Graph Clustering With Parallel Processing

Please enable javascript in your browser.

Bagha, Lila | 2014

357 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 47898 (02)
University: Sharif University of Technology
Department: Mathematical Sciences
Advisor(s): Daneshgar, Amir
Abstract:
One important way to discover patterns in data is clustering. Due to the limited human ability to solve problems in terms of volume and range of computing a viable solution for processing large amounts of data is using parallel processing. In this project, it is proposed some modification on Daneshgar-Javadi ‘s algorithm for data clustering using parallel processing. In this new algorithm , data will be divided among different processors to cluster them independently and in a parallel manner. Then results of each processor will be gathered and a final clustering will be performed on the gathered results. It is shown that the proposed modified algorithm along its high speed processing can improve the quality of big data clustering
Keywords:
Data Clustering ; Parallel Processing ; Big Data ; Graphs

No TOC