Social Media Platform Using K-Mean Clustering

Abstract

In today’s scenario there exists piles of information on any social media platform which raises the issue of finding and producing the most relevant piece of information which are data chunks and serves as feeds to be consumed by users that lies in the domain of interest and online behavioural activity pattern of the same users and help them find alike personalities with same interest. The current structure of social media platform is to have the most engagement by the user on their platform irrelevant of the content of information which may just serves as noise just another chunk of information for more user engagement instead of knowledge which turns out to be productive for the user. In this paper we present an acknowledgement to this information serving as chunks of noise to device a structure of a platform that will enable implementation of a social media platform that will help its users to develop and find skillset enabling them to be more productive and grow their skill tree of their knowledge. Our platform structure makes use of a machine learning technique called K-Mean clustering algorithm used in data mining over very large amount of data, an iterative approach to partition data sets into distinctive non-overlapping clusters which can be used for optimizing the content delivering engine to any specific individual or groups of users.