From traditional to large language models: a novel nlp-based model for sentiment analysis in social media

Entity: Universitat Rovira i Virgili (URV)

Confidenciality: No

Education area(s): Enginyeria de la Seguretat Informàtica i Intel·ligència Artificial

APS: No

Title in different languages: From traditional to large language models: a novel nlp-based model for sentiment analysis in social media

Abstract: This thesis addresses the need for Catalan sentiment analysis tools by creating a new 23,000-sample balanced corpus and developing a specialized classification model, the CSXSC. This fine-tuned, 125M-parameter, encoder-only RoBERTa model was benchmarked against two 7-billion-parameter decoder-only models trained efficiently using QLoRA. The results demonstrate the superiority of the specialized model, which achieved a final test set accuracy of 83.69\% while being over 24 times more computationally efficient. This study concludes that for this discriminative task, a smaller, architecturally appropriate model provides a more accurate and practical solution than larger, general-purpose LLMs.

Subject: Xarxes socials

Academic year: 2024-2025

Language: en

Work's public defense date: 2025-06-12

Subject areas: Computer engineering

Student: Arias Cámara, Daniel

Department: Enginyeria Informàtica i Matemàtiques

Creation date in repository: 2026-03-13

TFM credits: 9

Keywords: Large Language Models, Sentiment Analysis, Social Media

Title in original language: From traditional to large language models: a novel nlp-based model for sentiment analysis in social media

Access Rights: info:eu-repo/semantics/openAccess

Project director: Pascual Fontanilles, Jordi