Prediction of recurrence cancer using J48 algorithm

Data mining is the process of extracting the useful information from the large existing datasets. Data Mining allows users to analyze data from various dimension, categorize the data, and present a review on the identified relationship between the data. This paper describes the use of j48 data mining algorithm to predict the recurrence of cancer based on the dataset of breast cancer. If cancer comes back after treatment, whether in the same place it first started, or in any other part of the body is called recurrent cancer. In this paper we use J48 data mining algorithm on the data set of breast cancer. J48 is a java implementation for C4.5 algorithm provided by Weka tool. Here we analyze the decision tree generated by the algorithm using 10-fold cross validation method to predict the recurrent events, based on the attributes such as node-caps, degree of malignancy, age, tumor-size, menopause, irraradiate etc. This process of data mining will help to identify what range or value of various attribute causes the recurrence cancer.