department of computer science and engineering
spacer
News & Announcements 
 
2008-04-03
Time Aggregated Graphs: Modeling spatio-temporal
11:00am C4 Lab (DHE 244) Abstract: The underlying data of interest for many significant applications such as transportation networks is structured as a spatio-temporal network. With an increasing availability of data collected over such networks, it becomes important to utilize this data in improving the quality of analysis and query processing in such networks, suggesting a need for spatio-temporal network databases. In a spatio-temporal network, the network parameter values and the network topology can change with time, requiring them to be modeled as time-variant graphs. Conventional graph algorithms cannot be easily applied to the snapshot graphs at discrete time instants to evaluate frequent queries without accounting for relationships among snapshots. In this talk I will present Time Aggregated Graph, a model for representing spatio-temporal networks, that allows the modeling of time-variance of network parameters and topology. The model avoids the replication of the network for every instant of time, making it less memory expensive compared to existing representations. I will also present a case study of this model using routing algorithms on transportation networks. In this context, I will discuss the alternative semantics that the common queries can assume, given the time variance of the networks. I will also present the analytical and experimental analysis of the model and the algorithms in comparison with existing approaches. Bio: Betsy George is a PhD candidate in Computer Science at the University of Minnesota, Minneapolis, advised by Professor Shashi Shekhar. Her research interests lie in the area of spatial databases with emphasis on spatio-temporal networks.

 
2008-04-02
Graduate research at an iSchool
11:00am C4 Lab (DHE 244) Speaker: Sue Kase, Pennsylvania State University. Abstract: The College of Information Sciences and Technology (IST) is a new college at the Pennsylvania State University. The college offers multidisciplinary programs at the graduate and undergraduate level, with courses designed to provide students with a broad knowledge base and the skills needed to address complex problems through technology. Although not departmentalized, IST faculty and graduate students form a diverse group of researchers (e.g., human-computer interaction, artificial intelligence, organizational informatics, cyber security and privacy, geographic information systems, information and image fusion, information policy, social networks) working in collaboration to investigate information-centric research questions. IST PhD students often engage in projects spanning multiple labs and research domains. In this talk I will summarize three funded projects. The first is a light-weight agent-construction environment, called dTank, used as an experimentation and prototyping tool for adversarial behavior implementations. The second, Neo Nexus, is a 3-year community-oriented participatory design project with the goal of assisting community groups sustain technology learning and development in their organizations. Neo Nexus utilizes a patterns approach to identify organizational learning and IT skills learning in response to specific organizational challenges. The third is a computational modeling project investigating the effects of stress and task appraisal on human cognitive performance. A high-performance computing and evolutionary algorithm approach was developed to address the stochastic global optimization problem of fitting a complex cognitive model to individual subject data. A prototype of this optimization approach integrates the ACT-R cognitive architecture and a model of the serial subtraction task with a parallel implementation of a genetic algorithm. The optimization system is running on a supercluster at the National Center for Supercomputing Applications. Bio: Sue Kase is a Ph.D. candidate in the College of Information Sciences and Technology at the Pennsylvania State University. Before enrolling in the Ph.D. program, she received a M.S. degree in Computer Science and worked as a computer programmer for several companies within the PA, NJ, and NY area. More recently, she was an instructor in the Department of Computer Science and Engineering at Penn State for six years teaching both major and non-major courses. During summer sessions she participates as a mentor in diversity programs such as Women in Sciences and Engineering, Women in Sciences and Technology, middle- and high-school STEM workshops, and NSF Broadening Participation in Computing seminars. She has also attended the ACT-R Cognitive Architecture Summer School and the San Diego Supercomputer Center Summer Institute. In addition to cognitive modeling and human computer interaction, her interests include community informatics, information visualization, and high-performance computing.

