PREDICTION MODEL OF ASTHMA FROM LUNG DISEASE USING DATA MINING TECHNIQUE

Abstract

Data mining is the processes of extraction useful pattern and model from huge dataset. These model and pattern have an effective role in decision making task. Data mining basically depends on quality of data. Raw data usually have missing value, noise data, incomplete data, inconsistent data and outlier data. So any raw data should preprocess before mining. The existing problems The Doctor and Nurse Give treatment with guess without developing prototype. Data preprocess is the main task of data mining. This research is done using WEKA software because it can understand large data set and have several data preprocessing technique like cleaning, integration, transformation, discretization and reduction. This research study uses different algorithm to measure the efficiency of data and show detail description of data preprocessing technique which are used for data mining. The quality of data can be measure in different ways like confusion matrix, Roc curve and cross validation of the given dataset can test and train and can generate actual and predictive value. In this research the five mostly used classification technique such as Naïve Bayes(N.B), Bayes net(B.N), J48, Random forest(RF) and OneR is used to evaluating for Asthma prediction using Asthma dataset. The experiment result shows that the classification Accuracy using N.B(83.9%), Bayes net(84.95%), J48(85.29%), Random forest(82.49%) and OneR (84.95%