Experiments Towards Determining Best Training Sample Size for Automated Evaluation of Descriptive Answers Through Sequential Minimal Optimization

With number of students growing each year there is a strong need to automate systems capable of evaluating descriptive answers. Unfortunately, there aren’t many systems capable of performing this task. In this paper, we use a machine learning tool called LightSIDE to accomplish auto evaluation and scoring of descriptive answers. Our experiments are designed to cater to our primary goal of identifying the optimum training sample size so as to get optimum auto scoring. Besides the technical overview and the experiments design, the paper also covers challenges, benefits of the system. We also discussed interdisciplinary areas for future research on this topic.

Sunil Kumar C1, R. J. Rama Sree2
Bharathiar University, India1, Rashtriya Sanskrit Vidyapeetha, India2

Descriptive Answers, Auto Evaluation, LightSIDE, Machine Learning, SVM, Sequential Minimal Optimization
Published By :
Published In :
ICTACT Journal on Soft Computing
( Volume: 4 , Issue: 2 )
Date of Publication :
January 2014

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.