Mobile applications come in a variety of genres, including gaming, entertainment, and e-commerce, and users frequently rely on ratings and reviews to determine the applications’ dependability and quality. In order to match customer preferences, recommendation algorithms and sentiment analysis are essential. Personalized app suggestions improve discovery and satisfaction. Sentiment analysis, in particular, is essential for personalizing experiences since it evaluates the underlying sentiments of user comments and draws insights from user generated information. Sentiment analysis and recommendation algorithms are combined in a novel way in app recommendation systems. Sentiment analysis is used in the study to classify user ratings of six e-commerce apps from the Google Play Store, ranging from ”Very Negative” to ”Very Positive.” In an effort to better customize the user experience, these feelings were analyzed using supervised machine learning techniques. The creation of a hybrid recommendation model, which addresses Cold-Start and New User issues, is the central focus of this research. It combines collaborative filtering (CF) and content based filtering (CBF), and is further improved using a knowledge based filtering approach. The goal of this strategy is to offer more precise and customized app suggestions. Metrics including RMSE, Precision, Recall, and F1-score were used to evaluate the models’ performance, showing off their exceptional accuracy and predictive power in raising user engagement and satisfaction.