Video Recommendation System Based on Human Interest

Abstract

In today’s world watching online videos have become a popular trend and a daily habit of our new generation. Videos are a reliable source for gaining knowledge and it is easier to grasp information through videos than reading. The internet is flooded with billions of videos hence it is a time consuming task for user to find a relevant video. So to save time as well as efforts there is a necessity to build a strong, efficient and accurate recommendation system which will display appropriate videos for the users. Video recommendation system saves users from browsing lots of videos to choose the appropriate ones, and on the other hand, it also brings the video websites more network traffic and user stickiness. The main task of the system is to provide personalized recommendations using Web Crawler, Rating Factor Neural Network, Slope one, and Slope one based Map Reduce of two types, one is Content-based filtering, and the other is Collaborative Filtering. Presentation of recommendations is an important part of the overall user experience. Video recommendation algorithm is the core of the system. The proposed paper is about the system which allows user to search for their favorite videos and the system recommends videos relevant to their choice.