Prediction of Effective Heart Disease in Health care domain using Data Mining Techniques

  • Pathan Ahmed Khan, Dr. Yogesh Kumar Sharma


Data mining, using combined strategy of statistical analysis, machine learning and database technology, has been used in big databases to extract hidden patterns. Furthermore, because of its usefulness in developing different applications in the prosperous field of healthcare, medical data mining is an exceedingly essential research subject. The heart disease appears to be the primary cause, while summing the deaths worldwide. Identifying a person's potential for heart disease is a hard undertaking for doctors, as it involves years of experience and intensive medical testing. The health businesses acquire enormous amounts of data containing certain information that is useful for good decision-making. Certain advanced data mining techniques are utilised to provide adequate results and make good decisions on data. In this research, three classification data mining techniques, K-NN, Decision Tree, and Nave Bayes, are discussed and Used to develop a heart disease prediction system for analysis and prediction. The principal objective of this substantial study is to establish the optimal technique of classification for maximum precise categorization of normal and abnormal people. It is therefore feasible to avert loss of life earlier. The experimental setting was designed to evaluate the performance of algorithms through the UCI machine learning repository's dataset on heart disease. It is observed that the 98% accuracy of the Naïve Bayes algorithm is best compared to other heart disease prevention algorithms.