Implementation of Clustering Techniques for Analyzing Cancer Dataset

Abstract

This research implements one of the data mining techniques known as clustering. Clustering means grouped data which have the same feature, also focuses on two clustering algorithms, EM (Expectation Maximization) and K-means. These algorithms are applied on the dataset gathered from Khartoum Rick Hospital about cancer diseases, which contains information about patients, diagnosis, treatment, etc. The results from the two clustering algorithms were discussed and compared, and it was discovered that EM algorithm produced best results than K-means by grouping the patients into different clusters according to the sex, topography of cancer, region and age.