Article

Metadata for Recommending Primary and Secondary Level Learning Resources

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Recommender systems have been used in education to assist users in the discovery of learning resources. Unlike product-oriented recommender systems, the goals and behavior of users in education are influenced by their context; such influence may be stronger in formal scenarios such as primary and secondary education since context is highly regulated. Intuitively, we could assume that a biology teacher may be more interested in biology-related content rather than content from other fields. In this paper we explore such assumption by analyzing the impact of educational metadata that is associated to resources and teachers. We apply hierarchical clustering to determine clusters of interest and using a teacher profile, we classify new teachers and new items in order to predict their preferences. In order to validate our approach, we used a dataset derived from a repository of learning resources widely used by teachers in primary and secondary school in Chile in the role of old users, we also performed an experiment with teachers in training in the role of new users. Our results confirm the diverse impact of metadata on the formation of such clusters and on recommendation.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

Article
Educational resources can be easily found on the Web; most search engines base their algorithms on the resources text or popularity. Teachers must navigate the results until they find an appropriate resource making search a tedious and cumbersome task. Specialized repositories contain resources that are annotated with metadata that aims to facilitate the discovery of quality resources. Nevertheless, content abundance and variety makes searching a complex task. Recommender systems can assist teachers in finding the proper content either by determining clusters of similar users, and inferring the user interest on a resource; clusters of similar resources, or a mix of both. Probabilistic model based (PMB) techniques, on the other hand, make possible to classify resources into more than one cluster with diverse probability degree. In this paper we use recommender systems and PMB techniques to analyze a dataset produced by primary and secondary level teachers during four years and under natural conditions. We found that a hybrid recommendation, the CTR technique, performs better than other approaches despite the high sparseness of the dataset. In addition, learning resources annotated with curriculum metadata had a positive impact on recommenders accuracy whereas free-text or other metadata negatively impacted the results.
Article
Full-text available
This paper describes a software system that allows for discovering nontraditional educational resources, that is to say, those that go beyond educational content and incorporate elements such as: software applications that may support the teaching-learning process; events that take place outside school boundaries, such as new expositions in museums and theatre performances; and people who may participate as experts in some Learning Activity. The huge quantity of information available potentially all that can be extracted from the Internet- enforces us to adopt a strategy that enables filtering resources in accordance with their appropriateness and relevance, that is, an strategy based on recommendations. Besides, due to their particular nature (e.g. the most relevant events are those that will take place in the same city where the school is located) the apropriateness of those resources is highly dependent on the context where teaching and learning is produced. Therefore, the recommender system takes into account contextual factors when calculating the relevance of every resource. This system was evaluated with several focus groups in the scope of the iTEC project, which belongs to the European Commision's Framework Programme 7.
Article
Full-text available
Learner profiling is a methodology that draws a parallel from user profiling. Implicit feedback is often used in recommender systems to create and adapt user profiles. In this work the implicit feedback is based on the learner's answering behaviour in the Android application UnlockYourBrain, which poses different basic mathematical questions to the learners. We introduce an analytical approach to model the learners' profile according to the learner's answering behaviour. Furthermore, similar learner's profiles are grouped together to construct a learning behaviour cluster. The choice of hierarchical clustering as a means of classification of learners' profiles derives from the observations of learners behaviour. This in turn reflects the similarities and subtle differences of learner behaviour, which are further analysed in more detail. Building awareness about the learner's behaviour is the first and necessary step for future learning-aware applications.
Article
Full-text available
The personal learning environment (PLE) concept offers a learner-centric view of learning and suggests a shift from knowledge-push to knowledge-pull approach to learning. One concern with a PLE-driven knowledge-pull approach to learning, however, is information overload. Recommender systems can provide an effective mechanism to deal with the information overload problem in PLEs. In this paper, we study different tag-based collaborative filtering recommendation techniques on their applicability and effectiveness in PLE settings. We implement 16 different tag-based collaborative filtering recommendation algorithms, memory based as well as model based, and compare them in terms of accuracy and user satisfaction. The results of the conducted offline and user evaluations reveal that the quality of user experience does not correlate with high-recommendation accuracy.
