Authors:
Romasa Qasim
and
Rashedur M Rahman
Affiliation:
North South University, Bangladesh
Keyword(s):
Decision Tree, Data Mining, Hepatitis, LDA.
Related
Ontology
Subjects/Areas/Topics:
Applications of Expert Systems
;
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Data Mining
;
Databases and Information Systems Integration
;
Enterprise Information Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
Abstract:
In this paper we try to solve the challenge presented by the Chiba University and Hospital, Japan. Learning from the available liver biopsy data, the type of hepatitis of a test patient is found out without performing patient’s liver biopsy. The degree of liver fibrosis is also determined without performing biopsy. It is observed that for hepatitis type classification, linear discriminant classification performed well, and for finding the degree of liver fibroses decision tree’s results are encouraging. Later, the obtained decision tree is used to find out whether the interferon therapy, taken by set of patients, is effective or not. Result shows that linear discriminant analysis best suits to classify the type of hepatitis. However, to find the stage of fibrosis, decision tree performs well. The research finding reveals the fact that interferon therapy either reduces the liver fibroses level or does not let it increase from the diagnosed level.