Conference PaperPDF Available

Role of AI in effective eLearning

Authors:
  • Director, IIIT Kottayam, Kerala, India Institute of National Importance

Abstract and Figures

It was only ten years ago that the fax machine was revolutionary addition to communication tools, which consisted of regular mail, express mail and telephone. Education and Learning via media other than print was limited to the use of (video) cassettes tapes and television. The e-Learning system is moving from first generation to second generation. The integration of Artificial Intelligence and e-Learning is identified as Second Generation iLearning or ITS. In this paper, the potential contribution of AI to enrich e-Learning environment has been discussed. AI can contribute issues such as visual intelligence, automation reasoning, aural intelligence, automated intelligence, learning from experience, intelligent learning and intelligent interaction between humans and machines. To support issues that AI can contribute, three application areas viz. Language, Mathematics and Music have been discussed. Finally, it has been identified that the contribution of AI to e-Learning can greatly enhance the e-Learning environment and makes it more effective.
Content may be subject to copyright.
Role of AI in effective eLearning
Nitin Upadhyay
Department Of Computer Science & Information Systems
BITS, Pilani Goa Campus, Goa, India 403726
nitinu@bits-goa.ac.in
Rajiv V. Dharaskar
Department of Computer Technology,
MP Institute of Engineering & Technology, Gondia, MS, India 441614
rvdharaskar@rediffmail.com
Abstract
It was only ten years ago that the fax machine was revolutionary addition to
communication tools, which consisted of regular mail, express mail and telephone.
Education and Learning via media other than print was limited to the use of (video)
cassettes tapes and television. The e-Learning system is moving from first generation to
second generation. The integration of Artificial Intelligence and e-Learning is identified
as Second Generation iLearning or ITS. In this paper, the potential contribution of AI to
enrich e-Learning environment has been discussed. AI can contribute issues such as
visual intelligence, automation reasoning, aural intelligence, automated intelligence,
learning from experience, intelligent learning and intelligent interaction between humans
and machines. To support issues that AI can contribute, three application areas viz.
Language, Mathematics and Music have been discussed. Finally, it has been identified
that the contribution of AI to e-Learning can greatly enhance the e-Learning environment
and makes it more effective.
Keywords: E-Learning, ITS, CBT, CAI, Applications of AI in e-learning.
1. Introduction
Advancements standards specifications and subsequent adoption have led to major
increases in the extensibility, interoperability and scalability of e-learning technologies.
E-learning is fast becoming a major form of learning
.
It is noticeable that e-Learning
systems environment in the current educational scenario is unavoidable due to knowledge
delivery bottleneck
.
Gartner predicts that by 2006, "50 percent of enterprises will use
simulation to train sales and customer support personnel.” Knowledge grows at a fast speed
and it is too luxurious to have real contact with an expert. Therefore, we must seek to
create an effective learning environment.”
This opens up an exciting research area, which is multidisciplinary in nature. The
design of an e-Learning system governs issues such as infrastructure, curriculum designs,
content development, content management, student assessment, etc. Research from IDC
forecasts huge growth for the e-learning industry with a prediction for the US corporate
e-learning market to more than triple over the next four years from $6.6 billion in
2002 to $23.7 billion in 2006. And, according to Gartner, the future for corporate e-
learning adoption is indeed very bright. In recent years, Artificial Intelligence (AI) has
emerged as the main ingredient of any intelligent system. The design of the e-Learning
system could also benefit from AI technology. In this paper, some AI techniques that
could be the main technology for e-learning has been highlighted. Three possible
application areas has been identified viz. Language, Mathematics and Music. Finally, the
contribution of AI technology in these areas has been recognized.
2. E-learning and ITS
The e-Learning system is moving from first generation to second generation. The e-
Learning environment encompasses a wide range of issues ranging from the design of
infrastructure to the design of contents, from the design of content to the assessments of
whether the content has been successfully delivered. Here, an Intelligent Tutoring
System (ITS) technology that enhances content delivery has been discussed. The
integration of Artificial Intelligence and e-Learning is identified as Second Generation
iLearning or ITS; a branch of science which deals with helping machines find solutions to
complex problems in a more human-like fashion.
