, algorithmic efficiency, knowledge engineering, problem of generalizability, and data sparsity problem. In conclusion we addressed four critical future challenges in the area, namely, the lack of proper and rich data publicly available to the researchers, the lack of a data-driven method to identify conceptual features from log data, hybrid player modeling approaches, and data mining techniques for individual prediction.
This paper describes the development and evaluation of an affect-aware intelligent support component that is part of a learning environment known as iTalk2Learn. The intelligent support component is able to tailor feedback according to a student's affective state, which is deduced both from speech and interaction. The affect prediction is used to determine which type of feedback is provided and how that feedback is presented (interruptive or non-interruptive). The system includes two Bayesian networks that were trained with data gathered in a series of ecologically-valid Wizard-of-Oz studies, where the effect of the type of feedback and the presentation of feedback on students' affective states was investigated. This paper reports results from an experiment that compared a version that provided affect-aware feedback (affect condition) with one that provided feedback based on performance only (non-affect condition). Results show that students who were in the affect condition were less bored and less off-task, with the latter being statically significant. Importantly, students in both conditions made learning gains that were statistically significant, while students in the affect condition had higher learning gains than those in the non-affect condition, although this result was not statistically significant in this study's sample. Taken all together, the results point to the potential and positive impact of affect-aware intelligent support.
: Protocols of systematic reviews and meta-analyses allow for planning and documentation of review methods, act as a guard against arbitrary decision making during review conduct, enable readers to assess for the presence of selective reporting against completed reviews, and, when made publicly available, reduce duplication of efforts and potentially prompt collaboration. Evidence documenting the existence of selective reporting and excessive duplication of reviews on the same or similar topics is accumulating and many calls have been made in support of the documentation and public availability of review protocols. Several efforts have emerged in recent years to rectify these problems, including development of an international register for prospective reviews (PROSPERO) and launch of the first open access journal dedicated to the exclusive publication of systematic review products, including protocols (BioMed Central's Systematic Reviews). Furthering these efforts and building on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines, an international group of experts has created a guideline to improve the transparency, accuracy, completeness, and frequency of documented systematic review and meta-analysis protocols--PRISMA-P (for protocols) 2015. The PRISMA-P checklist contains 17 items considered to be essential and minimum components of a systematic review or meta-analysis protocol.This PRISMA-P 2015 Explanation and Elaboration paper provides readers with a full understanding of and evidence about the necessity of each item as well as a model example from an existing published protocol. This paper should be read together with the PRISMA-P 2015 statement. Systematic review authors and assessors are strongly encouraged to make use of PRISMA-P when drafting and appraising review protocols.
Face-to-face and one-to-one human tutoring is the best tutoring field. Human tutors are not always and everywhere available and that’s why computer based tutoring is developed. Intelligent Tutoring System (ITS) is the software system developed for tutoring human subjects, when human tutor is not available. ITS performance is not as good as human tutors, but ITS domain is improving and recently gaining much popularity. An ITS development project can involve an interdisciplinary team of experts from the domains of learning sciences, cognitive sciences, software engineering, and Human-computer interaction (HCI). ITS is a software intended to be used by human user; any software development project involves designing of user interfaces. Although ITS development has improved, but most of the development has focused on learning sciences and mostly ignoring the user interfaces. This has resulted in current ITS with efficient learning modules, but less efficient interface design. Performance of a software can be measured by it usability, and one way to define quality of a software is by its “ease of use”. Software that is not easy to use by its user can be considered of lower quality. This paper reviews the literature of ITS and usability; with the intention of indicating the currant gap between ITS development and usability of the system. This article also presents a discussion which can provide future directions in ITS development with usability context.
Intelligent e-learning systems have revolutionized online education by providing individualized and personalized instruction for each learner. Nevertheless, till now very few systems were able to leave academic labs and be integrated in real commercial products. One of these few exceptions is the Learning Intelligent Advisor (LIA) described in this paper, built on results coming from several research projects and currently integrated in a complete e-learning solution named IWT. The purpose of this paper is to describe how LIA works and cooperates with IWT in the provisioning of individualized e-learning experiences. Defined algorithms and underlying models are described as well as architectural aspects related to the integration in IWT. Results of experimentations with real users are discussed to demonstrate the benefits of LIA as an add-on in on-line learning.
In this paper we present a model of a system for integration of an intelligent tutoring system with data mining tools. The purpose of the integration is twofold; a) to power the system adaptability based on clustering and sequential pattern mining, and b) to enable teachers (non-experts in data mining) to use data mining techniques in their web browser on a daily basis, and get useful visualizations that provide insights into the learning progress of their students. We also present an approach to clustering results evaluation developed so that the system can independently deduce the best number of clusters for the k-means algorithm as well as order the clusters in terms of learning efficiency of cluster members (students).
