Accuracy Improvement of Short and Long Answer Grading Systems using Machine Learning
Simran Agrawal1, Avinash J. Agrawal2

1Simran Agrawal, Persuing Master of Technology Degree in Computer Science, and Engineering, From Shri Ramdeobaba College of Engineering and Management, Nagpur (Maharashtra) India.
2Dr. Avinash J. Agrawal, Associate Professor, in Shri Ramdeobaba College of Engineering and Management, Nagpur (Maharashtra) India.

Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 2084-2087 | Volume-8 Issue-7, May 2019 | Retrieval Number: G5590058719/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Grading of student answers with the help of language processing techniques has been a defacto standard for automatic marking systems. These systems generally do not take into consideration the errors which might have been introduced by the previous grading systems in order to incrementally improve the grading performance of the system itself. In this paper, we propose a machine learning based algorithm which uses Q-Learning and synset based language processing in order to incrementally improve the automatic grading accuracy for the both short and long answer texts. Usually systems have higher accuracy for short answer matching, but when the same system is applied to long answers then the accuracy reduces drastically. But the proposed algorithm works very well for both long and short answer grading due to it’s incremental nature, which allows the system to be used for any kind of automatic grading system. The proposed system provides atleast 10% higher grading accuracy when compared with it’s non-machine learning counterparts.
Keyword: Grading, Language, Processing, Short, Long, Accuracy.
Scope of the Article: Machine Learning.