 
2008-03-31
Automotive Electronics Product Design and Development at Johnson Controls, Inc.
March 31st (Monday) 11:00am C4 Lab (DHE 244) Speaker: Jeff Golden Abstract: The seminar discussion will open with a general introduction to Johnson Controls Inc. (JCI). It will then focus on JCI's Automotive Electronics product portfolio and key aspects related to the creation of such products including involved/emerging technologies, design and development methodologies, engineering disciplines utilized, noteworthy challenges, etc. Further attention will be given to the software intensive Embedded Systems based Telematics-Connectivity product line. Brief Bio: Jeff Golden is currently the Engineering Group Manager for Johnson Controls Inc., Automotive Experience Division, Advanced Electronics Development organization. Jeff has more than 30 years of progressively responsible experience in Electrical and Electronics Engineering and Technology Resource management. Prior to joining Johnson Controls, Jeff held various Engineering and Technical Leadership positions for Visteon, Ford Motor, and General Dynamics Corporations. Jeff holds a Bachelor of Science Electrical Engineering degree from Lawrence Technological University and a Master of Science Electrical Engineering degree from Wayne State University.

 
2008-03-30
People who received degrees in computer and information sciences (CS) in recent years are more likely to be employed in business and industry and to be working full-time than those who pursued several other majors. According to a recent NSF InfoBrief, 82 percent of CS majors who received bachelor's degrees in 2003, 2004, and 2005 were employed in business and industry, and 91 percent (along with engineering majors) had full-time jobs in April 2006. Among those with master's degrees, 76 percent were working in business and industry, and 93 percent were working full-time. The NSF InfoBrief also shows that CS graduates with bachelor's degrees had a median salary of $45,000, which was tied for second with health majors. CS graduates with master's degree made $65,000, tying engineering majors for first.

 
2008-03-28
Design Strategy for Knowledge Base Formation to Automate a Course Map Creation
11 AM @ C4 Lab (DHE 244) Speaker: Susan Lukose, University of Mississippi. ABSTRACT: In recent times, e-learning has become a popular alternative for the traditional classroom. The process of creating an online course which includes identifying the conceptual map, creating appropriate text, all the required multimedia object material and set up the tools for evaluating student performance is a very tedious and time consuming process. The Institute of Advanced Education in Geospatial Sciences (IAEGS) is a project initially funded by NASA, for developing state-of-the-art online courses in the field of Geospatial Sciences. IAEGS has already developed about 28 courses which are offered to any interested parties. According to them the process of creating a conceptual map (a detailed outline of the course) could take up to 2 months while entire course creation process could takes between 6 to 8 months. As the popularity of online learning is growing every day there is a need to create these courses quicker and at a lower cost. Automation of the course creation process could help to achieve this. An acceptable approach is in the form of a decision support system that guides the author during the process of concept and material selection. The goal of this research is to develop design strategies to create a knowledge base for a decision support system that could guide an author in these tasks. We propose to achieve this by automatically creating a ranked ontology from the domain specific glossaries using natural language processing techniques, statistical methods, graph theory and heuristic approaches. The application domain, "Remote Sensing and Photogrammetry”, is selected for the implementation and the effectiveness of automating this process will be analyzed.

 
2008-03-27
Induction of Multiclass Multifeature Split Decision Trees from Distributed Data
11 am @ C4 lab (DHE 244) Speaker: Dr. Nilesh Patel. ABSTRACT: The decision tree-based classification is a popular approach for pattern recognition and data mining. Most decision tree induction methods assume training data being present at one central location. Given the growth in distributed databases at geographically dispersed locations, the methods for decision tree induction in distributed settings are gaining importance. This paper describes one such method that generates compact trees using multifeature splits in place of single feature split decision trees generated by most existing methods for distributed data. Our method is based on Fisher's linear discriminant function, and is capable of dealing with multiple classes in the data. For homogeneously distributed data, the decision trees produced by our method are identical to decision trees generated using the Fisher's linear discriminant function with centrally stored data. For heterogeneously distributed data, a certain approximation is involved with a small change in performance with respect to the tree generated with centrally stored data. Experimental results for several well known data sets are presented and compared with decision trees generated using the Fisher's linear discriminant function with centrally stored data. BIO: Dr. Patel has received his PhD and MS in Computer Science in 1997 and 1993 from Wayne State University. He received his BS is in Control Engineering from Gujarat University, India in 1989. At present, Dr. Patel is working as Visiting faculty in the department of computer science and engineering at Oakland University, Rochester – MI. Prior to that he served as Assistant Professor and Software Engineering Manager at University of Michigan-Dearborn and Ford Motors Co / Visteon Corporations. At Visteon he pioneered the development of first in-vehicle computing system in 1999 while being at UMD he established their undergraduate and graduate Software Engineering Curriculum. His research interest is in the field of Multimedia Information Processing – specifically audio and video processing for indexing, retrieval and event detection, Pattern Recognition and Distributed Data Mining from a distributed heterogeneous data sources, Computer Vision with special interest in medical imaging and Mobile Computing – targeting to in vehicle computing and collaborative vehicular networks, vehicle data stream processing and its applications.

