NORMA eResearch @NCI Library

Sentiment Analysis In Tamil Language Using Hybrid Deep Learning Approach

Ramesh Babu, Suba Sri (2022) Sentiment Analysis In Tamil Language Using Hybrid Deep Learning Approach. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (1MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (3MB) | Preview

Abstract

Due to the rise of social media, the number of people using social media has also getting increased day by day from every corner of the world. This is the place for many people to share and discuss their opinion. However, these opinions were shared by various people in various languages. In recent studies, the new advancement in Machine learning and deep learning made the natural language processing task perform better in rich resources language such as English. However, these advancement has not been reached in Tamil language because of the less resources. Tamil language is one of the morphological rich language. There are a limited number of studies conducted in the field of sentiment analysis in Tamil language due to the complexity of the process and the limited resources. This research work aims to propose hybrid deep learning approaches that combines the capabilities of two different deep learning algorithms. These are the CNN-BiLSTM, CNN-LSTM, and CNN-BiGRU. In this study, to prepare the data various tools and libraries which supports Tamil language were used. The proposed methods will be evaluated and compared on various metrics such as accuracy, recall, and F1 to find the best performing model among them. The hybrid model will be able to classify the sentiments in the movie reviews in the Tamil language. The result shows that CNN-BiLSTM has achieved the higher accuracy of 80.2% and highest f1-score of 0.64 when compared to other two models.

Item Type: Thesis (Masters)
Uncontrolled Keywords: NLP; Sentiment Analysis; Deep Learning; Hybrid Deep learning; fasttext; word embedding; CNN-LSTM; CNN-BiLSTM; CNN-BiGRU
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing
P Language and Literature > PL Languages and literatures of Eastern Asia, Africa, Oceania > Dravidian Languages
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 01 Mar 2023 16:53
Last Modified: 01 Mar 2023 17:43
URI: https://norma.ncirl.ie/id/eprint/6274

Actions (login required)

View Item View Item