Topic：Using a Dynamic Artificial Neural Network for Pattern recognition and Forecasting
Lecturer：Professor M.Ghiassi Leavey School of Business, Santa Clara University, USA.
Time：2010.11.17，Wednesday, 3:00-5:00 p.m.
Location：New Main Building A1148
Host：Professor Liu Lu
Abstract：
Neural networks have shown to be an effective method for pattern recognition and forecasting time series events and nonlinear processes. Traditional research in this area uses a network with a sequential iterative learning process based on the feed-forward, back-propagation approach. We present a dynamic neural network model, called DAN2, for forecasting various processes that uses a different architecture than traditional models. DAN2 is a data driven; feed forward, multilayer, dynamic architecture that is based on the principle of learning and accumulating knowledge at each layer and propagating and adjusting this knowledge forward to the next layer. Model building is automatically and dynamically repeated until a model that accurately captures the behavior of the process is determined. The resulting model is then used to forecast future values. To assess the effectiveness of this method, we forecasted a number of standard benchmarks in time series and nonlinear forecasting research from the literature, as well real life applications. Results show that this approach is more accurate and performs significantly better than the traditional neural network, autoregressive integrated moving average (ARIMA), and nonlinear regression models.
Lecturer：
Manoochehr Ghiassi is professor of Information Systems, a Breetwor Fellow and Director of MSIS program at the Leavey School of Business at Santa Clara University, Santa Clara, California. Prior to becoming the director of the graduate program, he served as the department chair for 12 years.
He received a BS from Tehran University, Iran, and an MS in Econometrics from Southern Illinois University at Carbondale. He also received an MS in Computer Engineering and a PhD in Industrial & Mechanical Engineering both from the University of Illinois at Urbana-Champaign.
His research interests include artificial neural network, Business Intelligence, Information Retrieval, software engineering, software testing, Intelligent Supply Chain Management and simulation modeling. His publications have appeared in IEEE, ACM, Simulation and Industrial Engineering Journals. His current research projects address development of artificial intelligence models for cancer diagnostics and pre-release box office revenue forecasting of movies using social networking information.
School of Economics & Management
SEM Postgraduate Association
2010.11.16
