A Content-Based Collaborative Filtering Movie Recommendation System using Keywords Extractions

Abstract

The main focus of this research is to develop a Content-Based Collaborative Filtering model that uses different automated keyword extraction techniques to recommend movies to a user. Recommender systems predict consumers’ preferences for products and provide proactive suggestions for items they would enjoy. Collaborative filtering, content-based, and hybrid recommendation models are the most common types of recommendation models. Collaborative filtering generates suggestions based on previous interactions between the user and the item, whereas the majority of content-based recommendations are based on item comparisons. The majority of hybrid recommender systems are made up of a mix of collaborative filtering and content-based recommender models. The Content-Based method was used as the main model in this study, with Term Frequency - Inverse Document Frequency (TF-IDF) and Rapid Automatic Keyword Extraction (RAKE) algorithms serving as keyword extractors. A total of 244 movies were recommended using keywords from each extractor, with the highest average of 33% of the movies recommended from each being identical. Taking comparable movies into account, we can propose them to a user.