XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Tapak L, Shirmohammadi Khorram N, Hamidi O, Maryanaji Z. Predicting the Frequency of Human Brucellosis using Climatic Indices by three Data Mining Techniques of Radial Basis Function, Multilayer Perceptron and Nearest Neighbor: A comparative study. irje. 2018; 1111
URL: http://irje.tums.ac.ir/article-1-5886-en.html
1- PhD of Biostatistics, Modeling of Noncommunicable Diseases Research Center, Hamadan University of Medical Sciences, Hamadan, Iran
2- Ms.c of Biostatistics, Department of Biostatistics, School of Public Health, Hamadan University of Medical Sciences, Hamadan, Iran , n.shirmohamadi.kh@gmail.com
3- Department of Science, Hamadan University of Technology, Hamadan, Iran
4- PhD of Climatology, Department of Geography, SayyedJamaleddinAsadabadi University, Asadabad
Abstract:   (330 Views)
Background and Objectives: Identification of statistical models has a great impact on early and accurate detection of outbreaks of infectious disease and timely warning in health surveillance. This study evaluated and compared the performance of the three data mining techniquesin time series predicting of monthly brucellosis.
Methods: The present study includes monthly brucellosis counts and climatology parameters monthly from 2004 (March/April) to 2017 (February/March) in Hamadan located in west of Iran. The data were split into two subset of train (80%) and test (20%) sets. Three techniques, radial basis function (RBF) and multilayer perceptron (MLP) artificial neural network methods as well as K Nearest neighbor (KNN), were used to two subsets. The root mean square errors (RMSE), mean absolute errors (MAE), mean absolute relative errors (MARE), determination coefficient (R2) and intra-class correlation coefficient (ICC) criteria were used for comparing performances.
Results: Results indicated that RMSE (23.79), MAE (20.65) and MARE (0.25) for MLP were smaller compared with the other two models values. The ICC (0.75) and R2 (0.61) values for this model were also better. Thus the MLP model outperformed the other ones in predicting the used data. The most important variable was temperature among the climatology parameters.
Conclusion: MLP can be effectively utilized to diagnose the behavior of brucellosis over time. It is suggested that additional researches be done to detect the most suitable method to predict trend of this disease.
     
Type of Study: Research | Subject: General
Received: 2018/03/5 | Accepted: 2018/03/5 | Published: 2018/03/5

Add your comments about this article : Your username or Email:
Write the security code in the box

Send email to the article author


© 2017 All Rights Reserved | Iranian Journal of Epidemiology

Designed & Developed by : Yektaweb