Chapter
Full-text available
Technology enhanced learning (TEL) aims to design, develop and test socio-technical innovations that will support and enhance learning practices of both individuals and organisations. It is therefore an application domain that generally covers technologies that support all forms of teaching and learning activities. Since information retrieval (in terms of searching for relevant learning resources to support teachers or learners) is a pivotal activity in TEL, the deployment of recommender systems has attracted increased interest. This chapter attempts to provide an introduction to recommender systems for TEL settings, as well as to highlight their particularities compared to recommender systems for other application domains.
Book
Full-text available
Technology enhanced learning (TEL) aims to design, develop and test sociotechnical innovations that will support and enhance learning practices of both individuals and organisations. It is therefore an application domain that generally covers technologies that support all forms of teaching and learning activities. Since information retrieval (in terms of searching for relevant learning resources to support teachers or learners) is a pivotal activity in TEL, the deployment of recommender systems has attracted increased interest. This brief attempts to provide an introduction to recommender systems for TEL settings, as well as to highlight their particularities compared to recommender systems for other application domains.
Article
Full-text available
Recommender systems have been researched extensively by the Technology Enhanced Learning (TEL) community in the last decade. By identifying suitable resources from a potentially overwhelming variety of choices, such systems offer a promising approach to facilitate both learning and teaching tasks. As learning is taking place in extremely diverse and rich environments, the incorporation of contextual information about the user in the recommendation process has attracted major interest. Such contextualization is researched as a paradigm for building intelligent systems that can better predict and anticipate the needs of users, and act more efficiently in response to their behavior. In this paper, we try to assess the degree to which current work in TEL recommender systems has achieved this, as well as outline areas in which further work is needed. First, we present a context framework that identifies relevant context dimensions for TEL applications. Then, we present an analysis of existing TEL recommender systems along these dimensions. Finally, based on our survey results, we outline topics on which further research is needed.
Article
Full-text available
The apparent difficulty in assessing emotions elicited by movies and the undeniable high variability in subjects' emotional responses to film content have been recently tackled by exploring film connotative properties: the set of shooting and editing conventions that help in transmitting meaning to the audience. Connotation provides an intermediate representation that exploits the objectivity of audiovisual descriptors to predict the subjective emotional reaction of single users. This is done without the need of registering users' physiological signals. It is not done by employing other people's highly variable emotional rates, but by relying on the intersubjectivity of connotative concepts and on the knowledge of user's reactions to similar stimuli. This paper extends previous work by extracting audiovisual and film grammar descriptors and, driven by users' rates on connotative properties, creates a shared framework where movie scenes are placed, compared, and recommended according to connotation. We evaluate the potential of the proposed system by asking users to assess the ability of connotation in suggesting film content able to target their affective requests.
Article
Full-text available
Collaborative recommendation is effective at representing a user's overall interests and tastes, and finding peer users that can provide good recommendations. However, it re-mains a challenge to make collaborative recommendation sensitive to a user's specific context and to the changing shape of user interests over time. Our approach to building context-sensitive collaborative recommendation is a hybrid one that incorporates semantic knowledge in the form of a domain ontology. User profiles are defined relative to the ontology, giving rise to an ontological user profile. In this paper, we describe how ontological user profiles are learned, incrementally updated, and used for collaborative recom-mendation. Using book rating data, we demonstrate that this recommendation algorithm offers improved coverage, di-versity, personalization, and cold-start performance while at the same time enhancing recommendation accuracy.
Article
Full-text available
Collaborative filtering systems predict a user's interest in new items based on the recommendations of other people with similar interests. Instead of performing content indexing or content analysis, collaborative filtering systems rely entirely on interest ratings from members of a participating community. Since predictions are based on human ratings, collaborative filtering systems have the potential to provide filtering based on complex attributes, such as quality, taste, or aesthetics. Many implementations of collaborative filtering apply some variation of the neighborhood-based prediction algorithm. Many variations of similarity metrics, weighting approaches, combination measures, and rating normalization have appeared in each implementation. For these parameters and others, there is no consensus as to which choice of technique is most appropriate for what situations, nor how significant an effect on accuracy each parameter has. Consequently, every person implementing a collaborative filtering system must make hard design choices with little guidance. This article provides a set of recommendations to guide design of neighborhood-based prediction systems, based on the results of an empirical study. We apply an analysis framework that divides the neighborhood-based prediction approach into three components and then examines variants of the key parameters in each component. The three components identified are similarity computation, neighbor selection, and rating combination.
