A review on using various DM techniques for evaluation of performance and analysis of heart disease prediction

In the recent years, the most difficult task in medical sciences is to foresee a heart disease. This gives birth to the necessity for the decision support system developed to predict the heart disease of a patient. In case of disease prevention and disease treatment the decision-making marks a remarking step. The increasing demand of efficient healthcare organizations are encroached with the challenges to provide cost-efficient and superior quality patient care. To cope up with the huge databases of health care information systems to discover the relevant knowledge of decision making, Knowledge Data Discovery, is carried out through data mining. The procedure involved in the efficient use of the healthcare data through extraction and pattern discovery is referred to as data mining. It also acts as a study tool to learn the veiled relationships in heart disease healthcare data. These methods reveal various consistent primary recognition systems from clinical and diagnosed data. These methods require less time and produce more accurate results. This paper contains comparisons and results of different evaluation methods. Focusing greatly on developing a prototype that reveals the veiled knowledge of heart disease from past records with the help of a hybrid genetic algorithm. Thus, helping the medical practitioners to take up the right decision on right time providing appropriate treatments.