Voice Assistant based Movie Recommendation System utilizing User Filtered Correlation Factor

Abstract: The project deals with the idea of a Movie Recommendation System, which itself is one of the key tools being deployed across various OTT and Social Media platforms that serve the purpose of engaging a customer of a particular service to stick to their platform service by consuming more relevant content. Primarily the crucial factor to boost the momentum of the Recommendation Systems date back to the emergence of Search Engines (as a part of everyday lives). These search engines make use of multiple algorithms to page and derive the useful results by indexing every page of a hosted/ live site. Due to these technological advancements, there has been a burgeoning of data, which is being generated at every mill-second from numerous streams and multitudes across various fields. While the utility of platforms - like the afore mentioned - give rise to Historical Data, where the behavioural traits of multiple users give scope to Clustering upon a set of distinct characteristics. One such method to suggest (or) recommend movies to online users is to utilize the already available historical data. Within its entirety the Recommendation Systems have become one of the most prominent tools within the Machine Learning framework, serving various industries and sectors. Our project uses the already available Historical Data downloaded from Kaggle and develops the Voice Assisted Approach to feed input to our Movie Recommendation System. Keywords: Movie Recommendation System, User-based Collaborative Filtering, Voice Assistant, Pearson’s Correlation, Historical Data, Speech Recognition, Tkinter, Data Visualization