Abstract:
"In order to deliver more individualized eating recommendations, this study presents a novel restaurant recommendation system that makes use of large language models (LLMs) and aspect-based sentiment analysis (ABSA). Conventional methods frequently just consider overall ratings, which can hide the complex preferences of specific individuals. In order to tackle this issue, our system examines evaluations that users have left in order to identify and measure opinions regarding particular dining-related factors including food quality, service, atmosphere, cost, and cleanliness. This method improves the personalization of recommendations by allowing a more accurate and complete depiction of user preferences.
The system makes use of a dataset that has been enhanced with a variety of customer reviews. It then applies sophisticated machine learning algorithms to identify patterns and improve the precision of its recommendations. Preliminary experiments have indicated a notable enhancement in user contentment, implying that this approach has the potential to profoundly revolutionize the personalized recommendation system in the hotel sector. In addition to offering a breakthrough in recommendation technology, this project lays the groundwork for further studies by emphasizing how crucial it is to combine qualitative and quantitative data in order to better understand and satisfy customer preferences.
"
