News - Visitors' Abs

News: Abstracts of Selected Visitors' Talks

Wookey Lee's Visit

When Sensor Meets Face: Facial Detection Model with Sensor Device

by

Dr. Wookey Lee, Inha University

Tuesday, March 21, 2017 at 2:30pm

Abstract: Recently face recognition has widely been adopted to various kinds of application areas such as ATM machines,identity detection, face specification in smartphones and social network, security and CC cameras, etc. Hence, there exist many ongoing efforts in this area to provide devices which are capable to measure any face angles accurately. In this paper, we attempt to design a wearable sensor-based device to measure any face angles for the face recognition process. The novel device can measure any real degrees so that the problem of measuring the greater angles of faces can be overcome. Likewise, a face recognition method, FaceCube, is also proposed using clustering combined with Principal Component Analysis (PCA).The existing methods based on PCA suffer from the inevitable errors usually occur for great degrees, however, FaceCube by considering face angles through clustering approaches, gives more accurate results. By applying AP-clustering, PCA, and Support Vector Machine (SVM) using Radial Basis Function (RBF), the experiments on real face databases and synthetic faces demonstrate the super capability of FaceCube compared to the conventional methods.

Bio: Dr. Wookey Lee is a Full Professor at Informatics Lab at Inha University, South Korea. He is also the President of the KIISE Database Society of Korea, an Executive Committee member of the IEEE Technical Committee on Data Engineering (TCDE), and an Associate Editor of the Springer's World Wide Web journal.

Jian Pei's Visit

On Friday, September 16, 2016, Dr. Jian Pei, a CRC in Big Data Science at Simon Fraser University (SFU), gave a UofM Science interdisciplinary seminar on "Bringing data to life - building data science tools and platforms for business applications". He is a Professor of Computing Science, and an associate member in Department of Statistics and Actuarial Science as well as Faculty of Health Sciences, at SFU. He is also an ACM Fellow and an IEEE Fellow. Jiawei Han, Micheline Kamber, and Jian Pei. co-authored Data Mining: Concepts and Techniques, 3rd ed., Morgan Kaufmann, 2011.

Bringing Data to Life - Building Data Science Tools and Platforms for Business Applications

by

Dr. Jian Pei, Simon Fraser University

Friday, September 16, 2016 at 3pm

Abstract: In the last several years, we have conducted in-depth data science research in several exciting applications, such as health informatics. We built a series of tools for our industry partners to analyze and explore transaction, spatial, temporal and social data, and conduct contrast aspect analysis, viral marketing analysis, benchmark analysis, fraud detection, investigation and management, early prediction, and population index analysis. In this talk, I will present a brief overview of our data science analytics platform, and showcase some recent technical breakthroughs in contrast aspect analysis and continuous influence maximization for viral marketing, as well as some successful applications in industry.

Bio: Dr. Pei’s expertise is in developing business driven, technology-enabled data analytics for critical applications, including health informatics and viral marketing.

Kazutoshi Sumiya's Visit

Less-Conscious Information Retrieval Techniques for Location Based Services

by

Dr. Kazutoshi Sumiya, Kwansei Gakuin University, Japan

Tuesday, September 12, 2016 at 1pm

Abstract: We have developed methods which can deal with the users' interaction without conventional conscious searching. When a user generally performs map operations with certain information retrieval intentions (less-conscious), a system using our method can detect the specific operation sequences. For example, if the user performs zooming-in and centering operations, the user is narrowing down the search area to a certain location. We define such operation sequences as chunks. The system detects the chunks and uses them to analyze the user's operations and thereby detect the user's intentions. We have developed several prototype systems based on the proposed methods.

Bio: Dr. Kazutoshi Sumiya received his BE and ME degrees in instrumentation engineering from Kobe University in 1986 and 1988, respectively. Then he joined Panasonic (Matsushita Electric Industrial Co). He received his PhD in Information media from Kobe University in 1998. He left the company and became a lecturer at Kobe University in 1999, and then was promoted to an associate professor in 2000. He became an associate professor in 2001 at Kyoto University, a professor at the University of Hyogo in 2004 and a professor at Kwansei Gakuin University in 2015. He developed information dissemination systems and fusion technique for broadcast media and network media. He is developing next-generation information techniques. He was a chair of Database System special interest group (DBS) in the Information Processing Society of Japan (IPSJ) and a co-editor of IPSJ Transaction on Database. He is a director of IPSJ.