The idea of using computer in education exists in parallel with the development in the
computer industry. Here, ITS as an enhancement of a more traditional computer assisted
instruction applications (CAI) has been identified which usually lack intelligence in their
behaviors. Computers have been used in education for over 20 years. Computer-based
training (CBT) and computer aided instruction (CAI) were the first such systems
deployed as an attempt to teach using computers. Early instructional strategies used in
CAI are usually in the two major forms of drill-and-practice and guided instructions.
While both CBT and CAI may be somewhat effective in helping learners, they do not
provide the same kind of individualized attention that a student would receive from a
human tutor [1]. Intelligent tutoring systems have been shown to be highly effective at
increasing students' performance and motivation. For example, students using
Smithtown, an ITS for economics, performed equally well as students taking a traditional
economics course, but required half as much time covering the material [2].
2.1 Components of Intelligent Tutoring Systems
For the purposes of conceptualization and design intelligent tutoring systems can be
viewed as consisting of several interdependent components. Previous research by Woolf
[3] has recognized four major components: the student model, the pedagogical module,
the domain knowledge module, and the communication module. We have identified a
fifth component, the expert model. Woolf includes this component as part of the domain
knowledge, but we feel that it is a separate entity. Figure 1 provides a view of the
interactions between the modules. The goal for every ITS is to communicate its
embedded knowledge effectively
Figure1: Components interaction in an intelligent tutoring system
An ITS must insure the following:
Students' knowledge structures, skills and learning styles accurate diagnosis
Instead of preprogrammed responses, diagnose using principles
Decisive about the next working
Adapt instruction accordingly
Ensure feedback
3. What can AI contribute?
Teaching and learning is not a simple process. By far, only trained humans can teach
effectively. Computer assisted instructions (CAI) could be useful in learning at an
elementary level and even then cannot compete with trained humans in terms of teaching
quality. The complexity of the teaching task is the result of the following main factors:
Adaptive teaching strategies required for effective teaching.
Continuous assessment of students by effective teachers.
To suit individual students, effective teachers adapt their teaching strategy.
AI can enhance convention CAI systems by providing human-like intelligence to the
system. Here, an ITS has been identified as an enhancement of a more traditional
computer assisted instruction applications (CAI) which usually lack intelligence in their
behaviors. This is the general aim of ITS system. In order to enhance computer programs
efficiency in handling the above-mentioned factors, computers must be prepared with the
necessary sensors (to perceive). They must be able to make sense out of their senses (to
infer), then they can act logically (and perhaps emotionally as well). They should be able
to learn, do diagnose and be adaptable to new environments [4-5]. Below, current
developments of these issues and their applications in ITS has been discussed.
3.1. Aural potential
Domains such as language and music require computer for effective functioning and
learning. For this computer must equip with listening and hearing ability. States of
listening are intuitive, rational, passive, meditation, analysis, active, improvisation, and
Student
Model
Communication
Model
Expert Model
Pedagogical
Module
Domain
Knowledge
composition. The ability to hear here also implies to perceive what is being heard.
Advances in speech recognition, speech processing and natural language processing
provide useful applications in many areas including ITS. Recent advances in wireless
communication and speech recognition have made it possible to access the web from any
place, at any time, by using only a phone. Some possible applications are browsing the
web, getting stock quotes, verifying flight schedules, getting maps and directions,
checking email, etc. At the current state of the art, we have a flight booking system where
clients can book their flights with a computer using interactive telephone dialogue [6].
Interactive and natural dialogue is the keywords here: clients talk to a computer as if they
are talking to another human being (compare this to a voice instruction telling you to
press some digits as in a naive phone banking system). A screen reader is software that
works together with a speech synthesizer to read aloud everything contained on a
computer screen, including icons, menus, and text, punctuation, and control buttons.
Potential users for screen readers include students of pronunciation, people learning
languages with orthography different from their native language, and people learning to
read. Examples of screen readers are ASAW from Microtalk, HAL from Dolphin,
OutSpoken from Alva etc.
