Academic
Publications
Using code slicing technique to improve case classification: a suggested case slicing approach

Using code slicing technique to improve case classification: a suggested case slicing approach,Omar A. A. Shiba,Norwati Mustapha,Ali Mamat,Fatimah Ahm

Using code slicing technique to improve case classification: a suggested case slicing approach  
BibTex | RIS | RefWorks Download
Traditional approaches to classification involve generating a set of rules, based on induction from training examples such as decision tree algorithms. These approaches achieve classification by remembering individual instances, as done in case- based systems, or change connection weights, as the case in neural networks. The goal in general is to achieve correct classification for a given set of cases or situations. To achieve this goal the paper suggests and discusses a new case classification approach based on program slicing techniques, originally used in the area of software development. The proposed approach helps in identifying the subset of features used in computing the similarity measures needed by classification algorithms. When we slice a case, we are interested in automatically obtaining that portion of case "features" responsible for specific parts of the solution of the case at hand. The paper introduces the technique not as a replacement technology, but rather to complement other case classification approaches. The performance of the proposed approach is compared with other related early works using several domains. The experimental results showed that the case slicing approach achieved high classification accuracy.
Published in 2003.
Cumulative Annual