This review describes a meta-analysis of findings from 50 controlled evaluations of intelligent computer tutoring systems. The median effect of intelligent tutoring in the 50 evaluations was to raise test scores 0.66 standard deviations over conventional levels, or from the 50th to the 75th percentile. However, the amount of improvement found in an evaluation depended to a great extent on whether improvement was measured on locally developed or standardized tests, suggesting that alignment of test and instructional objectives is a critical determinant of evaluation results. The review also describes findings from two groups of evaluations that did not meet all of the selection requirements for the meta-analysis: six evaluations with nonconventional control groups and four with flawed implementations of intelligent tutoring systems. Intelligent tutoring effects in these evaluations were small, suggesting that evaluation results are also affected by the nature of control treatments and the adequacy of program implementations.
The evaluation of interactive adaptive and personalised systems has long been acknowledged as a difficult, complicated and very demanding endeavour due to the complex nature of these systems. This paper describes a web-based framework for the evaluation of end-user experience in adaptive and personalised e-Learning systems. The benefits of the framework include: i) the provision of an interactive reference and recommendation tool to encourage the evaluation of systems that fulfil certain methodological requirements; ii) the collaborative nature of the framework facilitates the sharing of information among researchers from the information technology, adaptive hypermedia, information retrieval and e-Learning communities; iii) the identification of pitfalls in the evaluation planning process as well as in data analysis; and iv) the translation of presented information into users language of choice. This paper also presents a review of User-Centred Evaluation approaches, methodologies and techniques adopted by current systems and frameworks. The results of this review are analysed. From these results, an architectural design for the framework was specified.
Intelligent tutoring and personalization are considered as the two most important factors in the
research of learning systems and environments. An effective tool that can be used to improve
problem-solving ability is an Intelligent Tutoring System which is capable of mimicking a
human tutor’s actions in implementing a one-to-one personalized and adaptive teaching. In this
paper, a novel Flowchart-based Intelligent Tutoring System (FITS) is proposed benefiting from
Bayesian networks for the process of decision making so as to aid students in problem-solving
activities and learning computer programming. FITS not only takes full advantage of Bayesian
networks, but also benefits from a multi-agent system using an automatic text-to-flowchart
conversion approach for engaging novice programmers in flowchart development with the aim
of improving their problem-solving skills. In the end, in order to investigate the efficacy of FITS
in problem-solving ability acquisition, a quasi-experimental design was adopted by this
research. According to the results, students in the FITS group experienced better improvement
in their problem-solving abilities than those in the control group. Moreover, with regard to the
improvement of a user’s problem-solving ability, FITS has shown to be considerably effective
for students with different levels of prior knowledge, especially for those with a lower level of
prior knowledge.
Cultural awareness, when applied to Intelligent Learning Environments (ILEs), contours the overall appearance, behaviour, and content used in these systems through the use of culturally-relevant student data and information. In most cases, these adaptations are system-initiated with little to no consideration given to student-initiated control over the extent of cultural-awareness being used. This paper examines some of the issues relevant to these challenges through the development of the ICON (Instructional Cultural cONtextualisation) system. The paper explores computational approaches for modelling the diversity of students within subcultures, and the necessary semantic formalisms for representing and reasoning about cultural backgrounds at an appropriate level of granularity for ILEs. The paper investigates how student-initiated control of dynamic cultural adaptation of educational content can be achieved in ILEs, and examines the effects of cultural variations of language formality and contextualisation on student preferences for different types of educational content. Evaluations revealed preliminary insight into quantifiable thresholds at which student perception for specific types of culturally-contextualised content vary. The findings further support the notion put forth in the paper that student-initiated control of cultural contextualisation should be featured in ILEs aiming to cater for diverse groups of students.
Systematic reviews should build on a protocol that describes the rationale, hypothesis, and planned methods of the review; few reviews report whether a protocol exists. Detailed, well-described protocols can facilitate the understanding and appraisal of the review methods, as well as the detection of modifications to methods and selective reporting in completed reviews. We describe the development of a reporting guideline, the Preferred Reporting Items for Systematic reviews and Meta-Analyses for Protocols 2015 (PRISMA-P 2015). PRISMA-P consists of a 17-item checklist intended to facilitate the preparation and reporting of a robust protocol for the systematic review. Funders and those commissioning reviews might consider mandating the use of the checklist to facilitate the submission of relevant protocol information in funding applications. Similarly, peer reviewers and editors can use the guidance to gauge the completeness and transparency of a systematic review protocol submitted for publication in a journal or other medium.
This paper presents one possible approach to providing individualised and immediate feedback to students' written responses to short-answer questions. The classroom context for this study is a large first-year undergraduate health sciences course. The motivation for our approach is explained through a brief history of intelligent tutoring systems, the philosophical and educational positions which inspired their development and the practical and epistemological issues which have largely prevented their uptake in a higher education context. The design and implementation of a new empirically-based tutorial dialogue system is described along with the results of in-class evaluation of the new system with 578 student volunteers. © 2013 Jenny McDonald, Alistair Knott, Sarah Stein and Richard Zeng.
