NORMA eResearch @NCI Library

Sentiment Analysis of Hindi Song Lyrics using a BiLSTM Model with BERT Embeddings

Kulkarni, Jay Milind (2023) Sentiment Analysis of Hindi Song Lyrics using a BiLSTM Model with BERT Embeddings. 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 (2MB) | Preview

Abstract

Songs, Poems, and Music has played an important role in expressing human emotions over centuries. Over a period of time, with the development of humans and with the introduction of movies, and albums there has been an increase in the popularity of songs. With the boon of digitalization, it has now become easy to access songs all over the world. There has been an increase in research that has been carried out to identify the sentiments of the songs. However, this has been carried out mostly for languages that have a sufficient amount of digitally available resources such as English, German, Chinese or Spanish, and a few others. However, there are still other languages such as Hindi, Marathi, Latin American languages, and many more where very little research has been carried out. This research is carried out for Hindi song lyrics data which have sentiment labels as Party, Sad, and Romantic. The model that was implemented was the BiLSTM model for classification with input as text data which was converted as BERT embeddings using the BERT model. However, there was a class imbalance in the dataset with “Romantic” being the majority class, and “Party” and “Sad” as the minority classes, and an attempt to resolve this was done by introducing class weights and K-Fold Cross-validation. Three models were implemented to classify the emotions of songs and out of which the best model obtained an accuracy of 63%.

Item Type: Thesis (Masters)
Supervisors:
Name
Email
Shahid, Abdul
UNSPECIFIED
Subjects: M Music and Books on Music > M Music
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
B Philosophy. Psychology. Religion > Psychology > Emotions
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 29 Nov 2024 12:33
Last Modified: 29 Nov 2024 12:33
URI: https://norma.ncirl.ie/id/eprint/7208

Actions (login required)

View Item View Item