Hierarchical K-Means Clustering Method for Friend Recommendation System

Friend suggestion is one of the most popular social networking sites since it helps users connect with others who are similar or known to them. The notion of friend recommendation originated on social media platforms such as Twitter and Facebook, suggesting individuals use the friends-of-friends mechanism. It may be claimed that users do not form friends randomly but rather with their friends' friends. Existing techniques have a limited suggestion scope and are inefficient. Therefore, a novel buddy recommendation model that overcomes the shortcomings of the present system has been proposed. To create a more accurate buddy recommendation system, collaborative filtering is utilized to analyze users' similar and dissimilar data, and a recommendation system that provides user-to-user recommendations based on their comparable choices, activities, and preferences is created. Location-based buddy recommendation systems are gaining popularity because they connect the actual world to the digital realm and provide a complete picture of users' preferences or interests. This recommendation method will broaden the reach of recommendations from one user to another who has similar interests and is located in the same area.