Intelligent Tutoring Systems (ITS) are computer programs that model learners’ psychological states to provide individualized instruction. They have been developed for diverse subject areas (e.g., algebra, medicine, law, reading) to help learners acquire domain-specific, cognitive and metacognitive knowledge. A meta-analysis was conducted on research that compared the outcomes from students learning from ITS to those learning from non-ITS learning environments. The meta-analysis examined how effect sizes varied with type of ITS, type of comparison treatment received by learners, type of learning outcome, whether knowledge to be learned was procedural or declarative, and other factors. After a search of major bibliographic databases, 107 effect sizes involving 14,321 participants were extracted and analyzed. The use of ITS was associated with greater achievement in comparison with teacher-led, large-group instruction (g = .42), non-ITS computer-based instruction (g = .57), and textbooks or workbooks (g = .35). There was no significant difference between learning from ITS and learning from individualized human tutoring (g = –.11) or small-group instruction (g = .05). Significant, positive mean effect sizes were found regardless of whether the ITS was used as the principal means of instruction, a supplement to teacher-led instruction, an integral component of teacher-led instruction, or an aid to homework. Significant, positive effect sizes were found at all levels of education, in almost all subject domains evaluated, and whether or not the ITS provided feedback or modeled student misconceptions. The claim that ITS are relatively effective tools for learning is consistent with our analysis of potential publication bias.
Within STEM domains, physics is considered to be one of the most difficult topics to master, in part because many of the underlying principles are counter-intuitive. Effective teaching methods rely on engaging the student in active experimentation and encouraging deep reasoning, often through the use of self-explanation. Supporting such instructional approaches poses a challenge for developers of Intelligent Tutoring Systems. We describe a system that addresses this challenge by teaching conceptual knowledge about basic electronics and electricity through guided experimentation with a circuit simulator and reflective dialogue to encourage effective self-explanation. The Basic Electricity and Electronics Tutorial Learning Environment (BEETLE II) advances the state of the art in dynamic adaptive feedback generation and natural language processing (NLP) by extending symbolic NLP techniques to support unrestricted student natural language input in the context of a dynamically changing simulation environment in a moderately complex domain. This allows contextually-appropriate feedback to be generated “on the fly” without requiring curriculum designers to anticipate possible student answers and manually author multiple feedback messages. We present the results of a system evaluation. Our curriculum is highly effective, achieving effect sizes of 1.72 when comparing pre- to post-test learning gains from our system to those of a no-training control group. However, we are unable to demonstrate that dynamically generated feedback is superior to a non-NLP feedback condition. Evaluation of interpretation quality demonstrates its link with instructional effectiveness, and provides directions for future research and development.
Students tend to retain naïve understandings of concepts such as energy and force even after completing school and entering college. We developed a learning environment called the Virtual Physics System (ViPS) to help students master these concepts in the context of pulleys, a class of simple machines that are difficult to assemble and use in the real world. Several features make the ViPS noteworthy: it combines simulation and tutoring, it customizes tutoring to address common misconceptions, and it employs a pedagogical strategy that identifies student misconceptions and guides students in problem solving through virtual experimentation. This paper presents the ViPS and describes studies in which we evaluated its efficacy and compared learning from the ViPS with learning from constructing and experimenting with real pulleys. Our results indicate that the ViPS is effective in helping students learn and remediate their misconceptions, and that virtual experimentation in the ViPS is more effective than real experimentation with pulleys.
Over the last decade, several research initiatives have investigated the potentials of the educational paradigm shift from the traditional one-size-fits-all teaching approaches to adaptive and personalized learning. On the other hand, mobile devices are recognized as an emerging technology to facilitate teaching and learning strategies that exploit individual learners’ context. This has led to an increased interest on context-aware adaptive and personalized mobile learning systems aiming to provide learning experiences delivered via mobile devices and tailored to learner’s personal characteristics and situation. To this end, in this paper we present a context-aware adaptive and personalized mobile learning system, namely the Units of Learning mobile Player (UoLmP), which aims to support semi-automatic adaptation of learning activities, that is: (a) adaptations to the interconnection of the learning activities (namely, the learning flow) and (b) adaptations to the educational resources, tools and services that support the learning activities. Initial evaluation results from the use of UoLmP provide evidence that UoLmP can successfully adapt the learning flow of an educational scenario and the delivery of educational resources, tools and services that support the learning activities. Finally, these adaptations can facilitate students to complete successfully the learning activities of an educational scenario.
Adaptive collaborative learning support (ACLS) involves collaborative learning environments that adapt their characteristics, and sometimes provide intelligent hints and feedback, to improve individual students’ collaborative interactions. ACLS often involves a system that can automatically assess student dialogue, model effective and ineffective collaboration, and provide relevant support. While there is evidence that ACLS can improve student learning, little is known about why systems that incorporate ACLS are effective. Does relevant support improve student interactions by providing just-in-time feedback, or do students who believe they are receiving relevant support feel more accountable for the collaboration, and thus more motivated to improve their interactions? In this paper, we describe an adaptive system we have developed to support help-giving during peer tutoring in high school algebra: the Adaptive Peer Tutoring Assistant (APTA). To validate our approach, we conducted a controlled study that demonstrated that our system provided students with more relevant support and was more effective at improving student learning than parallel nonadaptive conditions. Our contributions involve generalizable techniques for implementing ACLS that can function adaptively and effectively, and the finding that adaptive support does indeed improve student learning because of the relevance of the support.
