Study on Hybridized Gradient Descent Spectral Graph and Local-global Louvain Based Clustering of Temporal Relational Data

Abstract

The essential structure and other properties of the time series data are examined using temporal data clustering. The complexity of the data kinds makes it difficult for many approaches to perform more than just process the temporal dimension of the data. Hybridized Gradient Descent Spectral Graph and Local-Global Louvain Clustering (HGDSG-LGLC) technique is created to analyse the complex temporal data. The input dataset is used to gather the number of temporal data. The HGDSG-LGLC method then use graph-based clustering to divide the data's vertices, or points, into several clusters in accordance with the similarity matrix spectrum. The distance similarity between the data and the cluster mean is calculated. The Gradient Descent function determines the shortest distance between the data and the cluster mean. The Local-Global Louvain method then merges and filters temporal data to connect the graph's local and global edges with similar data. The change in modularity for each data is then calculated for filtering the unwanted data from its own cluster and merging it into the neighbouring cluster. As a result, optimal ‘k’ numbers of clusters are obtained with higher accuracy with minimum error rate. Experimental analysis is performed with various parameters like clustering accuracy ( C_ACC), error rate ( Err_Rate), computation time ( Time_C) and space complexity (S_com ) with respect to number of temporal data. The proposed HGDSG-LGLC technique achieves higher C_ACC and lesser Time_C, minimum Err_Rate as well as S_com than conventional methods.