Optimal Room and Roommate Matching System Using Nearest Neighbours Algorithm with Cosine Similarity Distribution

Abstract

The problem arises when you don’t have a desired room or friendly roommate and the goal of this project is to match preferred rooms and roommates according to their expectations. There are real world situations, especially when searching for room during transfer of job to another place, you’d probably struggle for three things room, roommate and location so that you won’t travel long distances to get to work. Also the budget, rooms will be shown with your suitable budget too. There are systems which throws rooms without taking your preferences and there are systems which finds roommates also without your preferences. So this project also takes that into account. So, this system finds you the perfect room or preferred roommates according to your needs. You can even apply filter based on amenities, gender, and sort by prices, everything will be shown according with your match score. Also rooms and mates will be sorted against highest probability of likeliness.

Keywords: NN (Nearest Neighbors), Cosine Similarity, Room and Roommate Matching, KNN (K-Nearest Neighbor), Gale-Shapleys Algorithm, Tag Based Matching, MLM-KHNN (Multi-Local Means-Based K-Harmonic Nearest Neighbor)