Computers have transformed every facet of our culture, most dramatically communication, transportation, finance, science, and the economy. Yet their impact has not been generally felt in education due to lack of hardware, teacher training, and sophisticated software. Another reason is that current instructional software is neither truly responsive to student needs nor flexible enough to emulate teaching. The more instructional software can reason about its own teaching process, know what it is teaching, and which method to use for teaching, the greater is its impact on education. Building Intelligent Interactive Tutors discusses educational systems that assess a student's knowledge and are adaptive to a student's learning needs. Dr. Woolf taps into 20 years of research on intelligent tutors to bring designers and developers a broad range of issues and methods that produce the best intelligent learning environments possible, whether for classroom or life-long learning. The book describes multidisciplinary approaches to using computers for teaching, reports on research, development, and real-world experiences, and discusses intelligent tutors, web-based learning systems, adaptive learning systems, intelligent agents and intelligent multimedia.
This paper proposes a generic methodology and architecture for developing a novel conversational intelligent tutoring system (CITS) called Oscar that leads a tutoring conversation and dynamically predicts and adapts to a student’s learning style. Oscar aims to mimic a human tutor by implicitly modelling the learning style during tutoring, and personalising the tutorial to boost confidence and improve the effectiveness of the learning experience. Learners can intuitively explore and discuss topics in natural language, helping to establish a deeper understanding of the topic. The Oscar CITS methodology and architecture are independent of the learning styles model and tutoring subject domain. Oscar CITS was implemented using the Index of Learning Styles (ILS) model (Felder & Silverman, 1988) to deliver an SQL tutorial. Empirical studies involving real students have validated the prediction of learning styles in a real-world teaching/learning environment. The results showed that all learning styles in the ILS model were successfully predicted from a natural language tutoring conversation, with an accuracy of 61–100%. Participants also found Oscar’s tutoring helpful and achieved an average learning gain of 13%.
In recent years, there is a growing need for computer technology to be used in a real school environment and/or higher education classrooms. However, educational software has often been criticized as it has not been specifically designed to meet the needs of real classrooms. In this study, we have tried to develop the system, what we have called as “ZOSMAT” that will respond almost every needs of a real classroom. ZOSMAT can be used for the purpose of either individual learning or real classroom environment with the guidance of a human tutor during a formal education process. This characteristic of ZOSMAT distinguishes itself from other intelligent tutoring systems. ZOSMAT follows a student in each stage of the learning process and guides him/her about what he/she will need to do. ZOSMAT with a web-based feature presents an equal opportunity of education for both the student in the classroom and the student in the far end of the world. This system is a student-centered one and the progress in student’s learning process depends on his/her effort.
Modeling and predicting player behavior is of the utmost importance in developing games. Experience has proven that, while theory-driven approaches are able to comprehend and justify a model's choices, such models frequently fail to encompass necessary features because of a lack of insight of the model builders. In contrast, data-driven approaches rely much less on expertise, and thus offer certain potential advantages. Hence, this study conducts a systematic review of the extant research on data-driven approaches to game player modeling. To this end, we have assessed experimental studies of such approaches over a nine-year period, from 2008 to 2016; this survey yielded 46 research studies of significance. We found that these studies pertained to three main areas of focus concerning the uses of data-driven approaches in game player modeling. One research area involved the objectives of data-driven approaches in game player modeling: behavior modeling and goal recognition. Another concerned methods: classification, clustering, regression, and evolutionary algorithm. The third was comprised of the current challenges and promising research directions for data-driven approaches in game player modeling.
This paper presents the results of an experiment, conducted on a sample of computer science students, using the adaptive learning system called ALS_CORR[LP] 1. Indeed, unlike the traditional LMS, the adaptive learning systems provide a personalized learning experience based on the objectives, the prerequisites or even the learning styles generating thereafter a specific learning path. However their main issue remains the fact, that they assume that the generated learning path is necessarily the leading one, which is far from being true, since we can always detect some failure cases during the evaluation phase. In this paper we conduct a learning experience using the system ALS_CORR[LP] which has the ability to correct the generated learning path by recommending the most relevant learning objects, and update the learner model based on a calculation of similarity in behavior between the struggling learner and the succeeding ones. We analyze later the results of behavior tracking within the system.
Recent years have seen a growing interest in the role that student drawing can play in learning. Because drawing has been shown to contribute to students’ learning and increase their engagement, developing student models to dynamically support drawing holds significant promise. To this end, we introduce diagrammatic student models, which reason about students’ drawing trajectories to generate a series of predictions about their conceptual knowledge based on their evolving sketches. The diagrammatic student modeling framework utilizes deep learning, a family of machine learning methods based on a deep neural network architecture, to reason about sequences of student drawing actions encoded with temporal and topological features. An evaluation of the deep-learning-based diagrammatic student models suggests that it can predict student performance more accurately and earlier than competitive baseline approaches.