 
2008-03-26
Discovering Community in Social Networks
11:00am @ C4 Lab (DHE 244) Speaker: Alvin Chin PhD candidate Department of Computer Science University of Toronto ABSTRACT: The web has moved from a network of machines, to a network of servers, and now to a network of people (Wellman, 2001). Leaders are more likely to be actively involved within communities and subgroups are likely to form around active and like-minded people. Thus the capability to find subgroups and leaders within social networks will be important in understanding and possibly influencing the goals and activities of a community. In this talk, I will address the problem of finding evidence of community in online environments. I will describe a methodology for identifying communities that is divided into three steps: selecting possible people, collecting those people into possible subgroups and choosing the cohesive subgroups. I will then explain techniques that I used for each of the steps that is based on social network analysis, clustering and similarity. The methodology is tested with two case studies, one involving an online group, and the other involving a group of people using web video. The results are validated with human judgement and content analysis to identify and characterize communities in these groups. Finally, I will discuss implications of this research for social computing applications in pervasive and mobile computing. Bio: Alvin Chin is currently completing his PhD in Computer Science at the University of Toronto in the Interactive Media Lab, working with Professor Mark Chignell. His research has been funded by Bell University Labs. He graduated with a Bachelors degree in Computer Engineering and a Masters degree in Electrical and Computer Engineering from the University of Waterloo. Prior to his Ph.D he worked in industry researching emerging technologies in the wireless and pervasive computing area, especially Bluetooth and 802.11. His current research interests include social networking, computer-supported collaborative work, human-computer interaction, and pervasive and mobile computing. He can be contacted at achin@cs.toronto.edu or his web site at: http://www.imedia.mie.utoronto.ca/~achin.

 
2008-03-20
Semantics-based Web Service Discovery and Composition
Dr. Srividya Kona, Georgetown University, 11:00am, C4 Lab. ABSTRACT: Service-oriented computing is gaining wider acceptance. For Web services to become practical, an infrastructure needs to be supported that allows users and applications to discover, deploy, compose and synthesize services automatically. For this automation to be effective, formal semantic descriptions of Web services should be available. In this talk I will present the language USDL (Universal Service-Semantics Description Language) for formally describing the semantics of web-services. USDL is based on the Web Ontology Language (OWL) and employs WordNet as a common basis for understanding the meaning of services. USDL can be regarded as formal service documentation that will allow sophisticated conceptual modeling and searching of available web-services, automated service composition, and other forms of automated service integration. I will also present a semantics-based automated service discovery and composition engine. This engine employs a multi-step narrowing algorithm and is efficiently implemented using constraint logic programming. The salient features of our engine are its scalability, i.e., its ability to handle very large service repositories, and its efficient processing times for discovery and composition queries. BIO: Srividya Kona is a Postdoctoral fellow and Lecturer at the Department of Computer Science in Georgetown University. Her research interests include Service-Oriented Computing, Semantic Web Services, Semantic Web, Software Engineering, Constraint Logic Programming, and Language-based Security. She received her Ph.D. in Computer Science from the University of Texas at Dallas in December 2007. She received her M.S. in Computer Science from Texas Tech Univ., Lubbock in 2002 and her B. Tech. in Computer Science from NIT (previously known as REC), Warangal, India in 1999. She has over 5 years of industry experience. She worked as a software developer in SAP Labs (1999-2000) and Tyler Technologies (2001-2005). She was involved in the design and development of a Web service description language called USDL (Universal Service-Semantics Description Language) which won the best paper award at the European Conference on Web Services in 2005.

