The proposed framework first extracts representative words from each sentiment class using Mutual Information, then enriches them through dictionary-based or embedding-based lexical substitutions, and finally generates synthetic reviews using GPT-4o-mini guided by these representative terms.
Abstract
Class imbalance remains one of the most challenging problems in sentiment analysis, particularly in tourism review datasets where positive opinions substantially outnumber neutral and negative comments. This issue is especially critical because minority classes often contain the most valuable information regarding customer dissatisfaction, service failures, and opportunities for improvement. In this work, we propose a hybrid balancing methodology for sentiment analysis in Mexican Spanish tourism reviews that combines undersampling and Large Language Model (LLM)-based oversampling. The proposed framework first extracts representative words from each sentiment class using Mutual Information, then enriches them through dictionary-based or embedding-based lexical substitutions, and finally generates synthetic reviews using GPT-4o-mini guided by these representative terms. Experiments were conducted on a corpus that contains approximately 300,000 tourism reviews collected from TripAdvisor, exhibiting severe sentiment imbalance. Three undersampling strategies and multiple oversampling configurations were evaluated across six traditional machine learning classifiers and one Transformer-based model (BETO). Results show that random undersampling consistently outperformed centroid-based and K-means-based alternatives while also requiring the lowest computational cost. The best overall performance was obtained by BETO, achieving a Macro-F1 score of 0.57 compared to 0.51 on the original imbalanced dataset, representing an improvement of 11.8%. Significant gains were also observed for minority classes, with improvements exceeding 16% for the most underrepresented category. Furthermore, the proposed methodology consistently outperformed direct prompt-based generation using GPT-4o-mini, Gemini 2.5 Flash, and Llama 3.3 70B. These findings suggest that guiding synthetic review generation through representative words effectively preserves domain-specific lexical and semantic patterns of Mexican Spanish tourism reviews, resulting in more balanced datasets and improved sentiment classification performance.
Experimental results show that the proposed approach improves the Macro F1 score over the Majority Class baseline, demonstrating that isolated lexical information contributes to the learning process and indicates that lexical features alone are insufficient for accurately representing minority sentiment classes.
Juan David Jurado-Buch, L. Bustio-Martínez, Vitali Herrera-Semenets et al.· 0 citations
An extended LSQ framework is proposed that combines AI-driven sentiment analysis, linguistic interpretation, and service quality dimensions to move beyond surface-level classification toward a deeper understanding of the customer experience.
S. Idris, Daljeet Singh Sedhu Janah Singh, S. Masrom et al.· 2026 7th International Confe...· 0 citations
Social media platforms like Twitter have now emerged as treasure houses of public opinion and provide deep insight into mass mood and behavioral patterns. This paper, therefore, proposes a systematic approach toward sentiment analysis using the Sentiment140 corpus, which contains over $\mathbf{1. 6}$ million labeled tw...
Pradeepa Sampath, Lavanya Raju, Suvedha S et al.· 2026 6th International Confe...· 0 citations
Social media sites, especially Twitter, has become an essential avenue for passengers over the years. They present their thoughts regarding airline services. The analysis of unstructured textual data discusses the important insights about client happiness and sentiments. In this paper, an integrated supervised Support...
Swati Arya, Saurabh Sharma, Shivam Singh et al.· International Workshop on Ar...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.