Automated essay scoring using feature extraction method

Author: 
Selvani Deepthi Kavila., Lakshmi S V S S and Rajesh B

Evaluation is the major metric in any institution to judge the student and their learning skills. Essay writing is one of the major ways in which students are tested. Now, a days this evaluation is done manually. This requires a lot of human effort and time. In this paper, evaluation technique is automated. Since machine cannot understand the evaluation measures we give them in terms of keywords which act as features. Here initially the student written essay is compared with keywords mentioned by admin. If the similarity is less than 20%, then the essay is not up to the standard. Otherwise, it is evaluated using segmentation, stop words removal, calculating word frequency, forming numerical features, sentence validation, spell check. By using features extraction methods a general essay have been evaluated and after evaluating this general type of essay the final result is produced. The result obtained by proposed model is compared with manual judgment and it showed that the proposed model is much better than the manual judgment.

Download PDF: 
DOI: 
http://dx.doi.org/10.24327/ijcar.2018.12165.2132
Select Volume: 
Volume7