An Effective Optimized Clustering Algorithm for Social Media

The increasing utilization of social media pose diverse challenges to the human society. A considerably larger number of users participate in social media discussions, wherein different users are assigned to different groups. The user will post both positive and negative comments while participating in the discussion. This research study proposes a novel method for categorizing different types of users and determining which category they belong to. Data mining methods may benefit the social media communications. After collecting the user conversations and network data from the social media traffic, the social media user relationships are investigated. The proposed method improves the clustering results while also introducing the concept of organising the user communications and network data from social media. This tool is used to determine how many people have seen and commented on a certain post. This assists in performing user classification. Once the user information such as user messages and network relationships has been collected, data mining techniques can be used to group different types of communities. On the other hand, the existing approaches struggle to capture the proper granularity of local stakeholders and their activity. This study provides a system for discovering different communities by grouping the messages from diverse networking data streams. To cluster data, the proposed system employs the K-Means clustering technique, Optimized Cluster Distance (OCD) method and genetic algorithm.