FLEX: A Content Based Movie Recommender

Abstract

Recommender systems are an efficient and powerful method for enabling users to filter through large information and product spaces. In this paper, we present a movie recommendation framework (FLEX) following a content based filtering approach. FLEX extends existing approaches like Doc2Vec and tf-idf by using a hybrid of the two methods. We use publicly available features such as movie plots, ratings, countries of production and release year to find similarity between movies and generate a recommendation list. By providing recommendations which are consistent with customer interests, the aim is to make a platform personalised and ensure customer engagement.