Elicitation of latent learning needs through learning goals recommendation
Introduction
A significant educational action able to guide the learner in a comprehensive learning process is not only focused on learning (cognition level) but also on fostering a correct learning behavior that empowers learners to achieve their learning goals in a controlled and directed way (metacognition level) (Mangione, Gaeta, Orciuoli, & Salerno, 2010).
Starting from this principle we defined and developed an e-learning system able to build personalized learning experiences starting from a set of target concepts selected on an ontology-based domain model (Capuano, Gaeta, Miranda, Orciuoli, & Ritrovato, 2008). We then extended such system in order to allow course generation from an explicit request in terms of needs to be satisfied and expressed by the learner in natural language (CapuaCapuano, Gaeta, Orciuoli, & Ritrovato, 2009).
The work presented in this paper deals with the definition of a further process of course building starting from an implicit request rather than from an explicit one. In other words, a methodology to recommend learning goals based on the analysis of a learner’ profile (including known topics) and on the comparison of this profile with profiles of similar learners is defined.
The proposed methodology upholds the social presence (Acampora et al., 2010, Capuano et al., 2010), while supporting the development of self-regulated learning. Educational recommendations serves as a pedagogical advance organizer for the learners, as it anticipates and spreads needs, knowledge and learning paths. Furthermore the proposed solution also supports help seeking processes improving the students’ control over learning. This makes the solution adequate nononly for educational settings but also for enterprise training (Capuano, Gaeta, Ritrovato, & Salerno, 2008).
The paper is organized in this way: Section 2 introduces some background about recommender systems; Section 3 briefly introduces the starting point of our research and then describes the proposed methodology; Section 4 introduces the developed prototype and presents some example of use; Section 5 compares our approach with some existing recommender systems for e-learning; eventually Section 6 presents conclusions and planned future work.
Section snippets
Background on recommender systems
Recommender Systems (RS) are aimed at providing personalized recommendations on the utility of a set of objects belonging to a given domain, starting from the information available about users and objects.
A formal definition of the recommendation problem can be expressed in these terms (Adomavicius & Tuzhilin, 2005): C is the set of users of the system, I the set of objects that can be recommended, R a totally ordered set whose values represent the utility of an object for a user (e.g. integers
The proposed approach
In this section we describe the methodology we have defined to recommend learning goals to users of an existing learning system named IWT (Intelligent Web Teacher). First of all a brief introduction to IWT is provided in the next sub-section as well as some fundamentals on Upper Level Learning Goals (ULLGs): a user friendly way (using natural language) for the expression of learning needs provided by IWT.
After having introduced the starting point, a methodology to recommend ULLGs basing on the
The developed prototype
In order to experiment the proposed approach, we designed and developed a prototype recommender system for ULLG and integrated it with IWT. In the following sub-sections we present a high-level view of the prototype architecture, give some details about its user interface and show an example of use.
Related work
Several recommender systems for e-Learning have been introduced to select and propose learning resources to users. Some of them are still at prototype stage while some of them are full systems (Bodea, Dascalu, & Lytras, 2012). One of the first collaborative recommenders for learning resources has been Altered Vista (Recker & Wiley, 2001) whose goal was to explore how to collect user-provided evaluations about learning resources, and to use them to recommend, to the members of a community, both
Experimentation context and approach
To evaluate the prototype and analyze its effects in a learning process, we experimented it with real users within a University setting. In particular, 170 students enrolled in an online course on Software Engineering were involved in the experiment.
68 out of 170 students (40%) participated actively in the experience. We considered active participation the submission of an evaluation form at the end of the experience. Since the experiment was optional for all students, 60% of them chose not to
Conclusions and future work
We defined in this paper a methodology to recommend learning goals and to generate learning experiences and a prototype component integrated in an commercial adaptive e-learning system named IWT. We compared the proposed approach with similar existing systems and prototypes facing the problem of learning resources and learning goals recommendation.