Article
Full-text available
Deploying semantic tools and services over a field of linked data could be a way to address many current challenges of higher education. The strengths of semantic technologies for learning and teaching, and their benefits for digital libraries, virtual communities, and e-learning, have been a major topic of discussion during recent years. Experts argue that semantic technologies can enhance the advanced learning experience by using the expressive power of metadata to describe learning content, people, and services, and then matching these intelligently.
Conference Paper
Full-text available
In the world of recommender systems, it is a common practice to use public available datasets from different application environments (e.g. MovieLens, Book-Crossing, or Each-Movie) in order to evaluate recommendation algorithms. These datasets are used as benchmarks to develop new recommendation algorithms and to compare them to other algorithms in given settings. In this paper, we explore datasets that capture learner interactions with tools and resources. We use the datasets to evaluate and compare the performance of different recommendation algorithms for learning. We present an experimental comparison of the accuracy of several collaborative filtering algorithms applied to these TEL datasets and elaborate on implicit relevance data, such as downloads and tags, that can be used to improve the performance of recommendation algorithms.
Article
Full-text available
In contrast to customers of bricks and mortar stores, users of online selling environments are not supported by human sales experts. In such situations recommender applications help to identify the products and/or services that fit the user’s wishes and needs. In order to successfully apply recommendation technologies we have to develop an in-depth understanding of decision strategies of users. These decision strategies are explained in different models of human decision making. In this paper we provide an overview of selected models and discuss their importance for recommender system development. Furthermore, we provide an outlook on future research issues.
Article
Full-text available
This article argues that there is a need for Personal Recommender Systems (PRSs) in Learning Networks (LNs) in order to provide learners advice on the suitable learning activities to follow. LNs target lifelong learners in any learning situation, at all educational levels and in all national contexts. They are community-driven because every member is able to contribute to the learning material. Existing Recommender Systems (RS) and recommendation techniques used for consumer products and other contexts are assessed on their suitability for providing navigational support in an LN. The similarities and differences are translated into specific requirements for learning and specific requirements for recommendation techniques. The article focuses on the use of memory-based recommendation techniques, which calculate recommendations based on the current data set. We propose a combination of memory-based recommendation techniques that appear suitable to realise personalised recommendation on learning activities in the context of e-learning. An initial model for the design of such systems in LNs and a roadmap for their further development are presented.
Article
Full-text available
E-learning has the potential to offer significant advantages over traditional classroom learning. It however requires a complete redefinition of the dynamics of interaction between the various actors of a classroom. Moreover, in this context, the authoring of instructional material requires much more time than in traditional learning. Therefore, a special care has to be posed to the definition of proper authoring approaches where educators can reuse and easily assemble existing materials. In this scenario, a comprehensive learning platform addressing the various interrelated aspects of authoring and fruition of instructional material is needed. Such a platform should enable reusability of materials, so that it is possible to make efficient use of preexisting experiences, and interoperability with existing platforms so that it is possible to take advantage of their strengths. The SCORM standard offers, among other features, a rich data model that can be used to define and share Learning Objects through different e-learning platforms. We argue that, despite the fact that it is the most emerging and promising standard, SCORM does not address some key issues properly, such as specification of metadata and LO composition. In this paper we focus on these issues and propose some extensions to SCORM that aim to address the above issues.
Article
Full-text available
Recommender systems have already been engaging multiple criteria for the production of recommendations. Such systems, referred to as multicriteria recommenders, demonstrated early the potential of applying Multi-Criteria Decision Making (MCDM) methods to facilitate recommendation in numerous application domains. On the other hand, systematic implementation and testing of multicriteria recommender systems in the context of real-life applications still remains rather limited. Previous studies dealing with the evaluation of recommender systems have outlined the importance of carrying out careful testing and parameterization of a recommender system, before it is actually deployed in a real setting. In this paper, the experimental analysis of several design options for three proposed multiattribute utility collaborative filtering algorithms is presented for a particular application context (recommendation of e-markets to online customers), under conditions similar to the ones expected during actual operation. The results of this study indicate that the performance of recommendation algorithms depends on the characteristics of the application context, as these are reflected on the properties of evaluations' data set. Therefore, it is judged important to experimentally analyze various design choices for multicriteria recommender systems, before their actual deployment.
