Large call-centers''arrival-rates prediction models based on the least squares support vector machine
大型呼叫中心人工呼入量的最小二乘支持向量机模型

In analyzing the data from a large call center, we find that arrival rates can be split into the daily-arrival-rate and the time-period-arrival-rate. Based on the least squares support vector machine theory(LS-SVM), predicting models of the daily-arrival-rate and the time-period-arrival-rate are established. Simulation experiments show that these models are good at regression and forecasting. Compared with Back-Propagation(BP) neural network prediction models, these models give better prediction results.