DESIGNING A MACHINE LEARNING MODEL  FOR CHILD BIRTH DELIVERY METHOD PREDICTION

AMSALU, FEKADU YENEALEM

dc.contributor.author	AMSALU, FEKADU YENEALEM
dc.date.accessioned	2022-03-07T07:33:44Z
dc.date.available	2022-03-07T07:33:44Z
dc.date.issued	2021-10-12
dc.identifier.uri	http://ir.bdu.edu.et/handle/123456789/13151
dc.description.abstract	Childbirth delivery method is an essential determinant for the health of mother and child. Choosing the wrong type of childbirth delivery causes different kinds of complications on the mother and new born child. This research work aims to design a machine learning model for prediction of childbirth delivery method which can support the gynecologists in decision making process. Predicting the correct delivery method has significance in reducing maternal mortality and morbidity rate by avoiding complications associated with wrong delivery method. KNN, RF and SVM algorithms are implemented as base learners and a Stacking Ensemble method with ANN algorithm as super learner to develop childbirth delivery method prediction model. Confusion matrix and its derivatives; Accuracy, Precession, Recall and F1-score are used to evaluate the performance of the proposed model. Execution Time is also used to compare the time taken to execute the models. The study result shows that the model implemented using Stacking Ensemble method with ANN achieved the best Recall and Accuracy results of 90.9% and 95.2% respectively.	en_US
dc.language.iso	en_US	en_US
dc.subject	INFORMATION TECHNOLOGY	en_US
dc.title	DESIGNING A MACHINE LEARNING MODEL FOR CHILD BIRTH DELIVERY METHOD PREDICTION	en_US
dc.type	Thesis	en_US