Article
Full-text available
Drachsler, H., Bogers, T., Vuorikari, R., Verbert, K., Duval, E., Manouselis, N., Beham, G., Lindstaedt, S., Stern, H., Friedrich, M., & Wolpers, M. (2010). Issues and considerations regarding sharable data sets for recommender systems in technology enhanced learning. In N. Manouselis, H. Drachsler, K. Verbert, & O. Santos (Eds.), Elsevier Procedia Computer Science: Volume 1, Issue 2. Proceedings of the 1st Workshop on Recommender Systems for Technology Enhanced Learning (RecSysTEL 2010) (pp. 2849-2858). doi: 10.1016/j.procs.2010.08.010.
Article
Full-text available
Recommender systems have been evaluated in many, often incomparable, ways. In this article, we review the key decisions in evaluating collaborative filtering recommender systems: the user tasks being evaluated, the types of analysis and datasets being used, the ways in which prediction quality is measured, the evaluation of prediction attributes other than quality, and the user-based evaluation of the system as a whole. In addition to reviewing the evaluation strategies used by prior researchers, we present empirical results from the analysis of various accuracy metrics on one content domain where all the tested metrics collapsed roughly into three equivalence classes. Metrics within each equivalency class were strongly correlated, while metrics from different equivalency classes were uncorrelated.
Article
Full-text available
Recommender systems for e-learning demand specific pedagogy-oriented and hybrid recommendation strategies. Current systems are often based on time-consuming, top down information provisioning combined with intensive data-mining collaborative filtering approaches. However, such systems do not seem appropriate for Learning Networks where distributed information can often not be identified beforehand. Providing sound way-finding support for lifelong learners in Learning Networks requires dedicated personalised recommender systems (PRS), that offer the learners customised advise on which learning actions or programs to study next. Such systems should also be practically feasible and be developed with minimized effort. Currently, such so called light-weight PRS systems are scarcely available. This study shows that simulation studies can support the analysis and optimisation of PRS requirements prior to starting the costly process of their development, and practical implementation (including testing and revision) during field experiments in real-life learning situations. This simulation study confirms that providing recommendations leads towards more effective, more satisfied, and faster goal achievement. Furthermore, this study reveals that a light-weight hybrid PRS-system based on ratings is a good alternative for an ontology-based system, in particular for low-level goal achievement. Finally, it is found that rating-based light-weight hybrid PRS-systems enable more effective, more satisfied, and faster goal attainment than peer-based light-weight hybrid PRS-systems (incorporating collaborative techniques without rating).
Article
Full-text available
this paper is published as: Koper, E.J.R. (2003). Combining re-usable learning resources and services to pedagogical purposeful units of learning. In A. Littlejohn (Ed.), Reusing Online Resources: A Sustainable Approach to eLearning (pp. 46-59). London: Kogan Page. technology-supported learning". Wiley (2002, p.6) advocates strongly that this definition is too broad to be of any practical or scientific value. He proposes as an alternative definition: "any digital resource that can be reused to support learning". I will further narrow down the scope of this definition for this chapter to: "any digital, reproducible and addressable resource used to perform learning activities or learning support activities, made available for others to use". This definition excludes a lot of things: for example non-digital materials, non-reproducible unique exemplars and non-addressable resources (for example those that are not connected with a URL and metadata for access). It excludes learning activities, and also learning objectives and prerequisites, since these are a function of the learning activities, not the resources. It also excludes courses (since these are aggregates of learning objects and learning activities) as well as 'people', 'activities' and 'services'. Further sub-classification of learning objects has been introduced in chapters 1 and 2 and, in this chapter, I will use the following definitions: - Knowledge objects are learning objects which contain information for people to learn from or to use while supporting the learning activities of others (for example teachers with students). An example is a web page with a series of information objects to learn, (e.g. about sensory systems); or a teachers' manual. - Tool objects are learning objects to learn with or to use whi...
Conference Paper
Collaborative Filtering recommendation algorithm is one of the most popular approaches for determining recommendations at present and it can be used to solve Information Overload issue in e-Learning system. However the Cold Start problem is always one of the most critical issues that affect the performance of Collaborative Filtering recommender system. In this paper an enhanced composite recommendation algorithm based on content recommendation tags extracting and CF is proposed to make the CF recommender system work more effectively. The final experiment results show that the new enhanced recommendation algorithm has some advantages on accuracy compared with several existing solutions to the issue of Cold Start and make sure that it is a feasible and effective recommendation algorithm.