3.2. Visual sensitivity
The state of the art in education technology addresses the issue of affects and emotions.
Various researches are going on to find out the ways to improve learning capabilities.
Information regarding students’ affects could be useful in the adaptation of teaching
strategies. Visual perception is training in seeing and interpreting visual stimuli. This area
is however still not mature. Perceiving the visual environment in a meaningful way
includes visually recognizing objects, people, and gestures. The ability to see can be
useful in teaching music performance as well. Music activities often demand visual skills
combined with motor responses. Visual motor skills are required to successfully play
instruments, engage in body movement activities, read and write music scores, and
follow a conductor. In many performing skills, the techniques are closely related to the
poster (e.g., hand-shape, sequence of movements, etc.).
3.3. Inference mechanisms
The most important aspect of intelligent behaviors is about inference power. This is the
area where most AI researchers have been investigating. There Logical inference is
suitable for domains at a course grain abstraction, for example, the encoding of domain
knowledge in a rule-based expert module. A complicated challenge for artificial
intelligence since its inception has been knowledge representation in problems with
uncertain domains. Probabilistic and statistical inferences are suitable in handling
uncertainties in the domain. Bayesian networks, artificial neural networks and kernel
machines are among the most popular inference techniques.
3.4. Machine learning and ILE
The crucial factor for the success of any ITS system is the accuracy of its student
model. Real students are dynamic. As such, ITS systems must possess the learning
capability so that changes in students understanding level and approach could be tracked.
Learning is a long-standing research in AI and this is one of the areas where AI could
make a useful contribution to ITS in e-learning environments. Finally, we would like to
comment on the interaction between human and machine. The interaction between
learner and computer in interaction learning environment (ILE) also acts as a key
component of the ITS success. Information from the interactions serves three purposes:
(i) to identify useful information from the environment, (ii) to model an accurate student
model, and (iii) to trigger the environment (e.g., feedback) with useful information.
4. Applications of AI in E-learning
Here, AI applications in ITS which are a part of the e-learning environment has been
discussed. An ITS will behave like a real teacher would, identifying students’ strengths
and weaknesses and deploying a teaching approach which will best fit to students’
learning profile and personality. The system will assist and guide and spontaneously add
supplementary material to the initial course for students’ better understanding. Real-
time feedback is also an essential part of this system, as it shows the instructor
and the learner which areas require additional attention. The system plans and teaches
individual students by maintaining its belief about each student in the student module
and acts according to its best teaching strategy to the virtual students. If the system has
an accurate picture of these students then the teaching should be quite effective. Below,
three applications in ITS has been discussed.
4.1. Language
In order to teach language one has to deal with teaching of reading and writing skills.
The process involves content development, the delivering of content and an assessment
of the delivering process. Here, the assessment dimension in language teaching has been
highlighted. Speech recognition is a difficult problem; it is hard to recognize a phrase or a
sentence with out specifying proper context. Imagine trying to recognize the two phrases:
recognize speech and wreck a nice beach. It is difficult to distinguish the two phrases
without specifying the context in which they are used. However, with a supplied context,
this is not too hard a problem. Using speech technology, assessment of pronunciations
could be automated for a given word, phrase, or sentence. Students can also be benefited
by ITS to enhance their essay writing skills. Students have a constant need for tutors to
give feedbacks on errors/mistakes as well as meritorious use of language, and to place the
essay marked in a suitable band ranging from A to E according to band specifications.
Highlighting errors/mistakes could be easily automated using the spelling check and
grammar check utilities. One of the major limitations in most spelling check or grammar
check utilities is the lack of context analysis (e.g., there are two meaning in the sentence
“I saw the boy with a telescope”). In order to handle context dependency, sophisticated
NLP techniques could be employed. The more complex task would be to identify and
discuss language, which deserves merit. If the system is able to accomplish both tasks,
and is equipped with the band descriptions, then learning of language using ITS will be
more effective. Automated Essay Evaluation: The Criterion Online Writing Service is a
web-based system that provides automated scoring and evaluation of student essays.
