Automating the Assessment of student’s Descriptive Answer Using Natural Language Processing

Authors

  • Parkavi A, Sowmya B J, K G Srinivasa

Abstract

In educational institutions a major concern is evaluation of answer sheets of subjective type questions. Descriptive Answer Assessment Using NLP aims at making this time consuming and tedious task of correcting students answer scripts very easy by utilizing the techniques of Natural Language Processing. The objective of Descriptive Answer Assessment Using NLP is to automate the process of evaluating descriptive answer scripts. Various algorithms are used to find the similarity between the reference answers and the student answer scripts. If the similarity is high, more marks will be awarded. Using several text similarity algorithms, the similarity and relative accuracy will be determined. Marks are assigned to students based on the degree of similarity with the student answer and reference answer. The work aims in exploring various Text Similarity algorithms in identifying their relative accuracies. There are several advantages of using this automated system like it helps in reducing the time taken by faculty to correct  the papers, the tests or examinations can be conducted online, the answers can be evaluated immediately and  it would be beneficial for universities, schools and colleges for academic purpose by providing ease to faculties and the examination evaluation cell.

Published

2020-11-01

Issue

Section

Articles