Charles Ling's Visit

On Thursday, September 01, 2016, Dr. Charles X. Ling, a Professor of Computer Science and a Science Distinguished Research Professor at Western University, gave a UofM CS seminar on "Mining lifestyle data for diabetes". He is a CAE Fellow and the Founder & CEO of GlucoGuide Corp. Charles X. Ling and Qiang Yang co-authored Crafting Your Research Future: A Guide to Successful Master's and Ph.D. Degrees in Science & Engineering, Morgan & Claypool, 2012.

Mining Lifestyle Data for Diabetes

by

Dr. Charles Ling, Western University (University of Western Ontario)

Thursday, September 01, 2016 at 1:30pm

Abstract: Diabetes is often called a "silent killer" affecting millions of people in Canada and worldwide, and is linked to and worsen by unhealthy lifestyle. In this talk, we will discuss a personalized recommendation system, called GlucoGuide, for Type-2 diabetes, based on mining users’ lifestyle data. Our system conveniently collects a variety of lifestyle data on patients' mobile device, mines the data with novel data-analytic algorithms, and sends personalized advice that can effectively improve users' condition. A small clinical trial with diabetic patients obtained promising results. GlucoGuide is also an university spin-off company, which strives to bring benefit to people with diabetes, and to cut down medical costs for organizations and insurance companies.

Bio: Dr. Charles Ling has been a Faculty member for over 25 years. He is a Professor in Computer Science, and Science Distinguished Research Professor at Western University. His research focuses on Big Data Analytics, machine learning, and data mining applications. He has published over 160 peer-reviewed research papers. Recently, he applies his research to mobile health, and created the GlucoGuide platform. It won the First Prize in Diabetes Research Day at the Schulich School of Medicine & Dentistry of Western.

Daniel Morgan's Visit

Challenges in a World of Large Databases and Business Intelligence

by

Daniel Morgan

Thursday, October 23, 2014 at 1pm

Abstract: This talk addresses some of the technical challenges you and your IT colleagues face managing large databases and business intelligence data. Learn the best practices for security and compliance. Discover the optimal architectures for integrating emerging big data technologies such as NoSQL and Hadoop. Expect to walk away with solutions that address issues confronting organizations.

Bio: Mr. Daniel Morgan, an Oracle ACE Director, is the former Chair of the Washington Software Association's Database SIG and has worked with many of our region's technology leaders in aerospace, e-commerce, and telecommunications. He began his IT career in 1969, working on IBM mainframes using Fortran IV. Later, Dan wrote the Oracle program for the University of Washington and he was the primary instructor there for 10 years. In addition to his work at UW, Dan served as an advisor to the outreach program at the University of California Berkeley, has been a consultant to Harvard University, and has been a guest lecturer on Oracle at the University of Oslo (Norway) and the University of Canterbury (New Zealand). Dan is a regular contributor at Oracle conferences and forums around the world presenting at conferences including OpenWorld, Collaborate, ODTUG Kaleidoscope, IOUC International Leadership Conference, and has presented or taught Oracle in Brazil, Bulgaria, Canada, Chile, Costa Rica, Denmark, Estonia, Finland, Germany, Japan, Latvia, Mexico, the Netherlands, New Zealand, Norway, Peru, Sweden, UK, Uruguay, and the US. Dan operates a full time consulting practice and is the "Morgan" behind the well-respected Morgan's Library (http://morganslibrary.org/) on the web.

Kyoji Kawagoe's Visit

Time Series Approximation for Temporal Data Management

by

Dr. Kyoji Kawagoe, Ritsumeikan University, Japan

Tuesday, October 21, 2014 at 1pm

Abstract: Time series, a sequence of data, can be easily obtained from sensors or simulation programs in various fields. A lot of work has been conducted on time series data management, especially its classification and similarity search over the past decades. However, the classification of a time series with high accuracy is still insufficient in applications such as ubiquitous or sensor systems. In this talk, a textual approximation of a time series, called TAX and l-TAX, is first presented to achieve high accuracy time series classification. Some ongoing approaches for multi-dimensional time series data management are also presented. The multi-dimensional time series, a sequence of high dimensional vectors, needs to be efficiently managed due to rapid progress in sensor technologies. Some approximation methods for its efficient data management are presented.

Bio: Prof. Dr. Kawagoe received his B.Eng. and M.Eng from Osaka University in 1975 and 1977, respectively. He received Ph.D from Tsukuba University in 1993. He joined Ritsumeikan University in 1997, while he had worked for NEC corporation since 1977. He is currently a full professor of College of Information Science and Engineering, Ristumeikan University. He has been serving as the vice-director of Ritsumeikan University Library since 2013. His research interests include multimedia data management and mining, information retrieval, and human communications. He was a visiting research associate at Lawrence Berkeley Labs of UC Berkeley from 1984 to 1985. He was also a visiting professor in both LMU Munich and Queensland University from 2009 to 2010. He is a member of IEEE, ACM, Data Base Society of Japan (DBSJ), IEICE, IPSJ, and ISSJ. He is currently the chair of ACM SIGMOD Japan Chapter.

