Predicting Maternal Health Care Seeking Pattern Using Data Mining Techniques In Ethiopia.

Background: Utilization of maternal health care services could save unnecessary severerncomplications and death among women during pregnancy, delivery and after delivery.rnEthiopia is one of the countries with high maternal morbidity and mortality in sub-SaharanrnAfrica which needs more public health care effort in the country. rnObjective: The general objective of the study was to construct a model that can predict thernmaternal health care seeking pattern of reproductive age in Ethiopia. rnMethodology: The study followed Hybrid methodology of Knowledge Discovery Process tornachieve the goal of building predictive model using data mining techniques. rnTherefore, the overall research design was to build a model that can predict the maternalrnhealth care seeking pattern using J48 Decision tree and Naïve Bayes algorithms in Ethiopiarnfrom EDHS 2011 dataset. WEKA 3.6.8 data mining tools and techniques were employed as arnmeans to address the research problem. rnResults: The result of the study showed that the J48 Decision tree algorithm outperformsrnNaïve Bayes on the three of the outcome variables. For antenatal care the model was selectedrnwith an accuracy of 74.78%. Then for the second outcome variable (delivery care) the modelrnwas selected with an accuracy of 91.23% and area under receiver operating characteristics ofrn0.89. Finally for postnatal care the model was selected with an accuracy of 87.5% and arearnunder receiver operating characteristics curve of 0.80. The best attributes selected for each ofrnthe outcome variables are Place of Residence, Household Wealth Index, Women’srnEducational level, Husbands Occupation, Region, Husbands Educational level, Total numberrnof children, Media Exposure. rnConclusion: In general, the results obtained from this study were interesting andrnencouraging; it can be used as decision support for healthcare practitioner. The finding showsrnthat there is a regional difference in utilizing maternal health care service in the country, thusrnit is recommended that all the concerned parties should give due consideration for thesernregions, increasing maternal education at least up to primary level in all regions of therncountry, provision of opportunities for employment and poverty reduction especially in ruralrnparts of the region.

Get Full Work

Report copyright infringement or plagiarism