JavaScript is disabled for your browser. Some features of this site may not work without it.
| dc.contributor.author | ASCHALEW, ABEBE BELETE | |
| dc.date.accessioned | 2021-10-01T10:36:29Z | |
| dc.date.available | 2021-10-01T10:36:29Z | |
| dc.date.issued | 2021-06 | |
| dc.identifier.uri | http://ir.bdu.edu.et/handle/123456789/12681 | |
| dc.description.abstract | Malaria affects more than 110 Million peoples worldwide with approximately 2414 diseases a day in average each year. Among this majority of the disease occur in Africa especially in sub Saharan including Ethiopia. More than 75% of the total area of Ethiopia is malarious, making malaria is the leading public health problem in Ethiopia. Neural Network was applied in a great variety of areas for data mining and knowledge discovery, Prediction, systems modeling, optimization and pattern recognition. Study has also shown the close relation between malaria incidences, weather pattern and clinical History. In these work we are implemented the ML algorithm to predict Malaria cases in Amhara region using clinical and environmental data. Those back history data are available from Ethiopian Meteorology Agency, Amhara Public health institute and various health centers of Amhara Region for train and predict the model. The collected data are used for input and output to the model for create awareness and alarm about malaria disease including transmission and their reproductive season using python simulation tool. The research singled out LSTM as the basis for the model, after considering various algorithms and their suitability for different problem domains . Four input and one output are used for deploy the model. The dataset are prepared using twelve years of data that are splitted in to 8 years of data used for traning,2 years of data used for validating and 2 years of data are used for testing. That means the model used approximately 70% data used for training, 15% for validating and 15% for testing. The performance of LSTM is shown to be the best fit model with minimum Mean Squared Error (MSE) of 0.096054 in amhara region, 0.049267 in south Gondar and 0.067914 in west Go jam which is small enough, and it indicates that the predictive model performs well. | en_US |
| dc.language.iso | en_US | en_US |
| dc.subject | ELECTRICAL AND COMPUTER ENGINEERING | en_US |
| dc.title | PREDICTION OF MALARIA ABUNDANCE USING ARTIFICIAL NEURAL NETWORK (ANN) IN ETHIOPIA: - A CASE STUDY OF AMHARA REGION | en_US |
| dc.type | Thesis | en_US |
