Paper Title
A Comparative Studyon Various Movie Recommendation Systems

Abstract
Online, there is a large amount of data and content that is growing dramatically every day. Users therefore require a product that can make movie recommendations with greater accuracy and efficiency. Each user has a different preference for the types of content. Every online service provider aims to attract as many customers as they can. The recommender systems are used in this situation. The main goal of this project is to create a quick and better method for movie recommendations that includes review analysis. In order to achieve effective results, we have suggested creating a model utilising a content-based filtering technique (supervised learning) in conjunction with a cosine similarity measure and a levenshtein distance. The issues with scalability, data sparsity, and automation are the main difficulties users encounter. By the end, we hope to solve these issues and create a logical and useful model that makes use of ajax requests, APIs, and a minimal number of other resources. Keywords - Content-Based Filtering, Supervised Learning, Cosine Similarity Measure, Levenshtein Distance, Naïve Bayes classification, TF-IDF Vectorizer