Significance of Accuracy Levels in Cancer Prediction using Machine Learning Techniques

Across the world, any cancer becomes a calamity for a person who is suffering from it, mainly women are facing a real challenge when it comes to breast cancer. Breast cancer can be diagnosed at an early stage to overcome the consequences at a later stage. In the field of Computer Science, Machine Learning (ML) techniques are competent enough to diagnose the stages of cancer. ML techniques work upon the data which are collected from hospitals of suspected patients. There are various ML techniques which can build a model in order to diagnose cancer on the basis of finding accuracy level. In this paper, we have discussed the significance of accuracy level for predicting the cancer. In previous works, it has been observed that 100% accuracy is found on data analysis by some researchers. Although 100% accuracy must have given perfect prediction but it is observed that prediction was not so, sometimes it gives incorrect prediction also. So, prediction technique is scaled up with inclusion of more parameters precision, recall, F1-measure, Receiver Operating Characteristics (ROC) area and Area Under Curve (AUC) score.