Abstract: Background/Objective: To analyse the efficiency and performance of pre-existing data mining techniques for development of novel predictor model for mining the response shown by physiological parameters. The medicines here being tested are physiological variability affecting medicines. Methods/Statistical Analysis: The data has been have been analysed using WEKA (version 3.7) tool for the following techniques – Classification Via Regression, Randomized filtered classifier, IBk and RandomForest technique. Findings: Usage pre-existing data mining techniques for development of novel prediction models for mining physiological variability data. An overall comparison of various techniques has also been made on the basis of various performance parameters like sensitivity, specificity, precision & F-measure. It has been found that Randomized Filtered classifier is the best suitable technique amongst all the techniques for such a use. Applications: it can be used for mining the physiological variability responses, thereby helping to check the effectiveness of medicines.

Keywords: Data Mining, Physiological Variability, ClassificationViaRegression, Randomized Filtered Classifier, Random tree classifier, IBk technique.