Movie Recommendation System Using Machine Learning

Background: We are living in the `Era of Evolution`. For any given product there are thousands of options, like online shopping, streaming videos, watching movies or series and so on. The recommendation system works for personalize a platform and recommended the user as their personal interests. In this article, our actual aim is to reduce the human effort by suggesting movies based on the user’s interests.

Objectives: The term `Technological Evolution` captures many advanced platforms such as Machine Learning, Deep Learning, Data Mining, the Internet of Things (IoT) etc. There are many real-life applications such as PowerShell [1], IoT [2-3], Cloud Computing [4] and so on. The objective of recommendation systems is to provide recommendation based on recorded data. These system is works by information filtering techniques to process information and provide the user relevant items. In this article, our actual aim is to reduce the human effort by suggesting movies based on the viewers interests.

Methodology: KNN Algorithm- KNN algorithm is called the K nearest neighbor algorithm. The primay thought of this algorithm is if most of the k most comparable neighbors of the test in the component space have a place with a specific class, at that point the example is considered to have a place with this category.

Results and discussion: When the user presses the “Recommendation” button it will recommend movies based on his previous ratings. If he is a new user and has not rated any movies then he is expected to search for a random movie or any movie of his interest in the search box. Here the user is new and has not rated any movies he searches for the word ‘Super’ in the search box and all the movies with words ‘Super’ in them will appear on the screen.

Conclusions and future work: These systems work on individual users’ ratings, hence limiting your choice to explore more. Deep learning, extreme learning machines this machine learning based algorithm maybe applied for better recommendations in future.