Predicting Under-Five Children Mortality Using Data Mining Techniques
| dc.contributor.advisor | Assefa, Temtim (PhD) | |
| dc.contributor.author | Tafesse, Hailemariam | |
| dc.date.accessioned | 2021-07-29T06:56:43Z | |
| dc.date.accessioned | 2023-11-18T12:47:41Z | |
| dc.date.available | 2021-07-29T06:56:43Z | |
| dc.date.available | 2023-11-18T12:47:41Z | |
| dc.date.issued | 2019-10-11 | |
| dc.description.abstract | The under-five deaths in Ethiopia represent 48% of all mortality. More than half of the under-five deaths occurred during the first year of life, and 53% of these before 2 months of age. Data mining is a collection of techniques for efficient automated discovery of previously unknown, valid, novel, useful and understandable patterns in large databases. The main objective of this study is to explore the potential applicability of data mining to predict the determinants, levels and pattern of under-five mortality in Ethiopia, particularly for the EDHS-CSA 2016 datasets. This can greatly support for policy makers, planners, and healthcare providers working on the control of under-five children mortality in Ethiopia. The methodology used for this research was a hybrid six-step Knowledge Discovery Process. The required data was received from DHS sites which covering the period 2016. The researcher used two popular data mining algorithms (Random Forest and PART Classifier) to develop the predictive model using a larger dataset (16,650 household). The researcher also used a 10-fold cross validation and 80% split test mode for data mining methods of the two predictive models for performance comparison purposes. The results indicated that the Random Forest is the best predictor with 99.32% accuracy. Thus, the study results reveal that several socioeconomic, demographic and health related variables associated with under-five mortality. This analysis results indicate that the best attributes selected for under-five mortalities are place of residence, household wealth; number of antenatal (ANC) visits, level of education, employment, religion, experience on pregnancy termination, experience on pregnant drug taking, vaccination of child, and size of child are major predictors. Therefore, attention should give to these predictors to reducing the risk of child mortality. The results from this study shows that applying data mining techniques could support in designing a predictive model for determining under-five mortality in Ethiopia. In the future, more classification studies by using a possible large amount of demographic and surveillance dataset records with epidemiological information and employing other classification algorithms, tools and techniques could yield better results. | en_US |
| dc.identifier.uri | http://etd.aau.edu.et/handle/12345678/27465 | |
| dc.language.iso | en | en_US |
| dc.publisher | Addis Ababa University | en_US |
| dc.subject | Predicting | en_US |
| dc.subject | Under-Five Children Mortality | en_US |
| dc.subject | Using Data | en_US |
| dc.subject | Mining | en_US |
| dc.subject | Techniques | en_US |
| dc.title | Predicting Under-Five Children Mortality Using Data Mining Techniques | en_US |
| dc.type | Thesis | en_US |