A Review of Big Graph Mining Research

1st Annual Applied Science and Engineering Conference

工业技术;自然科学

Atastina, I.^1 ; Sitohang, B.^1 ; Saptawati, G.A.P.^1 ; Moertini, V.S.^2

School of Electrical Engineering and Informatics, Bandung Institute of Technology, Indonesia^1

Informatics Department, Parahyangan Chatolic University, Indonesia^2

关键词: Compression techniques;    Distributed systems;    Frequent sub graphs;    Graph mining;    IMPROVE-A;    Preprocess;    Process algorithms;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/180/1/012065/pdf
DOI  :  10.1088/1757-899X/180/1/012065

来源: IOP

【 摘 要 】

Big Graph Mining" is a continuously developing research that was started in 2009 until now. After 7 years, there are many researches that put this topic as the main concern. However, there is no mapping or summary concerning the important issues and solutions to explain this topic. This paper contains a summary of researches that have been conducted since 2009. The result is grouped based on the algorithms, built system and also preprocess techniques that have been developed. Based on survey, there are 11 algorithms and 6 distributed systems to analyse the Big Graph have been improved. While improved pre-process algorithm only covers: sampling and compression technique. These improving algorithms are usually aimed to frequent sub graphs discovery, whereas slightly those of is aimed to cluster Big Graph, and there is no algorithm to classify Big Graph. As a conclusion of this survey, there is a need for more researches to be conducted to improve a comprehensive Graph Mining System, especially for very big Graph.