Optimizing Graph Mining Attributes based Inference for Social Network Analysis

Abstract

Graph mining is method of analysing and extracting useful knowledge to understand the relationships and content from structured data which is represented in graph. The research work focuses on proposing a methodology for multiple standard attributes of graph mining that can be considered to derive accurate and precise outcomes. In this research work, standard attributes were studied and further categorized based on the significance. The Graph Mining attributes are categorized into two phases-basic functionalities and implementation functionalities, on the basis of Graph Mining implementation FOSS to get specific Graph mining knowledge outcome. Researcher implement all attributes using FOSS- Gephi, SocNetV, yEd and Cytoscape tools. Researcher analysis all results obtained from attribute oriented comparative analysis which is based on supported input file formats, supported graph features. Researcher uses denotation to measure comparative analysis of SNAV tools based on attributes. Researcher derived optimization inferences from result analysis.