Criterion has two complementary applications: (1) Critique Writing Analysis Tools, a
suite of programs that detect errors in grammar, usage, and mechanics, that identify
discourse elements in the essay, and that recognize potentially undesirable elements of
style, and (2) e-rater version 2.0, an automated essay scoring system [6].
4.2. Mathematics
Mathematics and its application related areas recognized as one of the most popular
subjects investigated by AI researchers. In order to model mathematical problem skills in
computers Symbolic Automatic Integrator (SAINT) [8] and Symbolic Integrator (SIN)
[9] are two of the earliest computer programs developed.
Complex domain kind of problems implementation confirm that it is feasible to
implement domain specific knowledge in the ITS systems. The goal of SAINT and SIN is
to exploit the power of symbolic computations rather than to teach mathematical
integration skills to students. One of the important aspect of ITS is explanation; it
facilitates students with full view of the domain specific knowledge, most frequent errors,
guidelines to solve and hints to understand problem theme.
Figure 2: Problem solving tree
Htaik and Phon-Amnuaisuk implement the EGIP sys tem (Explanation Generation for
Integration Problem) [10]. The main purpose is to collect information from the problem
solving tree and use it to guide and teach integration skills to students [11]. The tree
derived from the problem solving process (Figure 2) is considered as a model answer
(of course, there could be different model answers). If for the given problem the answer
given by a student is wrong, then the system will analyses the student’s answer and
syntactically point out the mistakes (Figure 3 e.g., the errors in signs, exponential power,
etc.).
Figure 3: Problem solving tree
By using such approach it can be easy to identify the reason and approach that why
students’ have committed the errors/mistakes. For instance, one can examine and
understand how and why the mistakes are made by constructing a student problem-
solving tree (for the wrong answer). This kind of reasoning and approach is natural and
common in AI.
4.3. Music
At the current state of art, various technologies are used for music education. The
application of computers in music education has been around for many decades [12- 13].
The common teaching activities in music education revolve around music theory, aural
skills and performing skills. From the literature, we see research activities in various
domains e.g. aural training [14-15], performing [16], and learning music theory [17].
Learning music is usually carried out in a highly interactive environment. The ability to
see can be useful in teaching music performance as well. Music activities often demand
visual skills combined with motor responses. Visual motor skills are required to
successfully play instruments, engage in body movement activities, read and write music
scores, and follow a conductor. In many performing skills, the techniques are closely
related to the poster (e.g., hand-shape, sequence of movements, etc.). Learning music
composition is done in a small group, only music theory can be delivered in a
classroom style. To learn music effectively, AI can offer an interactive web-based
learning environment that encompassing aural intelligence (e.g., audio processing), visual
intelligence (e.g., analysis of performers’ movement) and intelligent interactivity [18].
Figure 4: Liberty to make errors
Interactivity in music learning environment has been discussed in [19-20]. We put back
overview of the issue here again:
The unrestricted degree of freedom in student interactions: To promote interactive
learning, it is necessary that the interface supports exploratory learning. Drill and
practice learning style could be implemented in a restricted and deterministic
environment. However, creative learning and exploratory learning require more freedom
in interactive designs (In Figure 4s, there are many errors in the notation and the system
allows these errors. The system uses this information to model the student model and
feedbacks appropriate feedbacks to students).
The logging of event streams and the inferences from the logged information: Event
streams are rich sources of information about the environment. Various inference tactics
have been applied (e.g., logical rule-based systems, Bayesian networks, data mining, etc.)
to accurately modeled student models.
5. Conclusions
Intelligent tutoring systems have been shown to be highly effective in increasing
student motivation and learning. Recent developments in artificial intelligence (AI) allow
course developers to incorporate diagnostic tools, intelligent role-playing, and tutoring
systems into the learning process. The benefits of AI include real-time interaction,
continuous improvement of content delivery, and an integrated approach to learning.
