Restaurant Recommender System Based On Sentiment Analysis

Sentiment analysis is used to design a recommender system many different sectors, including all the restaurant and food sectors. The majority of these recommenders, however, rely on fixed information like cuisine, price, and service quality. By determining the users’ dietary preferences and examining their choices, these systems produce customised recommendations. The usage of a scenario suggestion system is advised since it may identify people’s meal preferences based on their comments and propose restaurants that cater to those preferences. Food names are gathered and categorised using user comments to evaluate user sentiment. The constancy of the choices suggests a nearby restaurant that really is open for business. A data collection contains different user reviews. The platform’s precision, recalls, and f-measure are evaluated using three alternative scenarios top-1, top-3, and top-5. With such a 92.8% precision rate, the proposed technique is expected to provide users with exceptionally accurate recommendations. In proposed project we are including more features like the waiting time, main course, reachability, and deserts. We considered data on restaurants and over 2990 reviews on these restaurants as a dataset and we will be generating the data plot using LSTM and GRU algorithms.