Perceptual Computing Based Method of Evaluation in E-learning Systems

Abstract

Introduction. Implementation of e-learning has transformed the education system, expanding access to knowledge and providing flexibility in learning for both students and teachers. Application of information technologies in the education system allows to improve the learning process by introducing novel methods and approaches not only to teaching but also to knowledge assessment. However, along with the growing popularity of e-learning, new challenges arise. Among them the problem of objective evaluation of students' achievements—one of the key components of learning— is particularly significant. In an e-learning environment, evaluation becomes more challenging due to the lack of physical contact and limited access to informal communication. Therefore, it is necessary to develop and apply effective evaluation methods that would ensure objectivity, reliability, and adaptability to special features of e-learning. Evaluation using fuzzy sets enables the teacher to account for uncertainty in numerical grades and to reduce their subjective influence. Of particular importance are methods based on type-2 fuzzy sets, including perceptual computing. They enable both verbal evaluation of works and aggregation of various verbal and numerical assessments to determine the final grade. Objective. To formalize a model of quiz in e-learning systems, to develop a perceptual computing based evaluation method, which will enhance the objectivity and flexibility of the evaluation process. Results. Existing fuzzy set based evaluation methods in e-learning systems are reviewed. A graph model of a quiz is proposed, consisting of problems, knowledge units, and chunks connected by arcs loaded verbally. A method is described for calculating an aggregate grade using linguistically weighted average. Conclusions. The proposed method for quiz evaluation in e-learning systems, based on perceptual computing, allows to reduce the impact of the human factor and subjectivity. In particular, verbal evaluation of open-ended questions and processing of such grades using perceptual computing based on type-2 fuzzy sets lead to objective evaluation and have significant potential for enhancing the efficiency of e-learning.

Keywords

• perceptual computing
• e-learning
• chunk
• fuzzy sets