Modeling of dengue outbreak prediction in Malaysia: A comparison of Neural Network and Nonlinear Regression Model

Abstract

Malaysia has a good dengue surveillance system but there have been insufficient findings on suitable model to predict future dengue outbreak. This study aims to design a Neural Network Model (NNM) and Nonlinear Regression Model (NLRM) using different architectures and parameters incorporating time series, location and rainfall data to define the best architecture for early prediction of dengue outbreak. Four architecture of NNM and NLRM were developed in this study. Architecture I involved only dengue cases data, Architecture II involved combination of dengue cases data and rainfall data, Architecture III involved proximity location dengue cases data, while Architecture IV involved the combination of all criterion. The parameters studied in this research were adjusted for optimal performance. These parameters are the learning rate, momentum rate and number of neurons in the hidden layer. The performance of overall architecture was analyzed and the result shows that the MSE for all architectures by using NNM is better compared by NLRM. Furthermore, the results also indicate that architecture IV performs significantly better than other architecture in predicting dengue outbreak and it is therefore proposed as a useful approach in the problem of time series prediction of dengue outbreak.

An Innovative Big Data Predictive Analytics Framework over Hybrid Big Data Sources with an Application for Disease Analytics

“…Similarly, Husin et al [12] compared the NN with a nonlinear regression model (NLRM). Their comparison on applying the NN and NLRM to the same data (i.e., the same five Malaysian cities as Yusof and Mustaffa) showed that the NN led to smaller mean square errors (MSE): 0.028 for the NN and 26.054 for the NLRM.…”

“…Due to the relation between weather conditions and the proliferation of the dengue vector mosquito, statistical and machine-learning models can be used for forecasting the number of dengue cases based on weather conditions [11,12,21]. The detection of such outbreaks in advance would give to public health agencies time for dealing with the high number of cases and even decrease them.…”

Effective Search Engine Spam Classification

“…Normalization is mainly helpful for classification, as it progress accuracy and efficiency of mining. There are three types of normalization techniques namely Z-Score Normalization [11], Decimal Point Normalization [12], and Min-Max Normalization [13]. In Z(Zero Mean)-Score Normalization, the data is normalized based on the mean and standard deviation.…”

Predictive Modeling of Emerging Infections

“…The oldest and most widely used ANNs are feedforward neural networks with backpropagation learning rule (Rumelhart et al, 1986). Husin et al (2008) developed an ANN and a nonlinear regression model to predict dengue outbreaks in Malaysia and compared the results of those two methods. They examined weekly dengue cases in five districts for the years 2004-2005.…”