 
2008-03-19
Intelligent Agent-based Software Process Management for Large Software Development Projects – A Surveillance Tactical Software System as an Example
Dr. Ching-seh (Mike) Wu 11:00AM C4 Lab Abstract: Many software projects failed, because software is invisible. To successfully manage an invisible development project is highly challengeable. The intelligent agents use knowledge elicited from the project manager to define criteria of milestones. The intelligent agents work with the Project Attribute Monitoring and Predicting Associate II (PAMPA II), a Web-based automatic software project monitoring tool that I implemented to store knowledge describing an activity’s initial milestone and final milestone. The intelligent agents then use these activity’s criteria along with facts retrieved from the PAMPA II knowledge base to cooperate each other and to compare actual project progress to the planned progress. The resources, tasks, schedules, and milestones of the software project are described in the knowledge base. As the software development process evolves, the software development processes are monitored and the intelligent agents use the knowledge base to dynamically report recommendations, suggesting the software development that should be executed to best comply with the software project plan. The intelligent agents help the manager to determine the current phase of the project, identify the critical path, calculate the Earn Value, keep track of the progress, report problems, and suggest problem solutions during software development. The intelligent agents report risks and suggest problem solutions to help managers assure that a project is within budget, on time, and to customer satisfaction. BIO: Dr. Ching-seh (Mike) Wu is an Adjunct Faculty member of the Department of Computer Science and Engineering at Oakland University. He received his Master and Ph.D. both in Computer Science from U.S. Air Force Institute of Tech. in Dayton, OH and Texas A&M University in College Station, respectively. His past 20 years industrial experience and academic research has been focused on software development/engineering, software project management, and software process improvement, over which time he was both a faculty member of universities and a consultant for many software companies and organizations. He has published more than 30 peer-reviewed papers in the above areas. He had developed an artificial neural network simulator for the military weapon image processing/identification installed in F-16 Fighter. He was a team leader and a software project manager in developing the Tactical Software System for E2-T Hawkeye, a surveillance aircraft for protecting US aircraft carriers. He was also involved in the initial development of the ERP and Testing Software systems for both Hewlett Packard and Compaq computer corporations. He is a founding member of the Institute of Software Engineers in Dallas, Texas. He has been invited as a reviewer for IEEE Software Engineering Journal, Software Engineering Journal, and Journal of Information System.

 
2008-02-25
A widespread shortage of information technology (IT) graduates across North America is forcing Microsoft Corp. and other software companies to look to developing countries such as China to meet their needs, Microsoft Chairman Bill Gates says.

 
2008-02-13
Dr. Daniel Grosu Wayne State University Wednesday, February 20, 2008 11:50-12:50PM 244 DHE (C4 Lab) All are welcome. Pizza and Pop will be served at 11:45 a.m. ABSTRACT: Federated distributed computing systems are the next generation computing platforms for solving large scale problems in science and engineering. They are composed of geographically distributed resources owned by autonomous organizations. Resource management in such open distributed environments is a very complex problem. The existing resource management schemes do not explicitly address the fact that a rational participant is able to manipulate the resource management schemes in its own interest. To provide better performance and increase the efficiency it is essential to develop mechanisms for resource management that take into account the behavior of the participants and provide incentives to contribute resources. In this talk I will present my research in incentive-centered design for resource management in distributed computing. More specifically I will present the design of several incentive-based mechanisms for resource management in distributed computing systems. The design of these mechanisms is based on models and techniques from game theory and economics. BIO: Daniel Grosu is an assistant professor of Computer Science, Department of Computer Science at Wayne State University, Detroit, Michigan. His research interests include distributed systems and algorithms, resource allocation, computer security and topics at the border of computer science, game theory and economics. He has published more than 50 peer-reviewed papers in the above areas. He has served on the program and steering committees of several international meetings in parallel and distributed computing. He leads an NSF sponsored multidisciplinary research and training program in incentive-centered design at Wayne State University. He is a member of the IEEE, ACM, SIAM, AMS and the Game Theory Society. He received his Ph.D. in Computer Science from The University of Texas at San Antonio in 2003. (More information at: http://www.cs.wayne.edu/~dgrosu)

 
2007-12-10
Darrin Hanna, Assistant Professor in the CSE Department has been awarded 2007 Computer Science and Engineering Undergraduate Teaching Award by IEEE Computer Society. Congratulations!