This can be done in a variety of ways including through the use of virtual simulations. In
this paper, the power of AI in capturing domain knowledge, providing intelligent
assessment power, providing intelligent feedbacks, etc, are highlighted. Example
applications in the domain of language, mathematic and music are illustrated. Finally, it
can be inferred that this kind of intelligent activity enhances e-Learning experience and is
a necessary ingredient in the e-Learning environment to make it more effective.
References
[1] Bloom, B.S. (1984). The 2 Sigma Problem: The Search for Methods of Group Instruction as Effective
as One-to-One Tutoring. Educational Researcher, 13, pp. 3-16.
[2] Shute, V., R. Glaser, and K. Raghaven. (1989). Inference and Discovery in an Exploratory Laboratory.
Learning and Individual Differences, Ackerman, P., R. Sterberg, and R. Glaser, eds., pp. 279-326.
[3] Woolf, B. (1992). AI in Education. Encyclopedia of Artificial Intelligence, Shapiro, S., ed., John Wiley
& Sons, Inc., New York, pp. 434-444.
[4] Eliot, C. and Woolf, B. (1995). An Adaptive Student Centered Curriculum for an Intelligent Training
System. User Modeling and User-Adapted Interaction, 5, 1, pp. 67-86.
[5] Lajoie, S. and A. Lesgold. (1992). Apprenticeship Training in the Workplace: Computer-Coached
Practice Environment as a New Form of Apprenticeship. In Intelligent Instruction by Computer, Farr and
Psotka, eds., Taylor & Francis, Washington, DC, pp. 15-36.
[6] S. Seneff and J. Polifroni.(2000) Dialogue Management in Mercury Flight Reservation System. In
Proc. ANLP-NAACL2000, Workshop on Conversational Systems
[7] B. Jill, C. Martin , and L. Claudia (Fall 2004) Automated Essay Evaluation: The Criterion Online
Writing Service. AI Magazine 25(3),pp. 27-36.
[8] J.R.Slagle. (1961) A Heuristic Program that Solves Symbolic Integration in Freshman Calculus. PhD
thesis, Masachusetts Institute Of Technology, 1961.
[9] J. Moses. Symbolic Integration. PhD thesis, Massachusetts Institute of Technology, 1967.
[10] Thaw Ta Htaik and S. Phon-Amnuaisuk. (2005) Intelligent Tutoring System for Mathematical
Problems: Explanation Generations for Integration Problem (EGIP). March, 9-10. In Proceedings to the
third International Conference on Computer Applications (ICCA2005)
[11] Suthers, D. (1992). Answering Student Queries: Functionality and Mechanisms. In Proceedings of the
2nd International Conference on ITS (ITS-92), Montreal, June.
[12] M. Branda˜o, G. Wiggins, and H. Pain. (1999) Computers in Music Education. In Geraint
Wiggins, editor, Proceedings of The AISB’99 Symposium on Musical Creativity, pages 82–88. AISB.
[13] S. Holland. (2000) Artificial intelligence in music education: a crit- ical review. Readings in Music
and Artificial Intelligence, Contemporary Music Studies
[14] F. Hoffstetter. (1981) Computer-based aural trainning: The GUIDO system. Journal of Computer-
based Instruction, 7(3):8492
[15] E. B. Costa, H. O. Almeida, K. S. Silva , R. B. Paes, A. Perkusich (2002). Towards a methodology
for building intelligent tutoring systems in a distance learning environment based on a multi-agent
approach. In: Proceedings of the Workshop on Agent-based Learning Enviroment Brazilian
Symposium on Educational Informatics
[16] M. J. Baker.(1992) Design of an intelligent tutoring system for musical structure and
interpretation. In M. Balaban, K. Ebcioglu, and O. Laske, editors, Understanding Music with AI: Per-
spectives on music cognition, chapter 20, pages 467489. The AAAI Press/The MIT Press
[17] J. Cook. (1998) Mentoring, metacognition and music: Interaction analyses and implications for
intelligent learning environments. Journal of Artificial Intelligence in Education
[18] E. Ferneda, E. de B. Costa, H.O. de Almeida, D.G. Rodrigues, and E.S. de Almeida (2004), A Web-
based Cooperative E-Learning Environment for Musical Harmony Domain, Proceeding (416) Web-based
Education
[19] S. Phon-Amnuaisuk and Chee Keh-Siong, (2005) Interactivities in Music Intelligent Tutoring system.
