LEARNING ASSOCIATIONS FROM BLADDER CANCER DATA

Medical data have special status based upon their applicability to all people; their urgency (including life-or-death); and a moral obligation to be used for beneficial purposes. Thus, the medical knowledge is not certain and requires verification from the medical specialists. However, most of the models are strongly associated patient s age, tissue type, hormonal therapies and disease in family with the malignancy of cancers. In data mining, association rules are discovered by identifying relationships among sets of items, or attributes in a transactional database with two measures quantifying the support and confidence of each rule. Finding frequent item sets is computationally the most expensive step in association rules discovery and therefore, it has attracted significant research attention. The significance of association rules is evaluated using three metrics: support, confidence and strength. This paper focuses on discovering association rules on real medical data sets to predict the most important factors affecting bladder cancer patient survival Moreover, this can help physicians in predicting survival time for new unseen patients.