Scholarlink

This paper introduces ASTAS (Aspect-Based Sentiment Transformer Analysis & Summarization), a model for sentiment analysis and opinion summarization in the hospitality industry using transformer models and various NLP tools. ASTAS integrates Aspect-Based Sentiment Analysis (ABSA) with VADER and other NLP techniques to extract and analyze sentiments across different aspects of hotel reviews. It employs DistilBERT for accurate sentiment classification and T5 for generating concise opinion summaries. The model outperforms existing approaches and provides actionable insights for improving service quality and guest satisfaction. The study details the model architecture, evaluation results, strengths, and future directions, highlighting its potential to improve decision-making processes in the hospitality sector.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

ASTAS: Elevating Hospitality Insights Through Transformer-Based Aspect Analysis and Opinion Summarization

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kawtar Mouyassir,
  • Abderrahmane Fathi,
  • Noureddine Assad

摘要

This paper introduces ASTAS (Aspect-Based Sentiment Transformer Analysis & Summarization), a model for sentiment analysis and opinion summarization in the hospitality industry using transformer models and various NLP tools. ASTAS integrates Aspect-Based Sentiment Analysis (ABSA) with VADER and other NLP techniques to extract and analyze sentiments across different aspects of hotel reviews. It employs DistilBERT for accurate sentiment classification and T5 for generating concise opinion summaries. The model outperforms existing approaches and provides actionable insights for improving service quality and guest satisfaction. The study details the model architecture, evaluation results, strengths, and future directions, highlighting its potential to improve decision-making processes in the hospitality sector.