Self-learning process is an important factor that enables learners to improve their own educational experiences when they are away of face-to-face interactions with the teacher. A well-designed self-learning activity process supports both learners and teachers to achieve educational objectives rapidly. Because of this, there has always been a remarkable trend on developing alternative self-learning approaches. In this context, this study is based on two essential objectives. Firstly, it aims to introduce an intelligent software system, which optimizes and improves computer engineering students’ self-learning processes. Secondly, it aims to improve computer engineering students’ self-learning during the courses. As general, the software system introduced here evaluates students’ intelligence levels according to the Theory of Multiple Intelligences and supports their learning via accurately chosen materials provided over the software interface. The evaluation mechanism of the system is based on a hybrid Artificial Intelligence approach formed by an Artificial Neural Network, and an optimization algorithm called as Vortex Optimization Algorithm (VOA). The system is usable for especially technical courses taught at computer engineering departments of universities and makes it easier to teach abstract subjects. For having idea about success of the system, it has been tested with students and positive results on optimizing and improving self-learning have been obtained. Additionally, also a technical evaluation has been done previously, in order to see if the VOA is a good choice to be used in the system. It can be said that the whole obtained results encourage the authors to continue to future works. © 2017 Wiley Periodicals, Inc. Comput Appl Eng Educ 25:142–156, 2017; View this article online at wileyonlinelibrary.com/journal/cae; DOI 10.1002/cae.21787.
This study proposes an Intelligent Tutor System for assessing slide presentations from novice undergraduate students. To develop such system, two learner models (rule based model and clustering model) were built using 80 presentations graded by three human experts. An experiment to determine the best learner model and students' perception was carried out using 51 presentations uploaded by students. The findings show that the clustering model classified in a similar way as a human evaluator only when a holistic evaluation criterion was used. Whereas, the rule-base model was more precise when the evaluation rules were easier to be followed by a human evaluator. Furthermore, students agreed with the usefulness of the system as well as the level of agreement with the grading model, although the latter in a lesser extent. Results from this study encourage to explore this area and adapt the proposed Intelligent Tutor System to other existing automated grading systems.
In this paper, first we present an educational system that assists students in learning and tutors in teaching search algorithms, an artificial intelligence topic. Learning is achieved through a wide range of learning activities. Algorithm visualizations demonstrate the operational functionality of algorithms according to the principles of active learning. So, a visualization process can stop and request from a student to specify the next step or explain the way that a decision was made by the algorithm. Similarly, interactive exercises assist students in learning to apply algorithms in a step-by-step interactive way. Students can apply an algorithm to an example case, specifying the algorithm’s steps interactively, with the system’s guidance and help, when necessary. Next, we present assessment approaches integrated in the system that aim to assist tutors in assessing the performance of students, reduce their marking task workload and provide immediate and meaningful feedback to students. Automatic assessment is achieved in four stages, which constitute a general assessment framework. First, the system calculates the similarity between the student’s and the correct answer using the edit distance metric. In the next stage, it identifies the type of the answer, based on an introduced answer categorization scheme related to completeness and accuracy of an answer, taking into account student carelessness too. Afterwards, the types of errors are identified, based on an introduced error categorization scheme. Finally, answer is automatically marked via an automated marker, based on its type, the edit distance and the type of errors made. To assess the learning effectiveness of the system an extended evaluation study was conducted in real class conditions. The experiment showed very encouraging results. Furthermore, to evaluate the performance of the assessment system, we compared the assessment mechanism against expert (human) tutors. A total of 400 students’ answers were assessed by three tutors and the results showed a very good agreement between the automatic assessment system and the tutors.
Deductive logic is essential to a complete understanding of computer science concepts, and is thus fundamental to computer science education. Intelligent tutoring systems with individualized instruction have been shown to increase learning gains. We seek to improve the way deductive logic is taught in computer science by developing an intelligent, data-driven logic tutor. We have augmented Deep Thought, an existing computer-based logic tutor, by adding data-driven methods, specifically; intelligent problem selection based on the student’s current proficiency, automatically generated on-demand hints, and determination of student problem solving strategies based on clustering previous students. As a result, student tutor completion (the amount of the tutor the students completed) steadily improved as data-driven methods were added to Deep Thought, allowing students to be exposed to more logic concepts. We also gained additional insights into the effects of different course work and teaching methods on tutor effectiveness.