 
2007-12-07
Computer and mathematical science occupations are projected to add 822,000 jobs—at 24.8 percent, the fastest growth among the eight professional subgroups. The demand for computer-related occupations will increase in almost all industries as organizations continue to adopt and integrate increasingly sophisticated and complex technologies. Growth will not be as rapid as during the previous decade, however, as the software industry begins to mature and as routine work is outsourced overseas. About 291,000—or 35 percent—of all new computer and mathematical science jobs are anticipated to be in the computer systems design and related services industry. The management, scientific, and technical consulting services industry is projected to add another 86,000 computer and mathematical science jobs. This expected 93-percent increase is due to the growing need for consultants to handle issues such as computer network security. Self-employment among computer and mathematical workers is anticipated to increase 19 percent, with most growth appearing among network systems and data communications analysts.

 
2007-11-26
Dr. Nilesh Patel has been funded by Infogation Corporation of San Diego, California to evaluate different schemes for remote activation of mobile devices. Congratulations!

 
2007-11-07
Seminar - Realizing the Potential of Wireless Sensor Networks
Dr. Loren Schwiebert, Wayne State University, Wednesday, November 7, 2007 11:50-12:50PM 244 DHE (C4 Lab) All are welcome. Pizza and Pop will be served at 11:45 a.m. Abstract: Wireless sensor networking is becoming a major topic in computer networking, since the advantages of connecting the physical and digital worlds offer opportunities for a wide range of applications. The possibility exists for important contributions to all facets of life, from agriculture and industrial manufacturing to healthcare. But interacting with the real world also introduces many novel problems to using sensors effectively. Unless these problems can be addressed, the full potential of wireless sensor networking will not be realized. After describing the many opportunities of wireless sensor networking, and some of the existing challenges to deploying sensor networks, we will present some of our recent work on wireless sensor networking designed to address some of these challenges. Loren Schwiebert received the B.S. degree in Computer Science (with a dual major in Mathematics) from Heidelberg College, Tiffin, OH, and the M.S. and Ph.D. degrees in Computer and Information Science from the Ohio State University, Columbus, OH. Since 1995 he has been a faculty member at Wayne State University, Detroit, MI, where he is currently an Associate Professor in the Department of Computer Science and Chair of the Graduate Committee. His research interests include wireless sensor networks, wireless communication, and interconnection networks. He is a member of the ACM, IEEE, and IEEE Computer Society.

 
2007-10-17
Wednesday Oct 17, 2007 11.50 AM – 12:50 PM in C4 Lab Title: A systems biology approach for pathway level analysis Speaker: Sorin Draghici, Ph.D. Abstract: A common challenge in the analysis of genomics data is trying to understand the underlying phenomenon in the context of all complex interactions taking place on various signaling pathways. A statistical approach using various models is universally used to identify the most relevant pathways in a given experiment. Here we show that the existing pathway analysis methods fail to take into consideration important biological aspects and may provide incorrect results in certain situations. Using a systems biology approach, we developed an impact analysis that includes the classical statistics, but also considers other crucial factors such as the magnitude of each gene's expression change, their type and position in the given pathways, their interactions, etc. The impact analysis is an attempt to a deeper level of statistical analysis, informed by more pathway-specific biology than the existing techniques. On several illustrative data sets, the classical analysis produces both false positives and false negatives while the impact analysis provides biologically meaningful results. This analysis method has been implemented as a web-based tool, Pathway-Express, freely available as part of the Onto-Tools http://vortex.cs.wayne.edu). ______________________________________________________________________ Sorin Draghici, Ph.D. Director of the Bioinformatics Core, Karmanos Cancer Institute Associate Professor Tel: (313) 577-5484 Dept. of Computer Science Fax: (313) 577-6868 Wayne State University 5143 Cass Ave, Room 431 State Hall, Detroit, MI, 48202 WWW: http://vortex.cs.wayne.edu/Sorin/ (personal) WWW: http://vortex.cs.wayne.edu/Projects.html (lab) He is an Associate Professor in the Department of Computer Science at Wayne State University, and the Director of the Bioinformatics Core at Karmanos Cancer Institute. He is active as an Associate Editor of IEEE/ACM Transactions on Computational Biology and Bioinformatics, as a regular NSF and NIH panelist in the areas of bioinformatics and biotechnology, as well as a reviewer for various technical journals. His publications include one book, 7 book chapters, 37 journal papers and over 30 peer-reviewed conference publications. 4 of these papers have over 100 citations each on Google Scholar. The book, Data Analysis Tools for DNA Microarrays, is a 550 page technical monograph published by Chapman and Hall/CRC Press and has over 100 citations, as well. His best known work today is in the area of computerized ontological analysis of high-throughput gene expression experiments. His research in this area has produced a set of tools (Onto-Tools), which has been made available as a service to the community, from the web page at http://vortex.cs.wayne.edu. These tools are currently used by more than 3,000 valid registered users from over 50 countries. The impact of this work should also be assessed in the light of the fact that Onto-Tools was the first tool of its kind. Since 2001, when the first tool was made available, the analysis approach he proposed has become the de facto standard in the second-stage analysis of microarray experiments. Currently, over 20 similar tools are available from other groups. All these tools use one or more of the statistical models he proposed, and virtually all publications describing these tools cite his work. Other notable results include our research in the genetic mechanisms of esothelioma which was chosen to illustrate the cover of the February 1 2004 issue of Clinical Cancer Research. Some of the research in neural networks was also featured the cover of the Proc. of Artificial Neural Networks in Engineering 1997 (published as a book by ASME Press, ISBN 0-7918-0064-4). In Oct 2006, one of his papers received the Fast Breaking Paper Award in Computer Science, from ISI/Thompson Scientific. According to ISI, their Essential Science Indicators which include the above mentioned award, comprise the top 1% of the papers in each field and each year http://www.esi-topics.com/fbp/fbp-october2006.html).

