Classification Data Mining Techniques: A Comparative Study with respect to Heart Disease Prediction
Supervised data mining is the exploration for those algorithms which employ labeled data to train machine learning algorithms, and thus becomes capable to produce upcoming occurrences with acceptable results. The labeled data cited above, refers to those datasets which bears the predictor (independent features) as well as target features (dependent features). After the training phase of the employed techniques, classifiers or models are able to categorize or predict the unseen/unlabeled data into the respective target classes. A survey of literature depicts that a well distinguished classification/grouping of supervised techniques varies from one research article to another. So this research paper attempts to present classification of supervised techniques in well-organized manner. Of course a single research paper is not enough to define all the supervised techniques, yet it is proposed to guide the researchers or data scientists to enhance their contribution to the field of data science that yet needs to be revealed. This research work also accomplishes a comparative study for best-known classification data mining techniques among the engaged methods based on performance measures for the prediction heart diseases.
Keywords - Data Mining Techniques, Classifiers, Neural Networks, Support Vector Machine