An intelligent adaptive fuzzy-based inference system for computer-assisted language learning. (1st August 2019)

• Record Type: Journal Article
• Authors: Troussas, Christos; Chrysafiadi, Konstantina; Virvou, Maria
• Is Part Of: Expert systems with applications. Volume 127(2019)
• Page Start: 85
• Page End: 96
• Publication Date: 2019-08-01
• Subjects: Adaptivity -- Misconception detection and identification -- Inference system -- Intelligent Tutoring Systems -- Second language acquisition -- Personalization; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2019.03.003
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Fuzzy inference system for dynamic delivery of language learning material. Knowledge inference relationships among learning objects for personalized learning. Hybrid model for misconception diagnosis and identification using machine learning. Framework of machine learning and fuzzy logic for individualized language learning. Implementation of expert system and evaluation using an established framework. Abstract: Adaptive e-learning employs algorithmic mechanisms in order to orchestrate the pace of instruction and provide tailored learning objects to support the unique educational experience of each learner. Taking this into consideration, this research work presents a fully operating and evaluated adaptive and intelligent e-learning system for second language acquisition. This system uses a hybrid model for misconception detection and identification (MDI) and an inference system for the dynamic delivery of the learning objects tailored to learners' needs. More specifically, the MDI mechanism incorporates the Fuzzy String Searching and The String Interpreting Resemblance algorithms in order to reason between possible learners' misconceptions. Furthermore, the inference system utilizes the knowledge inference relationship between the learning objects and creates a personalized learning environment for each student. The paper presents examples of operation and the system is evaluated using an evaluation model. The results are very encouraging and promising since they reveal that the hybrid model for misconception detection and identification and the inference system operate collaboratively and enhance the adaptivity of the students' needs and preferences.