Heuristic Knowledge Discovery for Medical Data Using Cultural Algorithm
10 Pages Posted: 28 Jan 2019
Date Written: January 14, 2018
The proposed study is carried out in two phases, namely, i) classification and ii) prediction. In classification phase, C5.0 decision tree classifier is adopted for the classifying the data into its relevant class with the help of factor analysis-oriented feature selection process. With the classified result, an efficient decision-making process is done via cultural algorithm. This phase mainly concentrates on assisting each entities of the healthcare environment for predicting the thyroid diseases based on past behavior. The proposed method is evaluated on thyroid dataset acquired from UCI machine repository. Performance metrics such as precision, sensitivity and specificity are studied which proves that the proposed classifier performed better than the Support Vector Machine (SVM). In addition to, we have also discovered a valid knowledge for predicting the thyroid diseases.
Keywords: Cultural Algorithm, Data Mining, Decision Tree, Healthcare Environment, Medical Data, Prediction and Thyroid Disease
Suggested Citation: Suggested Citation