000			03213ntm a2200277 a 4500
001			vtls000099137
003			KUKTEM
005			20251117113334.0
008			170419t2016 my a f am 000 0 eng d
020			_aTHE0001112(Local)
039		9	_a201905141152 _bnazirah _y201704191028 _zsaini
040			_aUMP
090			_aFSKKP .F38 2016 r Bc.
100	0		_aFatin Athirah Nasoha
245	1	0	_aPrediction of chronic kidney disease using artificial neural network / _cFatin Athirah Nasoha
260			_aKuantan, Pahang : _bUMP, _c2016
300			_axvi, 56 p. : _bill. (some col.) ; _c30 cm. + _e1 CD-ROM
500			_aFaculty of Computer Systems and Software Engineering
502			_aProject paper (Bachelor of Computer Science (Software Engineering) With Honours) -- Universiti Malaysia Pahang – 2016
504			_aBibliography : p. 54-55
520			_aThis research paper is about Prediction of Chronic Kidney Disease Using Artificial Neural Network. As known, a pair of kidneys is one of the most important organs in the body to keep the composition or make-up of the blood stable which lets body function. Then, where the kidneys are damaged, it can possibly affect human body since the waste and extra fluid can build up themselves inside of the human body. Currently, the physician in a hospital or clinic will make a deep review and analyze of the complexity of patient data which many attributes. So, it takes longer time to make a prediction. Normally, physicians are predicting CKD by knowledge and experience disease and it is a difficult task in medical environment. So, sometimes it can cause the human error. The overall objectives in this research; to study the neural network method for prediction of Chronic Kidney Disease (CKD), to implement neural network prediction using dataset for reduce human error and to evaluate the result of prediction CKD. The methodology used for this research is Artificial Neural Network (ANN) by applying back-propagation algorithm and sigmoid function. Data normalization is performed such that the range is from 0 to 1 which is a max-min normalization or also known as sigmoid normalization. The neural network designed with 24-4-1, 24-6-1 and 24-8-1 model. In this research, it trains the dataset of kidney disease factor and test process for dataset to see the actual result. Then, test the dataset and verify the accuracy result of kidney disease condition with calculate the Mean Square Error (MSE) value to calculate the error function. Besides, Root Mean Square Error (RMSE) will be used as data fitting method. From this study, 24-6-1 model for prediction of chronic kidney disease by using artificial neural network validates the best RMSE value; within 0.035 to 0.04 respectively. However, from the test dataset, 24-8-1 model is finally chosen as the RMSE value is within 0.09 to 0.12 respectively. The reason is because the result shows more accurate when test a new data.
610	2	0	_aFaculty of Computer Systems and Software Engineering _xDissertations
650		0	_aUniversities and Colleges _xDissertations
650		0	_aTheses
999			_aVIRTUA40 _c7003 _d7009
999			_aVTLSSORT0080*0200*0400*0900*1000*2450*2600*3000*5000*5020*5040*5200*6100*6500*6501*9992