Prof. Jin Wang
Department of Mathematics and Computer Science, Valdosta State University, Valdosta, GA 31698, USA
Jin Wang is a Professor of Operations Research in the Department of Mathematics at Valdosta State University, USA. He received his Ph.D. degree from the School of Industrial Engineering at Purdue University in 1994. His research interests include Operations Research, Stochastic Modeling and Optimization, Supply Chain Management, Monte Carlo Simulation, Computational Finance, Portfolio Management, and Applied Probability and Statistics. Currently, he is working on Big Data and Data Mining fields. He has more than 28 years collegiate teaching experience in the field of quantitative methods and statistics at Purdue University, Florida State University, Auburn University, and Valdosta State University. Dr. Wang has been active in professional research activities. He has authored articles for publication in referred journals and conference proceedings. He has been active in INFORMS, IIE, and the Winter Simulation Conference and invited to give presentations, organize and chair sessions at national meetings. He has participated as a principal investigator in several research projects funded by federal and industrial agencies, including the National Science Foundation, General Motors, and the National Science Foundation of P.R. China. He was invited as a panel member at the National Science Foundation Workshop. Dr. Wang also served as a consultant for financial firms. His analytical Monte Carlo method using a multivariate mixture of normal distributions to simulate market data has made a great impact in education and the finance industry. This algorithm was selected as a graduate-level research project topic for many schools, such as, Columbia University Management Department, Carnegie Mellon University Economics and Finance Department, Tilburg University in Holland, Technische Universitaet Munich in Germany, Imperial College in London. This method was also implemented in many financial companies, such as, Zürcher Kantonal Bank, IRQ, Zürich Switzerland, Klosbachstrasse, Zürich, Switzerland, Norsk Regnesentral in Norway, Cutler Group, L.P., Altis Partners (Jersey) Limited, Windham Capital Management, LLC.
Speech Title:On Cluster Based Multivariate Outlier Detection in Data Mining
Abstract: In today’s big data age, mining outliers has become more and more important. It has wide applications in many fields, such as, detection of potential terrorist attacks, credit card fraud detection, network intrusion, clinical trials, severe weather prediction, and athlete performance analysis. On the other side, outlier removal also plays an important role in big data cleaning process. In this study, we propose a normal mixture model for the multivariate outlier detection. Due to the large volume and variety of big data, the classical Euclidean distance measure are not working well for mining multivariate outliers. Variables among different dimensions are usually correlated. Different distant measures will be discussed including the well-known Mahalanobis distance. The mixture model parameters are fitted via the EM algorithm. The K-means clustering algorithm is used to provide the initial inputs for the EM algorithm. The optimal k clusters and initial centers issues will be discussed. A nonsingular robust covariance estimation in calculating Mahalanobis distance will be introduce.
Prof. Haiying Ren
School of Economics and Management, Beijing University of Technology, China
Haiying Ren is an Associate Professor of Management
Science and Engineering in School of Economics and
Management at Beijing University of Technology,
China. He received his Ph.D. degree in Industrial
Engineering at University of South Florida in 2000.
His research interests include Technology and
Innovation Management, Multi-agent Simulation,
Knowledge Management, and Operations Management.
Currently, he is working on the behavioral modeling
of the processes of Inventions with knowledge
network representations. He has 14 years collegiate
teaching experience in the field of operations
research, computer simulation, decision theories and
analytical business methods at Beijing University of
Technology. Dr. Ren has published more than 40
papers in referred journals and conference
proceedings and authored or co-authored two
monographs and one book series. He has been
principal investigator in several research projects
funded by Beijing Municipal agencies and Ministry of
Education, including the Beijing Natural Science
Foundation and Beijing Social Science Foundation.
Dr. Ren won a Third Prize at 2010 Beijing Technology
Advancement Award as a key research team member.
Dr. Ren is an Associate Coordinator of Department of
Management Science and Engineering in School of
Economics and Management at Beijing University of
Technology, China. He is also an active member of
Operations Research Society of China.
Speech Title: Dynamic Micro-Mechanism of Breakthrough Inventions Based on Multilevel Knowledge Networks
Abstract: China has become a major patent nation, but not a strong innovation nation, as indicated by the lack of breakthrough inventions (BI’s). In order to increase the productivity of BI’s, we have to understand their mechanism, especially the dynamic and micro processes on individual and team levels. We tackle this problem by integrating invention-problem specific knowledge, individual knowledge, team knowledge and global knowledge in a multilevel knowledge network. Breakthrough inventions can then be modeled as the dynamic searching, learning and recombining operations of the multilevel knowledge network. We design methods for constructing such multilevel knowledge networks, propose building network models for the mental processes, inventive schemes and thinking methods of breakthrough inventors’, modeling dynamic micro-mechanism of breakthrough inventions and verifying the model by patent analysis and multi-agent simulation.
Address: Unit B on 15th Floor Eu Yan Sang
Tower,Nos.11/15, Chatham Road South
Kowloon, Hong Kong.
Tel : +86-28-86528478 (China)
+852-3500-0005 (Hong Kong)