In Proceedings to the 5th International Conference on Advanced Learning Technology (ICALT2005),
Kaohsiung, Taiwan.
[20] E. Ferneda, E. B. Costa, H. O. Almeid a, D. G. Rodrigues, E. S. Almeida (2004), A web-based
cooperative e-learning environment for musical harmony domain. In: Proceedings of the IASTED
International Conference on Web-based Education Insbruck..
Article
This paper presents a new methodology using graph theory and matrix algebra to analyze software architecture based on systems engineering approach. It proposes a set of analytical tool to capture the notion of structural model as the basis to analyze characteristics of software architecture. In the present work, architecture (structure) modeling and analysis of intelligent mobile learning environment (iMLE) are presented that describe characteristics of performance, quality and reliability.
Article
Full-text available
The perception of the Romanian Public Service Officer came out rather negative according to the most recent public opinion barometers. Our paper sets out to investigate using –inquiry technique- how it feels to be on the other side of the counter, and to what extent the multifaceted professional life of the Romanian Public Service Officer is influenced by various determinants like: work related conditions, education and training, citizens behaviour when facing administration staff, general social environment, specific laws and regulations, politics, aspects of personal life etc. in order to better satisfy their ultimate goal, namely to best serve the citizens.
Article
Full-text available
In this paper, the hierarchical ways for building a regression model by using bootstrap and jackknife resampling methods were presented. Bootstrap approaches based on the observations and errors resampling, and jackknife approaches based on the delete-one and delete-d observations were considered. And also we consider estimating bootstrap and jackknife bias, standard errors and confidence intervals of the regression coefficients, and comparing with the concerning estimates of ordinary least squares. Obtaining of the estimates was presented with an illustrative real numerical example. The jackknife bias, the standard errors and confidence intervals of regression coefficients are substantially larger than the bootstrap and estimated asymptotic OLS standard errors. The jackknife percentile intervals also are larger than to the bootstrap percentile intervals of the regression coefficients.
Article
Full-text available
This paper reviews the principal approaches to using Artificial Intelligence in Music Education. Music is a challenging domain for Artificial Intelligence in Education (AI-ED) because music is, in general, an open-ended domain demanding creativity and problem-seeking on the part of learners and teachers. In addition, Artificial Intelligence theories of music are far from complete, and music education typically emphasises factors other than the communication of 'knowledge' to students. This paper reviews critically some of the principal problems and possibilities in a variety of AI-ED approaches to music education. Approaches considered include: Intelligent Tutoring Systems for Music; Music Logo Systems; Cognitive Support Frameworks that employ models of creativity; highly interactive interfaces that employ AI theories; AI-based music tools; and systems to support negotiation and reflection. A wide variety of existing music AI-ED systems are used to illustrate the key issues, techniques and methods associated with these approaches to AI-ED in Music.
Article
Full-text available
An intelligent tutoring system customizes its presentation of knowledge to the individual needs of each student based on a model of the student. Student models are more complex than other user models because the student is likely to have misconceptions. We have addressed several difficult issues in reasoning about a student's knowledge and skills within a real-time simulation-based training system. Our conceptual framework enables important aspects of the tutor's reasoning to be based upon simple, comprehensible representations that are the basis for a Student Centered Curriculum. We have built a system for teaching cardiac resuscitation techniques in which the decisions abouthow to teach are separated from the decisions aboutwhat to teach. The training context (i.e., choice of topics) is changed based on a tight interaction between student modeling techniques and simulation management. Although complex student models are still required to support detailed reasoning about how to teach, we argue that the decision about what to teach can be adequately supported by qualitatively simpler techniques, such as overlay models. This system was evaluated in formative studies involving medical school faculty and students. Construction of the student model involves monitoring student actions during a simulation and evaluating these actions in comparison with an expert model encoded as a multi-agent plan. The plan recognition techniques used in this system are novel and allow the expert knowledge to be expressed in a form that is natural for domain experts.