Article
This paper presents an overview of the field of recommender systems and describes the current generation of recommendation methods that are usually classified into the following three main categories: content-based, collaborative, and hybrid recommendation approaches. This paper also describes various limitations of current recommendation methods and discusses possible extensions that can improve recommendation capabilities and make recommender systems applicable to an even broader range of applications. These extensions include, among others, an improvement of understanding of users and items, incorporation of the contextual information into the recommendation process, support for multicriteria ratings, and a provision of more flexible and less intrusive types of recommendations.
Chapter
Simply matching learner interest with paper topic is far from enough in making personalized paper recommendations to learners in the educational domain. As such, we proposed the multidimensional recommendation techniques that consider (educational) context-aware information to inform and guide the system during the recommendation process. The contextual information includes both learner and paper features that can be extracted and learned during the pre- and post-recommendation process. User studies have been performed on both undergraduate (inexperienced learners) and graduate (experienced learners) students who have different information-seeking goals and educational backgrounds. Results from our extensive studies have been able to show that (1) it is both effective and desirable to implement the multidimensional recommendation techniques that are more complex than the traditional single-dimensional recommendation; (2) recommendation from across different learning groups (with different pedagogical features and learning goals) is less effective than that from within the same learning groups, especially when collaborative filtering technique is applied.
Chapter
Recommender systems have already been engaging multiple criteria for the production of recommendations. Such systems, referred to as multi-criteria recommenders, early demonstrated the potential of applying Multi-Criteria Decision Making (MCDM) methods to facilitate recommendation in numerous application domains. On the other hand, systematic implementation and testing of multi-criteria recommender systems in the context of real-life applications still remains rather limited. Previous studies dealing with the evaluation of recommender systems have outlined the importance of carrying out careful testing and parameterization of a recommender system, before it is actually deployed in a real setting. In this paper, the experimental analysis of several design options for three proposed multi-attribute utility collaborative filtering algorithms is presented. The data set used is synthetic, with multi-criteria evaluations that have been created using an appropriate simulation environment. This synthetic data set tries to mimic the evaluations that are expected to be collected from users in a particular application setting. The aim of the experiment is to demonstrate how a synthetic data set may be created and used to facilitate the study and selection of an appropriate recommendation algorithm, in the case that multi-criteria evaluations from real users are not available.
Article
Knowledge contribution is a very important issue in online engineering education. This study investigates how to encourage knowledge contribution in a large Asia online engineering education project. Based on long-term empirical research, we found economic incentive is more critical to facilitate explicit knowledge contribution, while social incentive is more important in encouraging tacit knowledge contribution. Moreover, we also found incentives may have different effects due to students' individual difference, e.g., personal value. That is, economic incentive may have positive effect on some students, but none or negative to others. These findings provides some theoretical contributions on e-learning and knowledge management, and also provides some implications for choosing technology in online engineering education system, and designing incentive mechanism.
Article
Recommender systems are used to provide filtered information from a large amount of elements. They provide personalized recommendations on products or services to users. The recommendations are intended to provide interesting elements to users. Recommender systems can be developed using different techniques and algorithms where the selection of these techniques depends on the area in which they will be applied. This paper proposes a recommender system in the leisure domain, specifically in the movie showtimes domain. The system proposed is called RecomMetz, and it is a context-aware mobile recommender system based on Semantic Web technologies. In detail, a domain ontology primarily serving a semantic similarity metric adjusted to the concept of “packages of single items” was developed in this research. In addition, location, crowd and time were considered as three different kinds of contextual information in RecomMetz. In a nutshell, RecomMetz has unique features: (1) the items to be recommended have a composite structure (movie theater + movie + showtime), (2) the integration of the time and crowd factors into a context-aware model, (3) the implementation of an ontology-based context modeling approach and (4) the development of a multi-platform native mobile user interface intended to leverage the hardware capabilities (sensors) of mobile devices. The evaluation results show the efficiency and effectiveness of the recommendation mechanism implemented by RecomMetz in both a cold-start scenario and a no cold-start scenario.
