Topics In Demand
Notification
New

No notification found.

Investigating optimal Machine Learning Techniques for the Detection of 4 Devanagari Languages in Roman script

October 10, 2022 2382 4 Analytics Data Science & AI Community AI Inside

Investigating optimal Machine Learning Techniques for the Detection of 4 Devanagari Languages in Roman script

Due to diversity in languages in India and lack of support for Indic languages in digital and physical keyboards, a common phenomenon, especially in online modes of communication, is the utilization of the roman script for Indic languages. This form of transliteration is quite common. As such, identification of the root language which is being transliterated can have many potential uses in translation, messaging, and search systems. It is therefore necessary to develop a rapid, accurate, and light model for the purpose of this detection. This paper presents an exploration of various standard textual classification techniques to achieve such a model. The paper is focused on 4 Devanagari languages: Hindi, Gujarati, Marathi and Sindhi. The machine learning models tested were a Multinomial Naive Bayes algorithm, along with a Recurrent Neural Network and a Convolutional Neural Network. The highest accuracy achieved was 97.3%.

2382

4

Download

Listen to this article



Aditya Mehta , Reverie Language Technologies. Reverie mentors Ashis Samal and Bhupen Chauhan. 


That the contents of third-party research report/s published here on the website, and the interpretation of all information in the report/s such as data, maps, numbers etc. displayed in the content and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party research report/s published, are provided solely as convenience; and the presence of these research report/s should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these research report/s, you do so at your own risk.




LATEST REPORTS

© Copyright nasscom. All Rights Reserved.