Statistical Comparison of the Impacts of Weather Variablesand Particulate Matter on Bacterial Density:Methodological Research

Abstract

Objective: In this study, the number of bacteriae in theair at five different regions in Eskişehir (Turkey) according to thesettle plate method and some meteorological factors such as tempera ture and relative humidity were collected and analyzed statistically.The data were used to predict the amount of bacteriae using machinelearning algorithms. Material and Methods: Air samples were takenfrom specified sampling points using the settle plate method. Therandom forest method, multivariate adaptive regression splines, sup port vector machine and ordinary least squares models were em ployed to analyze a qualified data set collected through 5 differentmonitoring locations in Eskişehir city, Turkey. All analyses wereconducted using RStudio Software (version 1.2.1335). Results: Theresults revealed that all variables including weather characteristicsand particulate matters were found statistically significant on theamount of the bacteria. Among modeling methods, the random forestmodel overperformed the others in terms of root mean square. Con clusion: Results might be useful for researchers studying bacterialdensity. These findings may provide policy-makers with crucial in formation for a better climate and health policy development, whichhelps address the conflict between development and air pollution,may be applicable in Turkey and other countries as well.