Multifunctional And Function-Oriented Parliamentary Streamlining Structure For Community Identification On Social Media

Abstract

This paper presents a multi-purpose function-based parliamentary optimization (MPPOA) community detection methods. Initially, the population of parliamentary optimization algorithm (POA) was created in a python environment for the data used. The population formed was divided into a certain number of groups, and the power values of each group were calculated. While strong groups show joining according to the determined combination probability value, the vulnerable groups are eliminated from the population according to the determined deletion probability. The result for the problem has been approached. The program steps continued until all groups were combined and the termination condition of the algorithm was met; the individual with the highest eligibility values among the remaining data in the last stage was accepted as the solution to the overlapping community discovery problem of the proposed algorithm. Subsequently, performs a proposed work evaluated over one artificial and four real network-based social media data sets. The comparison of feature evaluation is carried out to identify the influence of single and multi-purpose functions on community detection performance. Finally, this paper gives a comparative analysis of proposed MPPOA algorithm worth three heuristic overlap community detection algorithm over six real social media data set.