Data Mining Applications in Healthcare: A Comparative Analysis of Classification Techniques for Diabetes Diagnosis Using the PIMA Indian Diabetes Dataset

Data mining involves extracting valuable information from a substantial quantity of data.Numerous applications for data mining can be found in different areas such as finance, healthcare & marketing. Presently, data mining hold major significance in healthcare as key field for precise disease prognosis and in-depth exploration of medical information. Researchers employ varied data minding technique help diagnose a range of illnesses, including cancer, diabetes, and heart conditions. Diabetes has emerged as major modern-day disease with causing almost the highest number of deaths around the globe. It severely affects renal & cardiac functions of the patient along with eyesight loss leading to other diseases in the body. In this paper, we made use of the Kaggle’s PIMA Indian Diabetes Data Set. We deploy four data mining strategies: logistic regression, SVM (Support Vector Machine), KNN (K-nearest neighbor), and Random Forest. An assessment of the four algorithms’ performances is made based on an array of metrics, spanning Recall, Accuracy, Precision, and F-measure. The Random Forest Classifier fares better compared to alternatives classification methods when speaking of accuracy (80.08%).