Breast Cancer Analysis Using WEKA

Breast Cancer happens in ladies consistently almost 1.7 million cases are expanding at the same time in India and additionally in other industrialized nations like the USA. The risks of breast cancer are extensively arranged into modifiable and non - modifiable elements. Datasets have been collected for UCI Machine Learning Repository which is of actual cancer patients. Diverse machine learning calculations were analyzed for the examination of the bosom tumor database. The following report examines a methodology where a breast cancer dataset has been preprocessed and grouped taking into account their passionate substance as no-recurrence events, recurrence events and survivability rate. There is a legitimate examination of the assessment of the distinctive requesting figurings in perspective of their exactness and survey in such cases. The three classification algorithms were used and there was the conclusion that J48 classifier gives the classification accuracy which is the highest among all. From this evaluation, it can be concluded that the J48 classification algorithm has the best results for this dataset. Simple k-means is the most ideal path for the analysis of the patients among the clustering methods.