摘要：This paper presents a method for particulatematerial PM10 modeling based on support vector regression(SVR). Specifically, we applied ε-support vector regression (ε-SVR) and ν-support vector regression (ν-SVR) to a set of datarecorded in the city of Santa Marta, Colombia, between 1999and 2016. The set of data was initially pre-processed, filteredand normalized, and then was used to fit the SVR models. Theparametrization and accuracy of each regression model arereported here. We used a month as the unit of time for themodels and analyzed the accuracy for one-step predictions.The final results of this work show the best parameters andprediction properties of the SVR models for pollution datamodeling in Santa Marta.
