Abstract:
"Recommendation system is a technology that suggests personalized options to users based on their preferences and historical data and is widely used in various domains all across the world, including e-commerce, entertainment, social media etc. When it comes to movie recommendations, there are three main types used namely content based filtering, collaborative filtering and hybrid filtering. Content-based filtering is a method used in recommendation systems to recommend movies to the user based on their previous behaviour and preferences. The content-based filtering method compares the features of the movies with the movie that the user has already watched. In content-based filtering, movies are represented by their features.
After analyzing the previous studies on the problem domain as well as the prospective techniques, for this research project the author has selected the content based filtering since it is more suitable to give personalized movie recommendations. Thus the author has aimed to create and build a system that can recommend top N personalized movies using the text representation, similarity as well as classification algorithms.
The system was tested by building the machine learning model and the tested model was able to generate the expected similar movies based on the input given. The model was able to achieve the accuracy score of 60%. This current research project makes a contribution significantly to the field of Movie Recommendation Systems and presents the novel prospects for further research and future improvement."
