FUSE: A Fuzzy-Semantic Framework for Personalizing Learning Recommendations

Abstract

The use of instructional Semantic Web rules to deliver personalized learning recommendations has become an emerging trend in intelligent tutoring systems (ITSs) because it enables experts’ domain knowledge to be easily transferred to machine-readable formats. However, many approaches to ITS design using instructional Semantic Web rules have evaluated learners’ performances without considering the uncertainty of the evaluation process or other factors such as learning behavior and speed of response. In this paper, we present an ITS framework named FUSE, which uses a novel recommendation-making mechanism based on a fuzzy-semantic reasoning process and a multiagent system. In this framework, fuzzy reasoning and semantic reasoning are integrated to form a unified reasoning process and provide personalized learning recommendations adaptively and semantically. A field experiment was conducted at a public university. The results indicated that FUSE enabled significant achievements in both enhancing learning performance and increasing learners’ participation in the learning process.

Discrete Dynamic Modeling Analysis Based on English Learning Motivation

“…In formulae (17) and (18), T denotes the test data, |T| denotes the size of the test dataset, P u denotes the resource set prerecommended to the user with a predicted score 􏽢 r u ≥ 8, and R u denotes the actual user score r u ≥ 8 users really like the resource set (in this experiment, the user score ≥8 is set as the user's favorite resource). In the case of taking different numbers of adjacent users, the results of the average precision rate and average recall rate are shown in Table 3.…”

“…Chen Jiemin summarized the current personalized recommendation algorithms for learning resources and concluded that there are mainly content-based filtering, associationrule-based, collaborative filtering, and hybrid-based models. Liang Tingting established a learning resource filtering model through content vector space filtering; multidimensional correlation analysis is carried out on learners, resources, situations, etc., to achieve personality matching between resources and learners; Shen Miao designs a collaborative filtering algorithm based on student attribute classification to realize the intelligence and personality of the student course selection system [18]. Resource recommendation: Lei W first obtains association rules through data mining to establish preference matrix and then mixes it with collaborative filtering algorithm for personalized recommendation [19].…”

A Compositional Adaptation-based Approach for Recommending Learning Resources in Software Development

“…Moreover, there are some adaptive/intelligent web-based educational systems, which, not only focus on adaptive presentation and navigation, but also provide intelligent solutions for problem analysis and solving as well as curriculum sequencing [13]. In this respect, a ITS named Fuse has been developed based on both fuzzy and semantic reasoning to provide learning recommendations adaptively [14]. In this way, the user-centered design has shown to be a suitable basis for developing various kinds of learning recommenders.…”

Recommendation Systems in Education: A review of Recommendation Mechanisms in E-learning Environments

“…The authors of [11][12][13] propose a new approach to e-learning systems by integrating fuzzy logic models, the big data framework and the semantic web to make content adaptation to LMS platforms more effective. In [14] and [15] the authors developed a recommendations platform using artificial intelligence that can be adapted and customized for different use cases, taking into account the ecosystem and the daily life of the user.…”