Ranking Movie Recommendations Using Content-Based Filtering and Word Embeddings

Finding similar items to your interest is beneficial when the number of choices is too big. Movies are one of the most effective entertainment methods that people like the most. The number of movies is increasing each year, and the choice of which movie to watch is becoming hard. Recommending movies is a solution to such a problem. The traditional approaches of recommending items such as content-based filtering and Itembased filtering are not enough these days. Natural Language Processing (NLP) it is advancements in recent years in developing word embedding methods encouraged us to utilize such techniques and combine them in ranking and improving movie recommended systems. This paper proposed the idea of utilizing word embedding techniques in ranking recommendations for several movies. The finding of the proposed system showing word embedding can improve the list of recommendations and most similar movies to what a user likes to watch.