Article
The aim of the paper is to present and compare so-called bottom-up and top-down approaches for evaluating quality and reusability of learning objects (LOs). The paper proposes bottom-up methodology that outlines the central role of learners' individual and social behaviour while working with LOs. This includes social tagging and some parameters of interaction for measuring context to describe LOs usage, attention, and other aspects of the context as well as helps to exploit context data towards making LOs repositories more useful, and thus enhance the reuse. The paper also presents top-down methodology for the expert evaluation of LOs quality and reusability. This methodology consists of consecutive application of several scientific approaches, methods, and principles. The authors investigate how these two different, however complimentary approaches for evaluating LOs quality and reusability can be better applied for the aims of eQNet project in order to select reusable (''travel well'') LOs for implementing in different educational contexts and countries. These approaches could be considered as suitable applications of information and communication technologies (ICT's) for development of human capital. The examples of practical application of these approaches for evaluating LOs quality and reusability in eQNet project are also presented in the paper.
Article
A recommender system (RS) aims to provide personalized recommendations to users for specific items (e.g., music, books). Popular techniques involve content-based (CB) models and collaborative filtering (CF) approaches. In this paper, we deal with a very important problem in RSs: The cold start problem. This problem is related to recommendations for novel users or new items. In case of new users, the system does not have information about their preferences in order to make recommendations. We propose a model where widely known classification algorithms in combination with similarity techniques and prediction mechanisms provide the necessary means for retrieving recommendations. The proposed approach incorporates classification methods in a pure CF system while the use of demographic data help for the identification of other users with similar behavior. Our experiments show the performance of the proposed system through a large number of experiments. We adopt the widely known dataset provided by the GroupLens research group. We reveal the advantages of the proposed solution by providing satisfactory numerical results in different experimental scenarios.
Article
The current study proposes an intelligent approach to compose optimal learning groups in which the members have different domain backgrounds. The approach is based on a well-known evolutionary algorithm - Particle Swarm Optimization. The authors claim that quantifying various indicators, such as background diversity and similarity between the type of interest of the participants, within a group and between groups can positively impact on building learning groups. The algorithm is integrated in an ontology-based e-learning system, designed to create self-built educating communities, in which a trainees goes through the education process, gains points through achievements and ultimately becomes a trainer. When creating a new account, the newly created trainee is asked to self asses himself by filling out a form. The resulting profile is used to assign the user to the most suitable learning group. We propose to assign him by the following rule: maximizing the diversity within a group (due to the fact that multidisciplinary teams are more challenging) and minimizing the diversity between groups (all the groups should have similar composition), meaning a group will have members with similar interests. The study is presented in the context of group building strategies in adults' education.
Article
With the advent of the Social Web and the growing popularity of Web 2.0 applications, recommender systems are gaining momentum. The recommendations generated by these systems aim to provide end users with suggestions about information items, social elements, products or services that are likely to be of their interest. The traditional syntactic-based recommender systems suffer from a number of shortcomings that hamper their effectiveness. As semantic technologies mature, they provide a consistent and reliable basis for dealing with data at the knowledge level. Adding semantically empowered techniques to recommender systems can significantly improve the overall quality of recommendations. In this work, a hybrid recommender system based on knowledge and social networks is presented. Its evaluation in the cinematographic domain yields very promising results compared to state-of-the-art solutions.
Article
In this age of information overload, people use a variety of strategies to make choices about what to buy, how to spend their leisure time, and even whom to date. Recommender systems automate some of these strategies with the goal of providing affordable, personal, and high-quality recommendations. This book offers an overview of approaches to developing state-of-the-art recommender systems. The authors present current algorithmic approaches for generating personalized buying proposals, such as collaborative and content-based filtering, as well as more interactive and knowledge-based approaches. They also discuss how to measure the effectiveness of recommender systems and illustrate the methods with practical case studies. The final chapters cover emerging topics such as recommender systems in the social web and consumer buying behavior theory. Suitable for computer science researchers and students interested in getting an overview of the field, this book will also be useful for professionals looking for the right technology to build real-world recommender systems. © Dietmar Jannach, Markus Zanker, Alexander Felfernig, and Gerhard Friedrich 2011.
