Pragmatic Assessment of Occurrence of Brain Cancer with Incidence Levels using Collaborative Big Data Mining Techniques

Abstract

: The major objective of this research is to identify the presence of brain cancer along with the incidence levels of beginning stage to advanced stage using collaborative analysis of big data and data mining techniques. The dataset collected from secondary sources had few errors and rectified using preprocessing techniques in MATLAB. Further, the testing dataset is processed with k-means algorithm to form cluster analysis and identify the presence of brain cancer in three levels of well, fair and poor levels using degree of difference between the normal and cancer cells in brain. The algorithm is modified according to the needs of the medical analysis of the current dataset. The results indicates the presence of brain cancer in various three levels under cluster values of initial stage (54%), Curable stage (38%) and incurable stage (8%), respectively. The accuracy of prediction is 93.4% and the error identification is 9.3% whereas the sensitivity and specificity accounts to 0.8 and 0.7, respectively. Hence, further analysis is conducted in tableau big data tool and the sheets with story boards are formed. This research indicates the occurrence of brain cancer is influenced by gender and age factors along with regular activities and streams. Thus, brain cancer is considered as one of the challenging prediction as the cell contains mixed patterns with variations according to gender and age of human beings.