Sentiment analysis of Malayalam tweets using bidirectional encoder representations from transformers: a study

Abstract

Sentiment analysis on views and opinions expressed in Indian regional languages has become the current focus of research. But, compared to a globally accepted language like English, research on sentiment analysis in Indian regional languages like Malayalam are very low. One of the major hindrances is the lack of publicly available Malayalam datasets. This work focuses on building a Malayalam dataset for facilitating sentiment analysis on Malayalam texts and studying the efficiency of a pre-trained deep learning model in analyzing the sentiments latent in Malayalam texts. In this work, a Malayalam dataset has been created by extracting 2,000 tweets from Twitter. The bidirectional encoder representations from transformers (BERT) is a pretrained model that has been used for various natural language processing tasks. This work employs a transformer-based BERT model for Malayalam sentiment analysis. The efficacy of BERT in analyzing the sentiments latent in Malayalam texts has been studied by comparing the performance of BERT with various machine learning models as well as deep learning models. By analyzing the results, it is found that a substantial increase in accuracy of 5% for BERT when compared with that of Bi-GRU, which is the next bestperforming model.

Utilizing BiLSTM For Fine-Grained Aspect-Based Travel Recommendations Using Travel Reviews In Low Resourced Language

“…By focusing on sentiment analysis, the research seeks to uncover and understand the emotional aspects conveyed within Malayalam text data [5]. In paper [6] Yoon presented an innovative real-time recommenda-tion system for tourism, named R2Tour, designed to adapt to dynamic scenarios, including external factors and distance information, providing personalized recommendations for tourists based on their preferences and characteristics.…”

“…Paper [7] proposes a novel approach that leverages deep learning techniques within the context of global information management to provide personalized travel recommendations for different users by utilizing deep learning models and the system can effectively analyse and interpret large-scale global information, including user preferences, location data, and cultural factors, to generate tailored recommendations. Mahadevaswamy proposed a technical overview of a Bidirectional LSTM network for Sentiment Analysis.…”

Sentiment analysis and classification of Ghanaian football tweets from the 2022 FIFA World Cup

“…Elankath and Ramamirtham [9] examined the efficiency of using the BERT model for sentiment analysis on a Malayalam dataset constructed from 2,000 tweets from 𝕏. They found that BERT outperforms other machine learning (ML) and deep learning (DL) models with an accuracy of 88.61%.…”