مقاله انگلیسی رایگان در مورد داده کاوی باز برای اپیدمی دنگ تایوانی – الزویر ۲۰۱۸

By using a quantitative approach, this study examines the applicability of data mining technique to discover knowledge from open data related to Taiwan’s dengue epidemic. We compare results when Google trend data are included or excluded. Data sources are government open data, climate data, and Google trend data. Research findings from analysis of 70,914 cases are obtained. Location and time (month) in open data show the highest classification power followed by climate variables (temperature and humidity), whereas gender and age show the lowest values. Both prediction accuracy and simplicity decrease when Google trends are considered (respectively 0.94 and 0.37, compared to 0.96 and 0.46). The article demonstrates the value of open data mining in the context of public health care.

Background

Dengue fever remains one of the most serious infectious diseases in Taiwan. The Taiwan’s Center for Disease Control (TCDC) website (https://nidss.cdc.gov.tw/en/) reported 43,698 confirmed cases of dengue in 2015. Although the number decreased to only 743 in 2016, the Taiwanese government is still cautiously monitoring the epidemic , and implementing innovative strategies such as open data applications, which involve data (or data-driven) science, model development, and domain knowledge (Dhar, 2013; Zuiderwijk & Janssen, 2014; Zeleti et al., 2016; Hsu et al., 2017). Open data applications to predict dengue epidemic cover a variety of models such as linear and multi regression, moving average, weighted moving average, intra- and inter-seasonal autoregression, time series, networks, correlation, analytical hierarchy process, and their combinations. Models and variables have been tested in various countries including Thailand (Wongkoon et al., 2012), Malaysia (Dom et al., 2013; Dom et al., 2016), Taiwan (Chien and Yu, 2014), Bengal (Banu et al., 2014), Vietnam (Phung et al., 2015), Colombia (Eastin et al., 2014; Delmelle et al., 2016) and Saudi Arabia (Ibrahim Alkhaldy, 2017). Differently from these studies, we apply data mining to discover prediction themes in dengue open data. Online communities have been increasingly attracting attention to possible insights that could be derived from data mining based on open data sources (Brownstein et al., 2009; Santillana et al., 2014; Yang et al., 2017; Strauss et al., 2017). For example, the use of Google trends to develop data-driven models (classification-oriented models) of dengue epidemic is still under investigation. We expect that the insights derived from this approach to be relevant to the development of government management and control policies related to dengue. This study has three main purposes. First, we examine the applicability of data mining technique (data-driven classification technique) to predict dengue epidemics. Second, we investigate the role of Google trend data on prediction accuracy and simplicity of decision trees. Finally, based on the results obtained above we discuss management insights by conducting the mining results.