Title :
Short-term forecasting model of web traffic based on genetic algorithm and neural network
Author_Institution :
Dept. of E-Commerce & Logistic, Donghua Univ., Shanghai, China
Abstract :
Network traffic is an important load indicator that reflects the performance of the web system. Short-term forecast of web traffic is the base of effective overload control. Because of the complex and ever-changing network environment, web traffic is shown the characteristics of random and unexpected at most of the time scales. Hence, it is more difficult to improve the accuracy of traffic forecasts to get satisfactory results. In this paper, genetic algorithm is used in artificial neural network to optimize the structure design and weights firstly. Then, a web traffic forecasting model based on genetic neural network is proposed. The simulation result shown that the forecast result of this model is better than that based on BP and Elman neural network prediction model.
Keywords :
Internet; genetic algorithms; neural nets; road traffic; traffic information systems; BP neural network prediction; Elman neural network prediction model; Web traffic; genetic algorithm; load indicator; network traffic; short-term forecasting model; Artificial neural networks; Computational modeling; Forecasting; Genetic algorithms; Predictive models; Wavelet analysis; artificial neural network; forecasting model; genetic algorithm; network traffic; web system;
Conference_Titel :
Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC), 2011 2nd International Conference on
Conference_Location :
Deng Leng
Print_ISBN :
978-1-4577-0535-9
DOI :
10.1109/AIMSEC.2011.6010375
