Back close

A Machine Learning Approach for Identifying Compound Words from a Sanskrit Text

Publication Type : Conference Paper

Thematic Areas : Scientific reports

Authors : Dr. Soman K. P., B. Premjith; Chandni Chandran V.; Shriganesh Bhat; Prabaharan P.

Source : Proceedings of the 6th International Sanskrit Computational Linguistics Symposium, Association for Computational Linguistics, IIT Kharagpur, India (2019)

Url : https://www.aclweb.org/anthology/W19-7504

Campus : Coimbatore

School : School of Engineering

Center : Computational Engineering and Networking

Department : Computer Science

Year : 2019

Abstract : In this paper, we propose a classification framework for finding the compound words from a given Sanskrit text. The compound word identification plays a significant role in learning the elucidations of verses in Ayurveda text books which are written in Sanskrit. This process was modelled using several classification algorithms and we examined their efficacy with varying word embedding dimensions. Sanskrit words were vectorized using fastText word embedding method. The results show that the performance of K-Nearest Neighbor is better than other classifiers and the prediction accuracy is 90.38%.

Cite this Research Publication : B. Premjith, Chandni Chandran V., Shriganesh Bhat, Dr. Soman K. P., and Prabaharan P., “A Machine Learning Approach for Identifying Compound Words from a Sanskrit Text”, in Proceedings of the 6th International Sanskrit Computational Linguistics Symposium, IIT Kharagpur, India, 2019.

Admissions Apply Now