Abstract:

Medical professionals have labeled cervical cancer as a possibly fatal illness. Late detection and therapy pose significant challenges and put patients' lives in jeopardy. Formal screening for disease detection is hindered by factors such as high medical costs, a lack of accessible healthcare facilities, cultural attitudes, and a delay in the onset of symptoms in both established and developing countries. Several illnesses, including cervical cancer, can be diagnosed early with low processing costs using machine intelligence. Machine-intelligent solutions make early detection of cervical cancer much easier, and patients don't have to go through modern, time-consuming medical treatments. The present machine categorization techniques for illness detection suffer from over-reliance on the forecast precision of a particular predictor. Due to prejudice, over-fitting, improperly managing noise data, and anomalies, the implementation of singular categorization techniques does not guarantee the optimal forecast. In order to arrive at a correct diagnostic that takes into account the patient's medical circumstances or symptoms, the authors of this study suggest using a majority-voting Ensemble categorization technique. Different classifications, such as Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR), are tested in this investigation. A 94% improvement in forecast accuracy was recorded in the research, which is fundamentally higher than the forecast exactness’s of individual arrangement strategies assessed on the equivalent bench-marked datasets. Hence, the proposed model furnishes clinical experts with a second view to aid in the early detection and management of illness.

Description:

Machine Learning Algorithms, Classifiers, Cervical Cancer, Ensemble Classification