Program for Multimodal Clustering

This paper is dedicated to distributed multimodal clustering. In this paper, we present the basic modern approaches for the data clustering and describe previous methods. Besides, we introduce a new version of multimodal clustering algorithm for distributed processing in Apache Hadoop on computer clusters. The aim of the work was to develop a program for multimodal clustering. The main algorithm of the work is introduced in this paper. Also we provide an analysis of time and space complexity of the algorithm and justify its relevance. This program allows user to conduct clustering on data with modality more or equal two. The algorithm is adapted for MapReduce distributed processing model. The program was implemented by means of Apache Hadoop framework to provide possibility of parallel computing on thousands of nodes. The paper contains 36 pages, 4 chapters, 12 illustrations, 2 tables, 26 bibliography terms, 4 appendices. Keywords: Formal Concept Analysis, multimodal clustering, multimodal data, distributed computing, MapReduce.