Enhanced Movie Recommendation and Sentiment Analysis Model Achieved by Similarity Method through Cosine and Jaccard Similarity algorithms

Nowadays, with technology at the fore front of almost every business, there is an abundance of data and information that is difficult to sort through and understand. Consequently, a recommendation system is beneficial in dealing with such a big amount of information and filtering out the valuable material that is rapid also, pertinent to the client’s choice. In order to suggest films that are similar to the one the user has chosen, this paper describes a method for a movie recommendation system that uses different similarity approaches. In order to improve the user experience, this system performs sentiment analyses the reviews of the movie that was selected using optimization techniques. To increase accuracy and efficiency, a novel recommends-based sentiment classification technique that combines the Support Vector Machine Classifier with the Harris Hawks optimization method is utilised. The proposed approach is capable of handling high-dimensional data, according to empirical findings on the dataset (Movie). In terms of classification accuracy, we found that the suggested method performs better than the other recommendation systems strategies examined in this study. The trial findings revealed that the suggested strategy had a success rate of 97% accuracy.