WIT Press


Global Model Of Distributed Data Mining Is Not Summing Of Local Models

Price

Free (open access)

Paper DOI

10.2495/DATA030091

Volume

29

Pages

10

Published

2003

Size

415 kb

Author(s)

M. Soliman & M. Kantardzic

Abstract

Global model of distributed data mining is not summing of local models M. Soliman & M. Kantardzic Computer Science and Computer Engineering, University of Louisville, KY, USA Abstract Discovering meaningful information from data is what is known data mining. By meaningful information we mean sequential patterns, rules and other indexes embedded in the data. So far, attention in the data mining process has always focused on extracting information from data physically located at one central site. However, most real life applications rely on data distributed in several locations. As a result, a feasible model for mining distributed data is required. The paper presents a model for mining distributed sequential patterns. The model is further verified by a simulator that can work with different mining tasks. 1 Introduction Data mining is the science of discovering rules, patterns, and various indexes embedded in large set of data [I]. This paper concentrates on the problem of discovering sequentia

Keywords