|
Title:
|
A DEMPSTER-SHAFER BELIEF THEORY BASED ALGORITHM FOR CLASSIFICATION OF IMPERFECT DATA |
|
Author(s):
|
Jinsong Zhang , Lakshitha Weerakkody , David Roelant |
|
ISBN:
|
978-972-8924-40-9 |
|
Editors:
|
Jörg Roth, Jairo Gutiérrez and Ajith P. Abraham (series editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Belief theory, Classification, Imperfect data |
|
Type:
|
Full Paper |
|
First Page:
|
11 |
|
Last Page:
|
17 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
In this paper, a Dempster-Shafer Belief Theory based algorithm is developed to tackle the problem of classification of
imperfect data. The two most common cases of imperfect data are incomplete and ambiguous data. Incomplete data refer
to missing attributes and ambiguous data refer to ambiguously labeled training examples. Belief Theory has intrinsic
capability of handling ambiguous information. By incorporating multiple imputation method to the framework of Belief
Theory, superior performance of classification can be achieved with the presence of imperfect data comparing to other
popular methods. Simulation results demonstrated our conclusion. |
|
|
|
|
|
|
