DATA MINING BASED CRITICAL ANALYSIS AND CLASSIFICATION OF INFECTIOUS DISEASE FOR EFFECTIVE DIAGNOSIS AND TREATMENT

DESALEW, MOLALIGN

dc.contributor.author	DESALEW, MOLALIGN
dc.date.accessioned	2020-03-24T05:46:10Z
dc.date.available	2020-03-24T05:46:10Z
dc.date.issued	2020-03-24
dc.identifier.uri	http://hdl.handle.net/123456789/10775
dc.description.abstract	Infectious diseases are the leading causes of illness and deaths in low and middle-income countries. They are easily transmitted from an infected individual to others through causative organisms, such as bacteria, virus, protozoa, fungi and parasites. The disease causes high morbidity, mortality and economic crisis on countries until they are treated properly. Due these cases the healthcare sectors are demanding high cost, less performance and poor patient safety care management. However, studies on such diversified issues have not done much. On the other hand, the huge amount of data is available in the health sector. Therefore, we proposed a data mining process model to extract significant patterns from patient records. We have collected a data from two government hospitals, Felege Hiwot Comprehensive Specialized Referral Hospital and Addis Alem Hospital located in Amhara National Regional State, Bahir Dar city. A total of 3017 sample datasets are collected in developing models to classify datasets of cholera, malaria and TB. The findings of this study showed that all the models built by ID3 and J48 Decision Tree, SVM and Artificial Neural Network classifiers in the three datasets has different classification accuracy. Model comparison is done based on TP (sensitivity) and FP (specificity) rates, precision, recall, F-measure, ROC area and accuracy. 10-fold cross validation test option is used to check the performances of each classifier. This suggests that the model built by the SVM classifier performs slightly better in the classification of cholera disease with a classification accuracy of 83.43%, the model built by J48 classifier performs better in the classification of malaria and TB with a classification accuracy of 98.55% and 99.2% respectively. The study showed that data mining techniques have used effectively for classifying infectious disease for effective diagnosis and treatments activities, optimize patient safety care, epidemic control and advanced preparation. The outcome of the study can be used as an inference material for physicians to support them to make more consistent diagnosis and treatments of the disease.	en_US
dc.language.iso	en	en_US
dc.subject	Information Technology	en_US
dc.title	DATA MINING BASED CRITICAL ANALYSIS AND CLASSIFICATION OF INFECTIOUS DISEASE FOR EFFECTIVE DIAGNOSIS AND TREATMENT	en_US
dc.type	Thesis	en_US