A Fuzzy Logic-based Personalized Learning System for Supporting Adaptive English Learning

As a nearly global language, English as a Foreign Language (EFL) programs are essential for people wishing to learn English. Researchers have noted that extensive reading is an effective way to improve a person's command of English. Choosing suitable articles in accordance with a learner's needs, interests and ability using an elearning system requires precise learner profiles. This paper proposes a personalized English article recommending system, which uses accumulated learner profiles to choose appropriate English articles for a learner. It employs fuzzy inference mechanisms, memory cycle updates, learner preferences and analytic hierarchy process (AHP) to help learners improve their English ability in an extensive reading environment. By using fuzzy inferences and personal memory cycle updates, it is possible to find an article best suited for both a learner’s ability and her/his need to review vocabulary. After reading an article, a test is immediately provided to enhance a learner’s memory for the words newly learned in the article. The responses of tests can be used to explicitly update memory cycles of the newly-learned vocabulary. In addition, this paper proposes a methodology that also implicitly modifies memory cycles of words that were learned before. By intensively reading articles recommended through the proposed approach, learners comprehend new words quickly and review words that they knew implicitly as well, thereby efficiently improving their vocabulary volume. Analyses of learner achievements and questionnaires have confirmed that the adaptive learning method presented in this study not only enhances the English ability of learners but also helps maintaining their learning interest.