Machine Learning Techniques with Ontology for Subjective Answer Evaluation

TL;DR

This paper explores how integrating domain ontology with machine learning techniques improves the accuracy of computerized evaluation of technical English essays, demonstrating enhanced performance in answer scoring.

Contribution

It introduces the use of domain-specific ontology to enhance machine learning-based essay evaluation methods, showing improved accuracy over non-ontology approaches.

Findings

Ontology integration improves evaluation accuracy

Enhanced semantic understanding in answer scoring

Better coverage of concepts and synonyms

Abstract

Computerized Evaluation of English Essays is performed using Machine learning techniques like Latent Semantic Analysis (LSA), Generalized LSA, Bilingual Evaluation Understudy and Maximum Entropy. Ontology, a concept map of domain knowledge, can enhance the performance of these techniques. Use of Ontology makes the evaluation process holistic as presence of keywords, synonyms, the right word combination and coverage of concepts can be checked. In this paper, the above mentioned techniques are implemented both with and without Ontology and tested on common input data consisting of technical answers of Computer Science. Domain Ontology of Computer Graphics is designed and developed. The software used for implementation includes Java Programming Language and tools such as MATLAB, Prot\'eg\'e, etc. Ten questions from Computer Graphics with sixty answers for each question are used for testing. The results are analyzed and it is concluded that the results are more accurate with use of Ontology.