The nature of statistical learning theory
The nature of statistical learning theory
Choosing Multiple Parameters for Support Vector Machines
Machine Learning
Software reliability forecasting by support vector machines with simulated annealing algorithms
Journal of Systems and Software
Time Series Forecasting for Dynamic Environments: The DyFor Genetic Program Model
IEEE Transactions on Evolutionary Computation
Support vector machine with adaptive parameters in financial time series forecasting
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Traffic forecasting is critical for mobile operators to grasp market trends and control network capacity. Therefore, an improved method of forecasting for mobile traffic is presented in this paper. The traffic is divided into the general trend part and seasonal part to forecast them respectively. The general trend is predicted by fitting the curve of general trend on tariff level; and the remaining seasonal part is predicted by simulated annealing-support vector regression machine (SASVR) which uses simulated annealing (SA) to select the super-parameters of SVR. The experimental results show that not only this method improves the prediction accuracy but it provides mobile operators with a visual expression of the relationship between traffic and the tariff level.