Article
Thanks to recent developments on automatic generation of metadata and interoperability between repositories, the production, man- agement and consumption of metadata is vastly surpassing the human capacity to review or process this information. However, we need to as- sure that low quality metadata does not compromise the performance of the services that the repository provides to its users. We contend there is a need for automatic assessment of the quality of metadata in digital repositories, so tools or users can be alerted about low quality records. In this paper, we present several quality metrics for metadata based on quality evaluation frameworks used for human quality review. We applied these metrics to a sample of records from a real repository and compared the results with the quality assessment given to the same records by a group of human reviewers. Through correlation and regression analysis, we found that one of the metrics, the text information content, could be used as a predictor of the human evaluation. While these metrics are not proposed as a definitive measurement of the complete multi-dimensional quality of the metadata record, we present ways in which they can be used to enhance the functionality of digital repositories.
Article
Statistical learning plays a key role in many areas of science, finance and industry. Here are some examples of learning problems:
Article
Repositories with educational resources can support the formation of online learning communities by providing a platform for collaboration. Users (e.g. teachers, tutors and learners) access repositories, search for interesting resources to access and use, and in many cases, also exchange experiences and opinions. A particular class of online services that take advantage of the collected knowledge and experience of users are collaborative filtering ones. The successful operation of such services in the context of real-life applications requires careful testing and parameterization before their actual deployment. In this paper, the case of developing a learning resources' collaborative filtering service for an online community of teachers in Europe was examined. More specifically, a data set of evaluations of learning resources was collected from the teachers that use the European Schoolnet's learning resource portal. These evaluations were then used to support the experimental investigation of design choices for an online collaborative filtering service for the portal's learning resources. A candidate multi-attribute utility collaborative filtering algorithm was appropriately parameterized and tested for this purpose. Results indicated that the development of such systems should be taking place considering the particularities of the actual communities that are to be served.
Article
The publication of different media types, like images, audio and video in the World Wide Web is getting more importance each day. However, searching and locating content in multimedia sites is challenging. In this paper, we propose a platform for the development of multimedia web information systems. Our approach is based on the combination between semantic web technologies and collaborative tagging. Producers can add meta-data to multimedia content associating it with different domain-specific ontologies. At the same time, users can tag the content in a collaborative way. The proposed system uses a search engine that combines both kinds of meta-data to locate the desired content. It will also provide browsing capabilities through the ontology concepts and the developed tags.
Conference Paper
Recommender systems for automatically suggested items of interest to users have become increasingly essential in fields where mass personalization is highly valued. The popular core techniques of such systems are collaborative filtering, content-based filtering and combinations of these. In this paper, we discuss hybrid approaches, using collaborative and also content data to address cold-start - that is, giving recommendations to novel users who have no preference on any items, or recommending items that no user of the community has seen yet. While there have been lots of studies on solving the item-side problems, solution for user-side problems has not been seen public. So we develop a hybrid model based on the analysis of two probabilistic aspect models using pure collaborative filtering to combine with users' information. The experiments with MovieLen data indicate substantial and consistent improvements of this model in overcoming the cold-start user-side problem.
Article
Throughout their lifecycle, learning resources undergo a multitude of processes by being created, used, provided or re-used. However, in order to be reusable, a learning resource often has to be adapted for a new context of use. This in turn implies multiple re-authoring processes being performed on a learning resource. During each of these processes different types of information emerge. When retained, this information can be helpful for the retrieval, authoring, use or re-use of learning resources thereafter. In this paper, the lifecycle of learning resources along with the information generated herein is analyzed and a distributed architecture proposed that allows for the capture, processing, management and utilization of the information mentioned in a generic way. Three steps have been conducted to implement the proposed framework. First evaluation results are promising.
