Sentiment Analysis of Digital Banking Reviews Using Machine Learning and Large Language Models DOI Open Access
Raghad Alawaji, Abdulrahman Aloraini

Electronics, Год журнала: 2025, Номер 14(11), С. 2125 - 2125

Опубликована: Май 23, 2025

Sentiment analysis, in the context of digital banking reviews, aims to assess customer satisfaction and support service enhancement. Despite increasing attention sentiment analysis across domains, Arabic reviews remain underexplored. To bridge this gap, we introduce a dataset 4922 from three major Saudi banks with categories positive, negative, or conflict—providing actionable insights for banks. We evaluate using several machine learning models four large language (LLMs)—GPT 3.5, GPT 4, Llama-3-8B-Instruct, SILMA—using zero-shot (no labeled examples) few-shot (a few strategies. Our results show that 4 performs best among LLMs settings, while traditional still outperform LLMs, Voting Classifier achieving 90.24% accuracy. This study contributes domain-specific comparative research practical improvements services.

Язык: Английский

Sentiment Analysis of Digital Banking Reviews Using Machine Learning and Large Language Models DOI Open Access
Raghad Alawaji, Abdulrahman Aloraini

Electronics, Год журнала: 2025, Номер 14(11), С. 2125 - 2125

Опубликована: Май 23, 2025

Sentiment analysis, in the context of digital banking reviews, aims to assess customer satisfaction and support service enhancement. Despite increasing attention sentiment analysis across domains, Arabic reviews remain underexplored. To bridge this gap, we introduce a dataset 4922 from three major Saudi banks with categories positive, negative, or conflict—providing actionable insights for banks. We evaluate using several machine learning models four large language (LLMs)—GPT 3.5, GPT 4, Llama-3-8B-Instruct, SILMA—using zero-shot (no labeled examples) few-shot (a few strategies. Our results show that 4 performs best among LLMs settings, while traditional still outperform LLMs, Voting Classifier achieving 90.24% accuracy. This study contributes domain-specific comparative research practical improvements services.

Язык: Английский

Процитировано

0