Enhancing Movie Recommendations: A Content Based Approach Using TF-IDF Weighted Word2Vec and Cosine Similarity

An essential component of online video streaming platforms is their recommendation system. These systems analyze user preferences, viewing history, and demographic data to recommend customized content. These systems can be categorized into three different types on the basis of the underlying algorithms - Content based recommendation systems, Collaborative Filtering based recommendation systems and Hybrid recommendation systems. These recommendation systems often struggle to offer variety in recommendations. This causes users to be stuck in a loop where they are suggested content which is very similar to what they have already watched. This restricts their exposure to new and diverse content and leads to bad user experience. This study presents an innovative approach for improving the accuracy and pertinence of movie recommendation systems. The core methodology involves the creation of a word soup by combining different attributes of movies, vectorizing the word soup using the TF-IDF weighted Word2Vec algorithm and performing recommendations based on the cosine similarity scores of the movies. The proposed method offers nuanced and personalized recommendations enhancing the user experience. The experimental results demonstrate that the proposed methodology is efficient because the generated recommendations are contextually relevant and aligned with user preferences.