The Yelp Dataset comprises data collected from 8,021,122 reviews and 209,393 businesses located in 10 major metropolitan areas. This comprehensive dataset includes multiple aspects related to the businesses. We are interested in assessing the reliability of Yelp's review sentiment algorithm by constructing our own specific sentiment analysis algorithm using data mining and machine learning techniques. The system, based on Natural Language Processing (NLP), generates structured text, followed by the application of machine learning (ML) techniques to classify the text as either a 'good' or 'bad' indicator, used for sentiment prediction. The ML models we utilized here include logistic regression, random forest, k-nearest neighbors, and naive Bayes. Our results demonstrate that three of these models can precisely classify the text and accurately predict sentiment.