Problem statement: Data acquired for knowledge discovery from large repositories, persistently contains noisy, inconsistent and incomplete data. From these raw data, the significant attributes have to be identified and separated for the effective analysis of knowledge discovery and for the accurate prediction of the disease. Approach: In order to retrieve the significant attributes, data pre-processing must be carried out on the data base. This research work focuses on pre-processing the cardiology data in the form of data cleaning, data integration, data transformation and data reduction. Results: This data preparation process analyses the retrospective data containing a large number of insignificant attributes. The approach leads to the identification of the important attributes to predict the result of Treadmill Test. Conclusion: The significance of the results is an important step in the diagnosis and treatment of the heart disease.