Tamer Özsu's Visit

On Tuesday, May 29, 2012, Dr. M. Tamer Özsu, a Professor at the David R. Cheriton School of Computer Science of the University of Waterloo, gave a UofM CS seminar on "RDF data management using graph algorithms". He is an ACM Fellow and an IEEE Fellow. M. Tamer Özsu and Patrick Valduriez co-authored Principles of Distributed Database Systems, 3e, Springer, 2011.

RDF Data Management Using Graph Algorithms

by

Dr. M. Tamer Özsu, University of Waterloo

Tuesday, May 29, 2012 at 1pm

Abstract: Resource Description Framework (RDF) has been proposed for modeling Web objects as part of developing the "semantic web". It has also gained attention as a way to accomplish web data integration. As the volume of RDF data has increased, interesting data management issues have arisen. In this talk I will discuss some of our recent work in this area, focusing on two results: answering SPARQL queries over RDF graphs, and processing aggregate SPARQL queries. The first problem focuses on evaluating SPARQL queries with wildcards over an RDF graph that sees frequent updates. We propose an approach that maps both the RDF data and the SPARQL query into graphs and converts the query evaluation problem to one of subgraph matching. In order to speed up query processing, we propose an indexing mechanism and pruning rules to reduce the search space. The second problem addresses the processing of aggregation queries over large RDF data sets. We propose a processing approach that partitions aggregate queries into smaller parts (called star queries), processes these efficiently, and joins the results of star queries to obtain more general results. We develop indexes to assist in executing star queries and to facilitate joining their results.

Bio: M. Tamer Özsu is Professor of Computer Science at the David R. Cheriton School of Computer Science of the University of Waterloo. His research is in data management focusing on large-scale data distribution and management of non-traditional data. He is a Fellow of the Association for Computing Machinery (ACM), and of the Institute of Electrical and Electronics Engineers (IEEE), and a member of Sigma Xi.

Alan Mackworth's Visit

On Thursday, March 3, 2011, Dr. Alan K. Mackworth, a Professor of Computer Science and CRC in Artificial Intelligence at UBC, gave a UofM CS seminar on "Designing constraint-based agents". He served as the founding Director of the UBC Laboratory for Computational Intelligence (LCI). He is a Fellow of AAAI, the Canadian Institute for Advanced Research (CIFAR) and the Royal Society of Canada (RSC). David L. Poole and Alan K. Mackworth co-authored Artificial Intelligence: Foundations of Computational Agents, Cambridge University Press, 2010.

Designing Constraint-Based Agents

by

Dr. Alan K. Mackworth, University of British Columbia

Thursday, March 03, 2011 at 1pm

Abstract: In order to thrive, an agent must satisfy dynamic constraints deriving from four sources: its internal structure, its goals and preferences, its external environment and the coupling between its internal and external worlds. The life of any agent who acts without respecting those constraints will be out of balance. Based on this view of agents, I shall give four perspectives on the theme of designing constraint-based agents. The first is a discussion of the evolution of the concept of constraints in intelligent systems, from static to dynamic constraints. Second, I'll present our theory of constraint-based agent design and a corresponding experiment in robot architecture. Third, I shall sketch our work on the design of two assistive technology prototypes for people with physical and mental disabilities, who are living with significant additional constraints. Finally, our collective failure to recognize and satisfy with various constraints could explain why many of the worlds we live in seem to be so out of kilter. This approach hints at ways to restore the balance. Some of the work discussed is joint with Jim Little, Alex Mihailidis, Pinar Muyan-Ozcelik, Robert St-Aubin, Pooja Viswanathan, Suling Yang, and Ying Zhang.

Biography: Alan Mackworth is a Professor of Computer Science and Canada Research Chair in Artificial Intelligence at the University of British Columbia. He was educated at Toronto (B.A.Sc.), Harvard (A.M.) and Sussex (D.Phil.). He works on constraint-based artificial intelligence with applications in vision, robotics, situated agents, assistive technology and sustainability. He is known for his work in constraint satisfaction, robot soccer, hybrid systems and constraint-based agents. He has authored over 100 papers and co-authored two books: Computational Intelligence: A Logical Approach (1998) and Artificial Intelligence: Foundations of Computational Agents (2010). He served as the founding Director of the UBC Laboratory for Computational Intelligence. He is a Fellow of AAAI, the Canadian Institute for Advanced Research and the Royal Society of Canada.