Article
Full-text available
This paper reports on an empirical study of the interactive means by which a human teacher supported higher-order, musical thinking in learners. The aim of the work presented here was to create guidelines on how to make practical use of empirical research in teaching agent development. The research overview given in the introduction is followed by an illustrative example of mentoring interactions in musical learning. An elaboration of a musical education problem is then given and a description of the mentoring approach that has been proposed to address this problem is presented. This work uses a theoretical approach to modelling mentoring interactions called the Knowledge Mentoring framework (KMf). An empirical study designed to investigate the musical education problem is reported. In the study, a human teacher attempted to support, in the learners, creative, metacognitive and critical thinking by making use of a specially designed computer-based learning environment called Coleridge. The KMf was used to guide the analysis and modelling of the study interactions. The four main results were a pause taxonomy, reciprocal modelling, seven distinct mentoring stages (with associated state transition networks), and a script of the most frequent mentoring interactions. The last section of the paper presents a discussion of the main implications of these results for Intelligent Learning Environment (ILE) design. (http://aied.inf.ed.ac.uk/members98/archive/vol_9/cook/full.html)
Conference Paper
This paper discusses the design of an explanation facility for answering student-initiated questions. The facility is based on a hybrid planning architecture that models functionally distinct subtasks of explanation planning with mechanisms suited for those tasks. The paper focuses on mechanisms for four of these tasks: selection of knowledge relevant to a query, choice of conceptual model on which to base the explanation, addition of pedagogically motivated material, and coherent ordering. Specific recommendations are made for the design of an ITS capable of providing these functionalities when responding to student initiated queries.
Article
In this article, we describe a deployed educational technology application: the Criterion Online Essay Evaluation Service, a web-based system that provides automated scoring and evaluation of student essays. Criterion has two complementary applications: (1) Critique Writing Analysis Tools, a suite of programs that detect errors in grammar, usage, and mechanics, that identify discourse elements in the essay, and that recognize potentially undesirable elements of style, and (2) e-rater version 2.0, an automated essay scoring system. Critique and e-rater provide students with feedback that is specific to their writing in order to help them improve their writing skills and is intended to be used under the instruction of a classroom teacher. Both applications employ natural language processing and machine learning techniques. All of these capabilities outperform baseline algorithms, and some of the tools agree with human judges in their evaluations as often as two judges agree with each other.
Article
A large high-speed general-purpose digital computer (IBM 7090) was pro- grammed to solve elementary symbolic integration problems at approximately the level of a good college freshman. The program is called SAINT, an acronym for "Symbolic Automatic INTegrator." This paper discusses the SAINT program and its performance. SAINT performs indefinite integration. It also performs definite and multiple integration when these are trivial extensions of indefinite integration. It uses many of the methods and heuristics of students attacking the same problems. SAINT took an average of two minutes each to solve 52 of the 54 attempted problems taken from the Massachusetts Institute of Technology freshman calculus final examinations. Based on this and other experiments with SAINT, some conclusions concerning computer solution of such problems are: (1) Pattern recognition is of fundamental importance. (2) Great benefit would have been derived from a larger memory and more convenient symbol manipulating facilities. (3) The solution of a symbolic integration problem by a commercially available computer is far cheaper and faster than by man.
Article
SIN and SOLDIER are heuristic programs in LISP which solve symbolic integration problems. SIN (Symbolic INtegrator) solves indefinite integration problems at the difficulty approaching those in the larger integral tables. SIN contains several more methods than are used in the previous symbolic integration program SAINT, and solves most of the problems attempted by SAINT in less than one second. SOLDIER (SOLution of Ordinary Differential Equations Routine) solves first order, first degree ordinary differential equations at the level of a good college sophomore and at an average of about five seconds per problem attempted. The differences in philosophy and operation between SAINT and SIN are described, and suggestions for extending the work presented are made.