Predicting the Status of Anaemia In Women Aged 15-49 By Applying Data Mining Techniques Using The 2011 Ethiopia Demographic And Health Survey (Edhs) Datasets
No Thumbnail Available
Date
2013-06
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
Background: Anaemia is recognized as a major public health problem throughout the world. Globally, more than one-third of women and two-fifths of young children in the world are affected by anaemia. In Ethiopia, based on the 2005 and 2011 EDHS, 27% and 17% of women aged 15-49 are anaemic, respectively. Indeed, as per the WHOs classification on anaemia level, the prevalence of anaemia in Ethiopia is distinguished as moderate public health problem. Previous studies conducted using conventional statistical methods witnessed the public health problem of anaemia in Ethiopia. Objective: The purpose of this research is to predict the status of anaemia in women aged 15-49 by applying data mining techniques using the 2011 EDHS dataset. Methodology: The Knowledge Discovery in Database (KDD) which is applied to academic research was adopted to extract useful information from the dataset containing 6697 records. J48 Decision Tree and PART Rule induction were employed to build the predictive model. And also WEKA open software was used as a data mining tool to implement the experiments. Results: The findings of this study revealed that all the models built using J48 decision tree and PART rule induction algorithms with all attributes have high classification accuracy and are generally comparable in predicting any anaemic cases. However, comparison that is based on the detailed performance measures suggests that the PART Unpruned model with all attributes perform better in predicting any anaemic cases with an accuracy of 95.2%. Conclusion: The study showed that applying data mining techniques on the secondary data of EDHS to build a model that predicts any anaemic cases is possible. Thus, the outcome of this study helps health care planners and policy makers to design a proper and suitable preventive and control programs to combat anaemia. Keywords: EDHS Dataset, Anaemia, Data Mining, Predictive Model
Description
Keywords
EDHS Dataset, Anaemia, Data Mining, Predictive Model