Article
Para lograr coherencia y flexibilidad en unidades de aprendizaje basadas en documentos multimedia, varios autores han recomendado estructurar los componentes de los cursos en grafos. En un grafo de curso, los recursos educacionales son encapsulados como objetos de aprendizaje (LO - Learning Objects) con sus respectivos metadatos (LOM - Learning-Object Metadata) y son interconectados con relaciones de varios tipos retóricos y/o semánticos. Los grafos de recursos son almacenados en repositorios en los cuales los metadatos sirven para facilitar su recuperación y reutilización. Sin embargo, tales sistemas se enfrentan con problemas serios en cuanto al uso de los LOMs: los metadatos son difíciles de instanciar y los autores de cursos generalmente no tienen estímulos para cumplir con esta tediosa tarea ya que ellos mismos no se benefician de los metadatos que generan. La generación automática de metadatos resuelve este problema. Sin embargo, este método se limita a ciertos metadatos excluyendo la mayor parte de los metadatos subjetivos tales como los metadatos educacionales. Esta limitación motivó el enfoque de esta tesis sobre una técnica complementaria: un método híbrido basado en la sinergia entre procesos automáticos e intervención humana. La generación híbrida de LOMs puede ser aplicada sobre los atributos que no pueden ser automáticamente generados. Sin embargo, este enfoque está basado en la contribución de usuarios no siempre cooperativos, quienes necesitarían ver beneficios para motivar su participación. Proponemos estudiar los usos de LOM durante la creación de cursos, no sólo desde la perspectiva de la generación híbrida sino también desde la perspectiva de los beneficios que pueden brindar los LOMs. Esta estrategia tiene como objetivo soportar una retroacción positiva en la cual los beneficios puedan motivar la generación de LOMs de buena calidad, y la buena calidad de los LOMs pueda mejorar los beneficios. En particular, esta tesis investiga métodos para (1) integrar sin transición la generación híbrida de LOMs dentro de una herramienta de creación de cursos, (2) procesar un conjunto de LOMs aunque ciertos metadatos quedaran incompletos, incorrectos, o faltantes, (3) mejorar los resultados de los métodos clásicos de recuperación de LOs usando los metadatos de los LOs que componen un curso. Desarrollamos una herramienta de código abierto para validar las propuestas de esta tesis. Experimentos preliminares mostraron que los LOMs pueden mejorar significativamente la recuperación de LOs adicionales durante el proceso de creación de cursos.
Improving sparsity problem in group recommendation
  • Ghazarian
[Ghazarian et al., 2014] Ghazarian, S., Shabib, N., and Nematbakhsh, M. A. (2014). Improving sparsity problem in group recommendation. In Cena, F., da Silva, A. S., and Trattner, C., editors, Hypertext 2014 Extended Proceedings: Late-breaking Results, Doctoral Consortium and Workshop Proceedings of the 25th ACM Hypertext and Social Media Conference (Hypertext 2014), Santiago, Chile, September 1-4, 2014., volume 1210 of CEUR Workshop Proceedings. CEUR-WS.org.
Using association rules to solve the cold-start problem in recommender systems Personalized recommendation in social tagging systems using hierarchical clustering
  • Shaw
[Shaw et al., 2010] Shaw, G., Xu, Y., and Geva, S. (2010). Using association rules to solve the cold-start problem in recommender systems. In Advances in Knowledge Discovery and Data Mining, pages 340–347. Springer. [Shepitsen et al., 2008] Shepitsen, A., Gemmell, J., Mobasher, B., and Burke, R. (2008). Personalized recommendation in social tagging systems using hierarchical clustering. In Proceedings of the 2008 ACM conference on Recommender systems, pages 259–266. ACM.
  • Colombo-Mendoza
[Colombo-Mendoza et al., 2015] Colombo-Mendoza, L. O., Valencia-Garcia, R., Rodriguez-Gonzalez, A., Alor-Hernandez, G., and Samper-Zapater, J. J. (2015).
On the use of learning object metadata: The globe experience
  • Ochoa
[Ochoa et al., 2011] Ochoa, X., Klerkx, J., Vandeputte, B., and Duval, E. (2011). On the use of learning object metadata: The globe experience. In Towards ubiquitous learning, pages 271–284. Springer.
Capture, management, and utilization of lifecycle information for learning resources. Learning Technologies
et al., 2008] Lehmann, L., Hildebrandt, T., Rensing, C., and Steinmetz, R. (2008). Capture, management, and utilization of lifecycle information for learning resources. Learning Technologies, IEEE Transactions on, 1(1):75-87.
Context-aware recommender systems for learning: a survey and future challenges. Learning Technologies
et al., 2012] Verbert, K., Manouselis, N., Ochoa, X., Wolpers, M., Drachsler, H., Bosnic, I., and Duval, E. (2012). Context-aware recommender systems for learning: a survey and future challenges. Learning Technologies, IEEE Transactions on, 5(4):318-335.