Improving Rules Quality Generated by Rough Set Theory for the Diagnosis of Students with LDs through Mixed Samples Clustering

Authors:
Tung-Kuang Wu;Shian-Chang Huang;Ying-Ru Meng;Yu-Chi Lin
Affiliations:
Dept. of Information Management, National Changhua University of Education, Changhua, Taiwan;Dept. of Business Administration, National Changhua University of Education, Changhua, Taiwan;Dept. of Special Education, National HsinChu University of Education, HsinChu, Taiwan;Dept. of Information Management, National Changhua University of Education, Changhua, Taiwan
Venue:
RSKT '09 Proceedings of the 4th International Conference on Rough Sets and Knowledge Technology
Year:
2009

Citing 7
Cited 0

C4.5: programs for machine learning

C4.5: programs for machine learning
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Discretization Problem for Rough Sets Methods

RSCTC '98 Proceedings of the First International Conference on Rough Sets and Current Trends in Computing
RSES and RSESlib - A Collection of Tools for Rough Set Computations

RSCTC '00 Revised Papers from the Second International Conference on Rough Sets and Current Trends in Computing
Evaluation of ANN and SVM classifiers as predictors to the diagnosis of students with learning disabilities

Expert Systems with Applications: An International Journal
Uncertain data mining: an example in clustering location data

PAKDD'06 Proceedings of the 10th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining
Feature selection with rough sets for web page classification

Transactions on Rough Sets II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Due to the implicit characteristics of learning disabilities (LDs), the identification or diagnosis of students with LDs has long been a difficult issue. In this study, we apply rough set theory (RST), which may produce meaningful explanations or rules, to the LD identification application. We also propose to mix samples collected from sources with different LD diagnosis procedure and criteria. By pre-processing these mixed samples with some simple and readily available clustering algorithms, we are able to improve the quality of rules generated by RST. Our experiments also indicate that RST performs better in term of prediction certainty than other rule-based algorithms such as decision tree and ripper algorithms. Overall, we believe that RST may have the potential in playing an essential role in the field of LD diagnosis.