000			03074ntm a2200277 a 4500
001			vtls000099209
003			KUKTEM
005			20251117113336.0
008			170419t2016 my a f am 000 0 eng d
020			_aTHE0001086(Local)
039		9	_a201905171109 _bnazirah _y201704191622 _zsaini
040			_aUMP
090			_aFSKKP .A85 2016 r Bc.
100	0		_aNurul Atikah Mahadi
245	1	0	_aClassification of heart disease using backpropagation neural network / _cNurul Atikah Mahadi
260			_aKuantan, Pahang : _bUMP, _c2016
300			_axi, 55 p. : _bill. (some col.) ; _c30 cm. + _e1 CD-ROM
500			_aFaculty of Computer Systems and Software Engineering
502			_aProject paper (Bachelor of Computer Science (Computer Systems & Networking)) With Honours) -- Universiti Malaysia Pahang – 2016
504			_aBibliography : p. 39-41
520	3		_aNormally, a heart disease patient is diagnosed with the disease at the hospital. Nowadays, there is no system available yet to diagnose a patient with heart disease in a fast pace. It is becoming necessary and crucial for human kind nowadays especially in Malaysia to have a solution for a faster diagnosis of heart disease. When a patient is diagnosed at the hospital, it is a time consuming as the complexity of data that has many attributes and doctor or physician is the only professional people that need to analyse the data. Besides, the high chances of fault diagnose may occur as the human error cannot be avoided. The aim of this research is to classify the heart disease problem. The study of Backpropagation Neural Network (BPNN) is performed in order to achieve the purpose of this research. The objective of this research is the implementation of BPNN and the evaluation of result of classification of heart disease after the implementation is done. The research methodology of this research consists of six fundamental step which are literature review, data collection, data normalization, BPNN design, BPNN training and BPNN testing and verification. From the research methodology, the heart disease classification using BPNN is discussed and the formulas that need to be computed in order to get the accurate result. For BPNN testing and verification, 30% from the dataset of statlog heart disease taken from University of California-Irvine Machine Learning is used. The other 70% of the dataset is used for BPNN training. In order to classify the heart disease, a Matlab neural network toolbox called nprtool is used in the experiment. After the training and testing process is done, the confusion plot is generated and it shows the training, testing and overall result of classification. In this research, the result indicates that BPNN produced better result than method that used Fuzzy Rule and Hidden Naïve Bayes.
610	2	0	_aFaculty of Computer Systems and Software Engineering _xDissertations
650		0	_aUniversities and Colleges _xDissertations
650		0	_aTheses
999			_aVIRTUA40 _c7051 _d7057
999			_aVTLSSORT0080*0200*0400*0900*1000*2450*2600*3000*5000*5020*5040*5200*6100*6500*6501*9992