With the increasing use of the Internet, the importance of sharing emotional and intellectual expressions on various platforms is emphasized and in this context, the effects of text mining and machine learning techniques on sentiment analysis are examined. User reviews on the Internet are widely used to provide feedback on products, films, and services. The focus of this study is to perform sentiment analysis using the Chicken Swarm Optimisation Algorithm (TSO) on 1000 English Amazon customer reviews obtained from the UCI Machine Learning Repository. The classification process using the Bag of Words attribute and metaheuristic algorithm is performed with linear and quadratic models. The results of the study show that the proposed method shows high success for both models and the TSO algorithm can be an effective tool in text mining studies. In future studies, it is thought that the use of more attributes may lead to better results and the TSO algorithm will be used more frequently in the field of text mining.