Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.supervisorBaliyan, Niyati-
dc.contributor.supervisorBassi, Vineeta-
dc.descriptionMaster of Engineering -CSEen_US
dc.description.abstractNowadays, heart diseases are very common and one of the major causes of death across the globe. This calls for accurate and timely diagnosis of the heart disease. Although, the healthcare industry has come a long way to treat patients with several kinds of diseases, yet the prediction of heart diseases is a complicated task in the healthcare field. Therefore, it is essential to develop a decision support system for analysing the heart disease in a patient. A heart disease is a dangerous disease which is not visible to naked eyes. Bad decision making by the physician can even cause death of a patient. To avoid these kinds of decisions, numerous hospitals use the clinical information system to manage the data of patients’ health. There is abundant data available with the health care systems; however, the knowledge about the data is rather poor. The accessibility of the enormous size of medical dataset hints towards the requirement of a tool which analyses data to extract valuable information. Unfortunately, this data is hardly used to support the healthcare decision making. There are huge amounts of hidden patterns in this data which are yet to be explored; this gives rise to the question that how we can extract useful information from these patterns. Thus, it is essential to form a model with the help of standard datasets to predict the heart disease of the patients even before it occurs. Data scientists have attempted several analytical methods in order to improvise the examination of heart diseases. Previously, various data mining techniques have been implemented in the healthcare systems, however, hybridization in addition to single technique in the identification of heart disease shows promising outcomes, and can be useful in further investigating the treatment of the heart diseases. Additionally, this can reduce the cost of treatment. This work attempts to survey some recent techniques applied towards knowledge discovery for heart disease and further proposes a novel prediction method using bagging and boosting to attain improved accuracy. Keywords—Data Mining, Heart Disease, Classification, Bagging, Boostingen_US
dc.subjectData miningen_US
dc.subjectHeart Diseaseen_US
dc.subjectBagging, Boostingen_US
dc.titleTowards Heart Disease Prediction using Hybrid Data Miningen_US
Appears in Collections:Masters Theses@CSED

Files in This Item:
File Description SizeFormat 
4649.pdf8.1 MBAdobe PDFThumbnail

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.