Multi-decision-tree classifier in Master Data Management System

Author

Xiaochen, Duan ; Xue, Hong

Author_Institution

Coll. of Comput. & Inf. Eng., Beijing Technol. & Bus. Univ., Beijing, China

Volume

3

fYear

2011

fDate

13-15 May 2011

Firstpage

756

Lastpage

759

Abstract

A simplified Decision Tree ID3 algorithm was advanced in this paper, and it overcame the existing bias of ID3 algorithm. And then, ADABOOST Algorithm and improved ID3 Algorithm were constituted a multi-decision-tree classifier, and it was applied in Master Data Management System to form the redundant data judgment module which responsibility is judging the redundant data. The result shows that the accuracy of this classifier is better than pure Decision-Tree classifier, and the training duration of this classifier is shorter than original-Decision-Tree-ID3 based ADABOOST classifier. It greatly reduces the manual labor after applying it in Master Data Management System, and saves the consumption of human and material resources.

Keywords

decision trees; learning (artificial intelligence); pattern classification; tree data structures; decision tree ID3 based ADABOOST classifier; master data management system; multidecision tree classifier; redundant data judgment module; Accuracy; Classification algorithms; Decision trees; Distributed databases; Manuals; Redundancy; Training; ADABOOST; Decision Tree ID3; Master Data; Master Data Management;

fLanguage

English

Publisher

ieee

Conference_Titel

Business Management and Electronic Information (BMEI), 2011 International Conference on

Conference_Location

Guangzhou

Print_ISBN

978-1-61284-108-3

Type

conf

DOI

10.1109/ICBMEI.2011.5920369

Filename

5920369