Machine Learning Algorithms based on Feature Selection Method used for the Prediction of Breast Cancer
Priya Prowal, Ajay Shankar Singh, K.S Thirunavukkarasu, Shahnawaz Khan, Mohammed Redha Qader · 2021 International Conference on Data Analytics for Business and Industry (ICDABI) · 2021
Cancer is a dangerous disease, which is driven by abnormal growth of cells. Breast cancer is one among the most common type of cancer occurs in women. In 2012, around 12 percent of all new malignancy cases and 25 percent of all tumours were seen in women [6]. Breast cancer growth is regularly analysed in 140 of 184 nations around the world [17]. In breast cancer, when the abnormal cells spread in other part of body, then it is considered last stage of cancer. At this stage chances of survival rate are very low. The only solution is to detect the signs in its early stage. In past scientists created Computer aided diagnosis (CAD) frameworks that help the radiologist to recognize variations from the norm in a proficient way. Because of complex nature of micro calcification and masses, radiologist technology fails to analyse the sign of breast cancer. These days, advancement of technology is beneficial for increase in detection rate. Machine learning techniques are used for improved accuracy in early stage of cancer. In the machine learning technology, train the model and it predict the output on the basis of past experience. This research work presents an ensemble model by applying machine learning techniques which are Support Vector Machine (SVN), Artificial Neural Network (ANN), Random Forest (RF) using Feature Selection method on UCI repository breast cancer dataset. Feature Selection methods are commonly utilized to eliminate the unnecessary features before machine learning method could be applied.