scispace - formally typeset
Search or ask a question
Posted ContentDOI

Sentiment Analysis on Indian Indigenous Languages: A Review on Multilingual Opinion Mining

TL;DR: In this paper, the authors analyze, review and discuss the approaches, algorithms, challenges faced by the researchers while carrying out the sentiment analysis on Indigenous languages on the web, and also discuss the challenges of conducting sentiment analysis in the web.
Abstract: An increase in the use of smartphones has laid to the use of the internet and social media platforms. The most commonly used social media platforms are Twitter, Facebook, WhatsApp and Instagram. People are sharing their personal experiences, reviews, feedbacks on the web. The information which is available on the web is unstructured and enormous. Hence, there is a huge scope of research on understanding the sentiment of the data available on the web. Sentiment Analysis (SA) can be carried out on the reviews, feedbacks, discussions available on the web. There has been extensive research carried out on SA in the English language, but data on the web also contains different other languages which should be analyzed. This paper aims to analyze, review and discuss the approaches, algorithms, challenges faced by the researchers while carrying out the SA on Indigenous languages.
Citations
More filters
Proceedings ArticleDOI
06 Nov 2020
TL;DR: In this article, the authors have evaluated top-performing classification models for classifying comments which are a mix of English and Malayalam, and the statistical analysis of results indicates that XLM was the top performing model with an accuracy of 67.31 %.
Abstract: Automatic classification of YouTube comments is a challenge especially when the comments are multilingual, as the messages are often rife with slang, symbols and abbreviations of the respective vernacular. In this work, we have evaluated top-performing classification models for classifying comments which are a mix of English and Malayalam. The statistical analysis of results indicates that XLM was the top-performing model with an accuracy of 67.31 %. Multi-layer Perceptron (MLP) with Term Frequency vectorizer produced the best results out of all the Deep Learning models with an accuracy of 65.98%. Random Forest with Term Frequency vectorizer was the top-performing model out of all the traditional classification models with an accuracy of 63.59%.

2 citations

Proceedings ArticleDOI
26 Dec 2020
TL;DR: In this paper, a framework for the multilingual dataset creation is proposed and the contribution of this research is the creation of multi-language dataset using proposed framework and practically presents the loopholes and consequences.
Abstract: In recent years, social media, especially Facebook have observed a massive growth of regular posts and their related comments. The users are free to post and comment any kind of information in any language, but there are no explicit mechanisms to reconcile the information expressed in different languages into the useful data set. So, in most of the cases, the contents of the Facebook expressed in different languages remain useless. This paper elucidates the motivation behind the multilingual dataset creation and proposed a framework for the multilingual dataset creation. Besides, the research illustrated the challenges associated with the data set generation, such as separating multilingual data etc. Finally, presents the consequences of multilingual dataset creation due to different challenges. Therefore, the contribution of this research is the creation of multilingual dataset using proposed framework and practically presents the loopholes and consequences.