Technological advancements within the educational sector and online learning promoted portable data-based adaptive techniques to influence the developments within transformative learning and enhancing the learning experience. However, many common adaptive educational systems tend to focus on adopting learning content that revolves around pre-black box learner modelling and teaching models that depend on the ideas of a few experts. Such views might be characterized by various sources of uncertainty about the learner response evaluation with adaptive educational system, linked to learner reception of instruction. High linguistic uncertainty levels in e-learning settings result in different user interpretations and responses to the same techniques, words, or terms according to their plans, cognition, pre-knowledge, and motivation levels. Hence, adaptive teaching models must be targeted to individual learners’ needs. Thus, developing a teaching model based on the knowledge of how learners interact with the learning environment in readable and interpretable white box models is critical in the guidance of the adaptation approach for learners’ needs as well as understanding the way learning is achieved.
This paper presents a novel interval type-2 fuzzy logic-based system which is capable of identifying learners’ preferred learning strategies and knowledge delivery needs that revolves around characteristics of learners and the existing knowledge level in generating an adaptive learning environment. We have conducted a large scale evaluation of the proposed system via real-word experiments on 1458 students within a massively crowded e-learning platform. Such evaluations have shown the proposed interval type-2 fuzzy logic system’s capability of handling the encountered uncertainties which enabled to achieve superior performance with regard to better completion and success rates as well as enhanced learning compared to the non-adaptive systems, adaptive system versions led by the teacher, and type-1-based fuzzy based counterparts.
The realisation of personalised e-learning to suit an individual learner's diverse learning needs is a concept which has been explored for decades, at great expense, but is still not achievable by non-technical authors. This research reviews the area of personalised e-learning and notes some of the technological challenges which developers may encounter in creating authoring tools for personalised e-learning and some of the pedagogical challenges which authors may encounter when creating personalised e-learning activities to enhance the learning experience of their students. At present educators who wish to create personalised e-learning activities require the assistance of technical experts who are knowledgeable in the area. Even with the help of an expert the creation of personalised e-learning activities still remains a complex process to authors who are new to the concept of tailoring e-learning to suit learner diversity. Before the successful utilisation of adaptive authoring tools can be realised, academic authors need to learn how to effectively use these tools. All learners come to education with a diverse set of characteristics; educators need to decide which learner characteristic(s) they wish to focus on addressing through the use of personalised e-learning activities. Further investigation, evaluation and analyses of authoring tools is required before personalised e-learning to support learner diversity can be achieved by many academics. Research members of the AMAS (2013) project team are currently involved in developing an authoring tool for adaptive activities for e-learning.
Programmable logic controllers (PLC) are used for many industrial process control applications. Learning to write ladder logic programs for PLC control is an important and challenging task. However, the learning of ladder logic is often hindered by limited PLC availability due to expensive lab setup, limited lab time, and high student/instructor ratios. With the help of the internet, teaching is not constrained in the traditional classroom pedagogy; the instructors can put the course material on the website and allow the students go on to the course webpage as an alternative way to learn the domain knowledge. However, there is no interaction between the users and learning materials; so, the learning efficiency is often limited. The problem here is how to design a web-based system that is intelligent and adaptive enough to teach the students domain knowledge in PLC. In this research, we proposed a system architecture which combines the pre-test, cased-based reasoning (i.e., heuristic functions), tutorials and tests of the domain concepts, and post-test (i.e., including pre- and post-exam) to customize students' needs according to their knowledge levels and help them learn the PLC concepts, effectively. We have developed an intelligent tutoring system which is mainly based on the feedback and learning preference of the users' questionnaires. It includes many pictures, colorful diagrams, and interesting animations (i.e., switch control of the user's rung configuration) to attract the users' attention. From the model simulation results, a knowledge proficiency effect occurs on problem solving time. If the students are more knowledgeable about PLC concepts, they will take less time to complete problems than those who are not as proficient. Additionally, from the system experiments, the results indicate that the learning algorithm in this system is robust enough to pinpoint the most accurate error pattern (i.e., almost 90 % accuracy of mapping to the most similar error pattern), and the adaptive system will have a higher accuracy of discerning the error patterns which are close to the answers of the PLC problems when the databases have more built-in error patterns. The participant evaluation indicates that after using this system, the users will learn how to solve the problems and have a much better performance than before.
This paper presents the evaluation of the synchronization of three emotional measurement methods (automatic facial expression recognition, self-report, electrodermal activity) and their agreement regarding learners’ emotions. Data were collected from 67 undergraduates enrolled at a North American University whom learned about a complex science topic while interacting with MetaTutor, a multi-agent computerized learning environment. Videos of learners’ facial expressions captured with a webcam were analyzed using automatic facial recognition software (FaceReader 5.0). Learners’ physiological arousal was recorded using Affectiva’s Q-Sensor 2.0 electrodermal activity measurement bracelet. Learners’ self-reported their experience of 19 different emotional states on five different occasions during the learning session, which were used as markers to synchronize data from FaceReader and Q-Sensor. We found a high agreement between the facial and self-report data (75.6%), but low levels of agreement between them and the Q-Sensor data, suggesting that a tightly coupled relationship does not always exist between emotional response components.
