Principal Component Analysis and Correlation Analysis on Wisconsin Breast Cancer Dataset

Abstract

Breast cancer is considered to beone of the serious malignant tumorthat originates from the cells present in the breast. The disease arises typically in women, but additionally men can also be rarelyget effected. During the diagnosis of breast cancer, odd growth of cells in breast takes vicinity and this increase may be in two sorts which are benign (non-cancerous) and malignant (cancerous). For data preparation tools such as IBM SPSSModeler 14.2, Access 2003 and Excel 2003 and IBM SPSSStatistics 16 was used to calculate Principal Component Analysis to find the adequacy of the dataset attributes for the prediction of the nature of Breast Cancer Disease. Further correlation analysis is also taken up to figure the dependencies among attributes. The paper focus on the foresaid experimentation and the results are justified to generated the appropriate and the sufficiency of the attributes for the prediction.