UPM Institutional Repository

Deep learning approach to tweet sentiment analysis for movie recommendation systems


Citation

Merissa, Nisa and Zolkepli, Maslina (2025) Deep learning approach to tweet sentiment analysis for movie recommendation systems. Journal of Advanced Research in Applied Sciences and Engineering Technology, 51 (1). pp. 245-257. ISSN 2462-1943

Abstract

A deep learning approach to analyze the sentiment of user tweets is proposed to provide recommendations for movies and TV shows on streaming services. Recognizing the importance of implicit user feedback on social media, we focus on how viewers express their opinions on Twitter regarding drama series and films. To achieve this, we collect Twitter data using a Python library and REST API, then employ natural language processing techniques like TextBlob and VADER to extract show names and analyze sentiment. Our deep learning model demonstrates positive results with strong accuracy and minimal error rates. The proposed approach targets social media users who also subscribe to streaming services and opens potential applications in diverse domains like purchasing preferences, political opinions, and educational choices.


Download File

[img] Text
120643.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial.

Download (3MB)

Additional Metadata

Item Type: Article
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.37934/araset.51.1.245257
Publisher: Semarak Ilmu Publishing
Keywords: Deep Learning; Levenshtein Fuzzy String Matching; NLP; Sentiment Analysis; Twitter; VADER
Depositing User: MS. HADIZAH NORDIN
Date Deposited: 17 Apr 2026 08:31
Last Modified: 17 Apr 2026 08:31
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.37934/araset.51.1.245257
URI: http://psasir.upm.edu.my/id/eprint/120643
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item