Hybridizing J48 and SVM for Predicting Sale and Purchase of Property

Abstract

In this paper diverse characterization procedures of Data Mining are thought about utilizing primary dataset and secondary dataset from Kaggle site. The Data Mining alludes to removing or mining learning from enormous volume of information. Classification is a critical information mining method with expansive applications. It characterizes information of different sorts. It is utilized as a part of each field of our life. Also, utilized to arrange everything in an arrangement of information into one of predefined set of classes or gatherings. This work has been done to make an execution assessment of J48 decision tree and Support Vector Machine (SVM) classification methods. J48 algorithm depends on decision tree, actually, SVM algorithm depends on statistical learning methods. The paper embarks to make near assessment of classifiers J48 and SVM with regards to real estate dataset. The experiment results are completed utilizing MATLAB 2013. From exploratory examination, it has been analyzed that the proposed method (SVM) gives more effective outcome when contrasted with previous method (J48) in the terms of Specificity and Sensitivity.