A seminar was held by Asst. Prof. Dr. Ashwan Anwar Abdul-Mun’im, a lecturer in the Computer Department/College of Computer Science and Information Technology. The seminar submitted a proposal to apply machine learning techniques to the problem of breast cancer classification. This classification depends on the use of the unsupervised linear transformation technique together with Cos Similarity Machine Learning. In the unsupervised transformation, the Principal Components Analysis (PCA) technique was used. PCA is used in analyzing a breast cancer dataset to identify effective and unique features. In addition, the use of the cosine similarity classification model has been proposed to test the accuracy of the given features. The performance acceptability of the proposed PCA-Cos technology has been demonstrated by conducting large-scale practical experiments on a binary class breast cancer dataset. Experimental results prove that PCA-Cos is a promising method for dealing with the challenges of feature selection and cancer classification with great accuracy