Breast cancer is the most common cancer in the world amongwomen according toWorld health organization’s Globocan2012report 1.As per the report, Indian women are most affected bythis disease and,therefore, it is the most common cause of deathtoo.Early detection of this cancer increases the survivabilitychances of patients suffering from this disease. Many biologicaltechniques can be used for early detection of breast cancer so thatpreventive measures can be taken.In this paper,we use different data mining algorithms topredict all those cases of breast cancer that are recurrent usingWisconsin Prognostic Breast Cancer (WPBC) dataset from theUCI machine learning repository 2. Different clustering andclassification algorithms of data mining techniques have beenused to find the performance of these prediction models. Fourclustering algorithms (K means, EM, PAM and Fuzzy c-means)and four classification algorithms (SVM, C5.0, Naive bayes andKNN) are selected for this research. R programming tool is usedfor the implementation purpose that provides free softwareenvironment for data analysis 3.In short, this research is toidentify the most successful dataminingalgorithm that helps to predict those cases of cancer,which can recur. The objective here is also to find criticalattributes which play major role in determining and predicting inadvance the possibility of recurrence of breast cancer using C5.0algorithm.This paper is organized indifferent sections as follows.Section 2 highlights the already published literature in the area ofbreast cancer survivability prediction models using data mining.Section 3, explains the detail description of data, variousprediction algorithms and measures for performance evaluationon the said models.The prediction results of all clustering andclassification algorithms along with the accuracy, sensitivity andspecificity are presented in section 4.Section 5 concludes withsummary of results eventually leading to the future directions.