Automatic Scikit-learn based detection and classification of Breast Cancer using Machine./ Learning techniques

Abstract

The most severe disease nowadays is breast cancer. The abnormal development of breast cells is what causes breast cancer. The breast ducts may include this type of malignancy. Women over 40 are the ones who get diagnosed with breast cancer most frequently. Women over 50 have the highest incidence (about 77% of invasive cases). A clear understanding of breast cancer and its kinds, as well as the creation of prophylactic measures, have made significant strides. Many genes have been linked to breast cancer. Malignant and benign cells both fall within this category. Cancerous malignant conditions exist. Non-cancerous benign cancer. Malignant cells affect other cells of the body as well as the tissues continue to divide uncontrollably. While benign tumors are significantly less dangerous than malignant ones since they do not spread to other parts of the body. Treatment is not necessary for these cancers. Malignant tumors can be treated with chemotherapy, radiation therapy, and immunotherapy, among other forms of treatment. The algorithms used to develop this prediction system are machine learning-based. Machine learning became a research focus and was found to be an effective method. Recently, many scholars have become interested in the developing technique of machine learning. Scikit-learn is used to create a model that categorizes breast cancer as either malignant or benign. A type of library that is included with the collab platform is called Scikit-learn. Mammograms have historically been the primary method of finding breast cancer. Any calcium deposits or calcifications in the breasts might be seen on a mammogram. Considering how pricey it is, few people can afford it. It will be great if an early diagnosis occurs so that treatment can get underway right away. Many women's lives could be saved by this. This dataset analysis will benefit from the use of machine learning algorithms. Six machine learning (ML) methods are compared in this paper: Naive Bayes (NB), Random Forest (RT), K Nearest Neighbor (KNN), Support Vector Machine (SVM), and Decision Tree (DT). These methods will be applied to accurately forecast outcomes. Due to early detection, the 5-year survival rate for breast cancer patients is above 82 %. The eventual goal of this project is to assist clinicians in analyzing enormous databases of available cancer data and discovering patterns using patient data and that cancer data. This endeavor to a modest stride in the protracted battle against breast cancer.