E-materials and various e-learning systems have become regular features in lower secondary schools in Slovenia and around the world. Many different systems and materials have been created for students, but only a few offer the same amount of individualisation that is present in traditional one to one teaching (one teacher to one student). The purpose of this research is to demonstrate the design and evaluation of an adaptive, intelligent and, most important, an individualised intelligent tutoring system (ITS) based on the cognitive characteristics of the individual learner. The TECH8 model presented is designed modularly, based on a system for collecting a range of metadata and variables that are vital for the teaching process. Prepared in such a way, the proposed system supports individualization and differentiation; because of this, it can be adapted to each individual's level of knowledge and understanding of the subject matter.
Personalization and intelligent tutor are two key factors in the research on learning environment. Intelligent tutoring system (ITS), which can imitate the human teachers' actions to implement one-to-one personalized teaching to some extent, is an effective tool for training the ability of problem solving. This research firstly discusses the concepts and methods of designing problem solving oriented ITS, and then develops the current iTutor based on the extended model of ITS. At last, the research adopts a quasi-experimental design to investigate the effectiveness of iTutor in skills acquisition. The results indicate that students in iTutor group experience better learning effectiveness than those in the control group. iTutor is found to be effective in improving the learning effectiveness of students with low-level prior knowledge.
In this paper, we present an intelligent tutoring system developed to
assist students in learning logic. The system helps students to learn how to construct
equivalent formulas in first order logic (FOL), a basic knowledge representation
language. Manipulating logic formulas is a cognitively complex and
error prone task for the students to deeply understand. The system assists students
to learn to manipulate and create logically equivalent formulas in a stepbased
process. During the process the system provides guidance and feedback
of various types in an intelligent way based on user’s behavior. Evaluation of
the system has shown quite satisfactory results as far as its usability and learning
capabilities are concerned.
Effective mathematics teachers have a large body of professional knowledge, which is largely undocumented and shared by teachers working in a given country’s education system. The volume and cultural nature of this knowledge make it particularly challenging to share curricula and instructional methods between countries. Thus, approaches based on knowledge engineering—designing a software system by interviewing human experts to extract their knowledge and heuristics—are particularly promising for cross-cultural curriculum implementations. Reasoning Mind’s Genie 2 system demonstrates the viability of such an approach, bringing elements of Russian mathematics education (known for its effectiveness) to the United States. Genie 2 has been adopted on a large scale, with around 67,000 United States students participating in the 2012–2013 school year. Previously published work (some of it in peer reviewed articles and some in non-peer-reviewed independent evaluations) has associated Genie 2 with high student and teacher acceptance, increases in test scores relative to “business as usual” conditions, high time on task, and a positive affective profile. Here, we describe for the first time the design, function, and use of the Genie 2 system. Based on this work, we suggest general principles—which collectively represent a proposed methodology—for the design of intelligent tutoring systems intended for cross-cultural transfer of curriculum and instructional methods.
Nowadays, adaptive and intelligent tutoring system AITS is one of regarded topics. So researchers are trying to optimize and expand its application in the field of education. This paper integrates AITS with expert system technology. It is intelligent because it can interact with the learners and offer them some subjects based on pedagogy view. Learning process in this system is as follows. First, the expert system plans a pre-evaluation and then calculates his score. If the learner gets the required score, the activities will be trained. Then the learner will be evaluated by a post-evaluation. After that, the system offers guidance in learning other activities. For that purpose it takes into account achievements, learning context and skill levels, by analyzing the other activities already carried out. The analysis is based on a set of rules. The proposed system covers all important properties such as hypertext component, adaptive sequencing, problem-solving support, intelligent solution analysis and adaptive presentation while available systems have only some of them. It can significantly improve the learning result. In other words, it helps learners to study in "the best way".
Programming is a subject that many beginning students find difficult. The PHP Intelligent Tutoring System (PHP ITS) has been designed with the aim of making it easier for novices to learn the PHP language in order to develop dynamic web pages. Programming requires practice. This makes it necessary to include practical exercises in any ITS that supports students learning to program. The PHP ITS works by providing exercises for students to solve and then providing feedback based on their solutions. The major challenge here is to be able to identify many semantically equivalent solutions to a single exercise. The PHP ITS achieves this by using theories of Artificial Intelligence (AI) including first-order predicate logic and classical and hierarchical planning to model the subject matter taught by the system. This paper highlights the approach taken by the PHP ITS to analyse students’ programs that include a number of program constructs that are used by beginners of web development. The PHP ITS was built using this model and evaluated in a unit at the Queensland University of Technology. The results showed that it was capable of correctly analysing over 96 % of the solutions to exercises supplied by students.
In this paper, we evaluate the effectiveness and accuracy of the student model of a web-based educational environment for teaching computer programming. Our student model represents the learner’s knowledge through an overlay model and uses a fuzzy logic technique in order to define and update the student’s knowledge level of each domain concept, each time that s/he interacts with the e-learning system. Evaluation of the student model of an Intelligent Tutoring System (ITS) is an aspect for which there are not clear guidelines to be provided by literature. Therefore, we choose to use two well-known evaluation methods for the evaluation of our fuzzy student model, in order to design an accurate and correct evaluation methodology. These evaluation models are: the Kirkpatrick’s model and the layered evaluation method. Our system was used by the students of a postgraduate program in the field of Informatics in the University of Piraeus, in order to learn how to program in the programming language C. The results of the evaluation were very encouraging.