The first evaluation provided encouraging results considering the prototype nature of the environment. A more extensive experimentation is currently
Acknowledgements
The research reported in this paper is partially supported by the European Commission under the Collaborative Project ALICE “Adaptive Learning via an Intuitive, interactive, Collaborative, Emotional system”, VII Framework Program, Theme ICT-2009.4.2, Grant Agreement n. 257639.
References (33)
- et al.
Advanced ontology management system for personalised e-learning
Journal of Knowledge-Based Systems
(2009) - et al.
QSIA-a Web-based environment for learning, assessing and knowledge sharing in communities
Computers, Education
(2004) - Acampora, G., Gaeta, M., Orciuoli, F., & Ritrovato, P. (2010). Exploiting semantic and social technologies for...
- et al.
Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions
IEEE Transactions on Knowledge and Data Engineering
(2005) - et al.
IWT: An innovative solution for AGS e-Learning model
International Journal of Knowledge and Learning, Inderscience
(2007) - Anderson, M., Ball, M., Boley, H., Greene, S., Howse, N., Lemire, D., & McGrath, S. (2003). RACOFI: A Rule-Applying...
- et al.
User recommendation for collaborative and personalised digital archives
International Journal of Web Based Communities
(2005) - et al.
Fab: Content-based, collaborative recommenddation
Communications of the ACM
(1997) - Bercovitz, B., Kaliszan, F., Koutrika, G., Liou, H., Zadeh, Z.M., Garcia-Molina, H. (2009). CourseRank: A social system...
- et al.
A recommender engine for advanced personalized feedback in e-Learning environments
International Journal of Engineering Education
(2012)
SUS: A “quick and dirty” usability scale
Hybrid Web recommender systems lecture notes in computer science
Cited by (37)
A proposal for an adaptive Recommender System based on competences and ontologies
2022, Expert Systems with ApplicationsCitation Excerpt :First criterion indicates whether the recommender system is accessed through a web application (25 of 27), a desktop application (0 of 27) or both possibilities (Duran et al., 2016; Mao et al., 2015). Second criterion details if the recommender system is a stand-alone application (18 of 27), if it is embedded in other system (Capuano et al., 2014; Chavarriaga et al., 2014; Emmenegger et al., 2016; Montuschi et al., 2015; Paquette, 2016; Santos & Boticario, 2008; Sielis et al., 2011; Wang, 2016) or if both values are possible (Duran et al., 2016). The target user is a very relevant criterion since it refers to the person who receives the recommendation.
Assisted education: Using predictive model to avoid school dropout in e-learning systems
2021, Intelligent Systems and Learning Data Analytics in Online EducationA fog based recommendation system for promoting the performance of E-Learning environments
2020, Computers and Electrical EngineeringFuzzy Group Decision Making for influence-aware recommendations
2019, Computers in Human BehaviorCitation Excerpt :In recent years, Recommender Systems (RSs) have become increasingly popular to support several human activities by anticipating needs and generating personalized suggestions for users. They are currently adopted in a variety of areas for the selection of items like movies, music, news, books, courses, papers, products, etc. (Adomavicius & Tuzhilin, 2005; Capuano, Mangione, Pierri, & Salerno, 2012; Capuano, Gaeta, Ritrovato, & Salerno, 2014). Even if the majority of RSs provides recommendations for individual users, there are activities that can be performed by groups of people, like watching a movie, going to a restaurant or traveling with friends.
A learner oriented learning recommendation approach based on mixed concept mapping and immune algorithm
2016, Knowledge-Based SystemsCitation Excerpt :Collaborative filtering is limited to cold start and sparsity problems [3]. Hybrid approach is a more reliable method for personalized recommendation [4,5], and this approach is widely applied, especially on the recommendations of e-commerce [6–9]. However, it is obvious that content-based recommendation is the fundamental approach and it is important for personalized recommendation.
DHM-OCR: A Deep Hybrid Model for Online Course Recommendation and Sustainable Development of Education
2024, International Journal of Electrical and Computer Engineering Systems