Ensembled Adaptive Fuzzy K-Means With Stochastic Extreme Gradient Boost Big Data Clustering on Geo-Social Networks

The szwfast growth of Geo-Social Networks (GeoSNs) offers a novel and unlikely form of data. Handlers of GeoSNs arrest their geographic locations and segment them with other users to form a community. Public detection is an efficient tool for analyzing the social relationship between users. The existing algorithm typically focuses on clustering the data but the number of expected clusters with higher accuracy is the major challenging one. In order to improve the clustering accuracy, Ensembled Adaptive Fuzzy K-Means with Stochastic Extreme Gradient Boost data clustering (EAFK-SEGBBDC) technique is introduced. The main aim of the EAFK-SEGBBDC technique is to analyze the geosocial network data and to form the cluster with higher accuracy and minimal error rate. In the EAFK-SEGBBDC technique, Stochastic Extreme Gradient Boost Cluster is an ensemble technique to construct a strong cluster by combining the number of weak learners as adaptive Fuzzy K-Means Cluster. The input geosocial network data are collected from the dataset and it is given to the adaptive Fuzzy K-Means Cluster for grouping the similar data. Adaptive Fuzzy C-Means Clustering model partitions the number of input data into different groups. Minkowski distance is calculated between the cluster centroid and the geo-social network data for grouping similar data to form the cluster. Finally, entirely the weak learners are combined to obtain the final strong clustering results with higher clustering accuracy. The observed results indicate that the proposed EAFK-SEGBBDC technique provides better performance in terms of achieving higher accuracy and lesser time than the conventional methods.