Excerpts available on Google Books (see link below). For more information, go to publisher's website : http://www.taylorandfrancis.com/books/details/9780805800548/
The thesis aims to illustrate the way how to construct a dualistic and dynamic Student Model in SQLTutor intelligent teaching system. The initial level of the Student Model is to collect the data of the student performance, and paradigmatically record and sort out the performance data of the individual student. The senior level of the Student Model is to apply the “fuzzy cluster” method to analyze further the data drawn from the initial level so that we can syntagmatically evaluate the overall students. Practice has proved that the Student Model can deliver a qualitative and quantitative evaluation of the individualized online learning of the students.
The most important feature of the intelligent tutoring systems (ITS), one of the most popular study topics of recent years, is that it provides an opportunity for individual learning by taking students' individual differences into account. In order to be able to realize this feature, it is necessary that the system recognizes students well. The process of recognizing student is performed as a result of observations which ITS applies on students. A number of uncertainties arise during these observations. In order to minimize learning uncertainties and create a productive and effective ITS, type-2 fuzzy logic, one of the artificial intelligence techniques, is used in the system developed in this study. In order to show the effectiveness of the developed web-based ITS, it is applied to the teaching of a basic Control Course. The educational evaluation of the system is presented in the paper. Comput Appl Eng Educ © 2011 Wiley Periodicals, Inc.
With the rapid growth of information technology, the e-learning has become a major trend in the computer assisted teaching and learning field. Previously, many researchers put efforts into e-learning system with emphasizing the application of multimedia elements; they often neglected the importance of three crucial elements-personalization, contextual understanding and platform-independent standardized learning materials, which are rather important for students of diverse disciplines background and learning abilities. To build a system for better implementing Web-based learning, the paper proposes using agents to design an intelligent tutoring system implemented for Web tutoring by combining learning theory. Evaluation results indicate that applying our system can efficiently help students increase learning efficiency while receiving traditional classroom instruction.
Personalized curriculum sequencing is an important research issue for web-based learning systems because no fixed learning paths will be appropriate for all learners. Therefore, many researchers focused on developing e-learning systems with personalized learning mechanisms to assist on-line web-based learning and adaptively provide learning paths in order to promote the learning performance of individual learners. However, most personalized e-learning systems usually neglect to consider if learner ability and the difficulty level of the recommended courseware are matched to each other while performing personalized learning services. Moreover, the problem of concept continuity of learning paths also needs to be considered while implementing personalized curriculum sequencing because smooth learning paths enhance the linked strength between learning concepts. Generally, inappropriate courseware leads to learner cognitive overload or disorientation during learning processes, thus reducing learning performance. Therefore, compared to the freely browsing learning mode without any personalized learning path guidance used in most web-based learning systems, this paper assesses whether the proposed genetic-based personalized e-learning system, which can generate appropriate learning paths according to the incorrect testing responses of an individual learner in a pre-test, provides benefits in terms of learning performance promotion while learning. Based on the results of pre-test, the proposed genetic-based personalized e-learning system can conduct personalized curriculum sequencing through simultaneously considering courseware difficulty level and the concept continuity of learning paths to support web-based learning. Experimental results indicated that applying the proposed genetic-based personalized e-learning system for web-based learning is superior to the freely browsing learning mode because of high quality and concise learning path for individual learners.
While problem-based learning has become widely popular for imparting clinical reasoning skills, the dynamics of medical PBL require close attention to a small group of students, placing a burden on medical faculty, whose time is over taxed. Intelligent tutoring systems (ITSs) offer an attractive means to increase the amount of facilitated PBL training the students receive. But typical intelligent tutoring system architectures make use of a domain model that provides a limited set of approved solutions to problems presented to students. Student solutions that do not match the approved ones, but are otherwise partially correct, receive little acknowledgement as feedback, stifling broader reasoning. Allowing students to creatively explore the space of possible solutions is exactly one of the attractive features of PBL. This paper provides an alternative to the traditional ITS architecture by using a hint generation strategy that leverages a domain ontology to provide effective feedback. The concept hierarchy and co-occurrence between concepts in the domain ontology are drawn upon to ascertain partial correctness of a solution and guide student reasoning towards a correct solution. We describe the strategy incorporated in METEOR, a tutoring system for medical PBL, wherein the widely available UMLS is deployed and represented as the domain ontology. Evaluation of expert agreement with system generated hints on a 5-point likert scale resulted in an average score of 4.44 (Spearman's ρ=0.80, p<0.01). Hints containing partial correctness feedback scored significantly higher than those without it (Mann Whitney, p<0.001). Hints produced by a human expert received an average score of 4.2 (Spearman's ρ=0.80, p<0.01).