Abstract: The early and precise detection of breast cancer is one of the most crucial measures in the fight against it. Unfortunately, breast cancer is asymptomatic in the early stages, but certain symptoms may appear later on. However, when breast cancer is symptomatic, therapy may be difficult or even impossible, which can result in death. Future technique, info gain method, and random forest method are the three approaches employed. Thus, accurate risk assessment is crucial for lowering mortality. Due to the different risk profiles of women, such as delayed menarche, low drug misuse, and low smoking rates, certain computational algorithms for assessing breast cancer risk have been established in the developed world. However, these strategies do not function well in developing countries. We attempted to demonstrate the superiority of the random forest approach. In this study, we use the Random Forest Classifier (RFC) machine learning approach drinking, dangers at work, and menopausal age. Four strategies — utilizing Chi-Square, common data gain, Spearman relationship, and all elements — were exactly utilized in the component choice. When all risk factors were taken into account. The findings of the selected characteristics for mutual information gain and Chi-Square were identical. The Random Forest Classifier has a fair chance of accurately predicting a woman's risk of developing breast cancer. The study assisted in identifying the female breast cancer risk factors. This is important information that can assist women in focusing on those risk factors in an effort to lower the incidence of breast cancer.

random forest, breast, cancer, classifier