Development of parallel FRiS-Tax text document clustering algorithm based on MPI technology

M. E. Mansurova, V. B. Barakhnin, S. S. Aubakirov, Ye Khibatkhanuly, A. B. Mussina

Research output: Contribution to journalConference articlepeer-review

Abstract

This paper describes a parallel implementation of FRiS-Tax text document clustering algorithm. The clustering algorithm is based on an assessment of the similarity between objects in the competitive situation that leads to the concept of competitive similarity function (FRiS-function). As the scales for determination of the similarity measures are selected attributes of bibliographic description of documents. The parallelization is performed on the step of coefficient tuning in similarity measure formula of the genetic algorithm, as well as directly in step of clustering. The clustering algorithm is implemented on a highperformance MPJ Express platform. Quantitative evaluation of the execution time of the process is performed, clearly demonstrating the advantages of parallel implementation of the algorithm.

Original languageEnglish
Pages (from-to)244-256
Number of pages13
JournalCEUR Workshop Proceedings
Volume1576
Publication statusPublished - 2016
Event10th Annual International Scientific Conference on Parallel Computing Technologies, PCT 2016 - Arkhangelsk, Russian Federation
Duration: 29 Mar 201631 Mar 2016

Keywords

  • Clustering text documents
  • Genetic algorithms
  • Parallel algorithms

Fingerprint Dive into the research topics of 'Development of parallel FRiS-Tax text document clustering algorithm based on MPI technology'. Together they form a unique fingerprint.

Cite this