Degree of Certainty in Students’ Academic Performance Evaluation Using a New Fuzzy Inference System

Abstract

The academic performance assessment of students helps teachers, administrators, and policymakers to initiate corrective measures on academically poor students. This paper revisits the Zadeh-Deshpande formalism for evaluating students’ answer scripts using the concept of the reliability of information (degree of confidence) via the “degree of match” and fuzzy inference system in students’ performance evaluation. The case study infers that the overall performance of all the students is “average”. Furthermore, 206 of 237 students (87%) are declared as “average” with “high degree of certainty” by the evaluators (teachers). The aim of the proposed method is not to replace the traditional method of evaluation. Instead, the proposed technique is a step forward to enrich the present system of students’ performance assessment. Policymakers can use this method as it provides reliable information. A comparison between the results obtained from the academic performance of students using multiexpert and single-expert systems is also discussed in the paper.

Keywords

• academic performance assessment
• degree of certainty
• degree of match
• examiner
• fuzzy inference system
• κ coefficient
• probability density function
• zadeh-deshpande (zd) formalism