Certain investigation on breast Cancer data analysis using Artificial intelligence techniques

Abstract

A major field of research involved in the clinical investigation to newlinediagnose the disease. The knowledge extraction from medical dataset is really imperative to make an effective medical diagnosis. Artificial Intelligence plays an important role in the field of prediction. The main purpose of Artificial Intelligence is to train the algorithm based on previous data and find the prediction for future data. Breast cancer is one of the common cancers among the women worldwide. Survival from breast cancer disease has altogether improved and the effects on personal satisfaction have turned out to be progressively essential. The primary target of this research is centred upon the diagnosis of breast cancer in healthcare using the Artificial Intelligence (AI) algorithms. The thesis made Three significant contributions such as feature selection, classification and optimization to overcome these issues. The first phase of the research work involved in the feature selection. The dataset contains 32 features. The feature selection is very important to reduce the computational time and cost. There are so many approaches can be used such as filter, wrapper and embedded. The objective is to increase the performance of the model to diagnose the disease as benign and malignant. newlineThe features like radius, perimeter and texture are extracted from the newlineimage. The investigation is used to increase the accuracy, sensitivity, newlinespecificity and to reduce the False Positive Rate (FPR), False Negative Rate (FNR) by feature selection. There are two proposed phases such as feature grouping and feature selection. In the first phase, Pearson correlation techniques are used to find the correlation within the feature and cluster the features based on ranking. In the second phase, the collinearity within the features is reduced by adopting the Triplet Feature Selection (TFS). newline