Restaurant Review Classification Using Naives Bayes Model

Numerous restaurants fight for the best quality for clients in the increasingly competitive restaurant sector. A restaurant is a business that demands more attention to customer care through continually enhancing customer service. The situation has an effect on the restaurant’s brand image, which is shaped by whether or not consumers are happy. Restaurant patrons may choose to benefit from others’ experiences by evaluating restaurants based on a range of factors, including meal quality, service, ambience, discounts, and deservingness. Users may leave reviews and ratings of companies and services, or just comment on other reviews. From one standpoint, bad (negative) reviews may influence how potential consumers make purchasing decisions. Sentiment analysis is a technique for determining the emotional content of a text that may be used to evaluate product/service reviews. Additionally, we may categorise them as positive or negative emotions. Understanding how the general public feels about various entities and products enables more relevant marketing, recommendation systems, and market trend research. Prepossessed data is collected, and then categorization is performed using a confusion matrix. This study enables us to create a report on the public’s perception of a particular restaurant. We developed a machine learning model and trained it using Bernoulli’s Naive Bayes classifier. Additionally, we evaluated the classifier’s performance on the test sample using evaluation matrices such as prediction, accuracy, recall, and F1 score. Customer review research has a significant influence on a business’s growth strategy.