Open AccessMALARIA PREDICTION MODEL USING ADVANCED ENSEMBLE MACHINE LEARNING TECHNIQUES
Author(s) -
Yusuf Aliyu Adamu
Publication year - 2021
Publication title -
journal of medical pharmaceutical and allied sciences
Language(s) - English
Resource type - Journals
ISSN - 2320-7418
DOI - 10.22270/jmpas.v10i6.1701
Subject(s) - ensemble learning , artificial intelligence , computer science , machine learning , ensemble forecasting , random forest , regression , decision tree , support vector machine , classifier (uml) , statistics , mathematics
Malaria is a life-threatening disease that leads to death globally, its early prediction is necessary for preventing the rapid transmission. In this work, an enhanced ensemble learning approach for predicting malaria outbreaks is suggested. Using a mean-based splitting strategy, the dataset is randomly partitioned into smaller groups. The splits are then modelled using a classification and regression tree, and an accuracy-based weighted aging classifier ensemble is used to construct a homogenous ensemble from the several Classification and Regression Tree models. This approach ensures higher performance is achieved. Seven different Algorithms were tested and one ensemble method is used which combines all the seven classifiers together and finally, the accuracy, precision, and sensitivity achieved for the proposed method is 93%, 92%, and 100% respectively, which outperformed better than machine learning classifiers and ensemble method used in this research. The correlation between the variables used is established and how each factor contributes to the malaria incidence. The result indicates that malaria outbreaks can be predicted successfully using the suggested technique.