Download PDFOpen PDF in browserModelling of ambient noise levels in urban environmentEasyChair Preprint 20965 pages•Date: December 5, 2019AbstractThe study conducts time-series approach for analysing one year noise monitoring data. Support Vector Machine (SVM) technique is used as a modelling technique for time-series approach. The noise data is trained using 10-fold cross validation to get optimum hyper-parameters (γ, ε, C). The performance and accuracy of model is determined by statistical parameters like MSE, RMSE, MAPE in %, R2. The paper predicts an error of ± 2dB(A) with the implementation of Support Vector Machine (SVM). Keyphrases: Ambient Noise Level, Support Vector Machine, noise monitoring
|