Indian Journal of Science and Technology
Year: 2017, Volume: 10, Issue: 22, Pages: 1-6
Narander Kumar1 and Sabita Khatri2
1Department of Computer Science, Babasaheb Bhimrao Ambedkar University (A Central University), Vidya Vihar, Raebareli Road, Lucknow – 226025, Uttar Pradesh, India; [email protected] 2Department of Computer Engineering, University Institute of Engineering and Technology (UIET), Babasaheb Bhimrao Ambedkar University (A Central University), Vidya Vihar, Raebareli Road, Lucknow – 226025, Uttar Pradesh, India; [email protected]
Objective: To intend towards increasing classification efficiency of J48 classifier by introducing attribute set on the basis of applied genetic programming. The constructed set of attribute not only enhances the data classification capabilities of J48 but also increased the data space for the algorithm towards giving more accurate results. Methods/Analysis: The datasets related to heart and liver disease were selected from the UCI machine learning repositories. The experiment has been conducted with the help of WEKA tool, which is an open source tool for data mining. Finding: After experimentation it is found that the efficiency of J48 is giving better classification accuracy with reduced error rate when applied with datasets after inclusion of newly generated attributes by genetic programming. After adding attributes induced by genetic programming, significant efficiency boost can be seen in classification capabilities of J48 by 74% to 83% and 68% to 72% for heart and liver disease datasets respectively. Improvement: We obtained better results when compared to the existing literature for the chosen clinical datasets.
Keywords: Classification, Clinical Data, Decision Tree, Data Mining, Genetic Attribute
Subscribe now for latest articles and news.