 
2007-09-24
The CSE Department, jointly with IEEE South East Michigan Computer Chapter and GL-Spin, is offering a free one day workshop on September 29. For details, please contact Professor Subra Ganesan at ganesan@oakland.edu or at 248-370-2206.

 
2007-08-16
The department is pleased to welcome Drs. Nilesh Patel, Gaungzhi Qu and Mohammad Siadat as new faculty members. Dr. Patel comes to OU from University of Michigan at Dearborn. He has worked at Visteon and Johnson Controls. Dr. Qu is from University of Arizona and has worked in the area of networking and security. Dr. Siadat comes from Henry Ford Hospital where he was working after completing his Ph.D. from Wayne State University.

 
2007-03-31
Automotive Electronics Product Design and Development at Johnson Controls, Inc.
March 31st (Monday) 11:00am C4 Lab (DHE 244) Speaker: Jeff Golden Abstract: The seminar discussion will open with a general introduction to Johnson Controls Inc. (JCI). It will then focus on JCI's Automotive Electronics product portfolio and key aspects related to the creation of such products including involved/emerging technologies, design and development methodologies, engineering disciplines utilized, noteworthy challenges, etc. Further attention will be given to the software intensive Embedded Systems based Telematics/Connectivity product line. Brief Bio: Jeff Golden is currently the Engineering Group Manager for Johnson Controls Inc., Automotive Experience Division, Advanced Electronics Development organization. Jeff has more than 30 years of progressively responsible experience in Electrical and Electronics Engineering and Technology/Resource management. Prior to joining Johnson Controls, Jeff held various Engineering and Technical Leadership positions for Visteon, Ford Motor, and General Dynamics Corporations. Jeff holds a Bachelor of Science Electrical Engineering degree from Lawrence Technological University and a Master of Science Electrical Engineering degree from Wayne State University.

 
2007-03-31
Automotive Electronics Product Design and Development at Johnson Controls, Inc.
March 31st (Monday) 11:00am C4 Lab (DHE 244) Speaker:Jeff Golden Abstract: The seminar discussion will open with a general introduction to Johnson Controls Inc. (JCI). It will then focus on JCI's Automotive Electronics product portfolio and key aspects related to the creation of such products including involved/emerging technologies, design and development methodologies, engineering disciplines utilized, noteworthy challenges, etc. Further attention will be given to the software intensive Embedded Systems based Telematics/Connectivity product line. Brief Bio: Jeff Golden is currently the Engineering Group Manager for Johnson Controls Inc., Automotive Experience Division, Advanced Electronics Development organization. Jeff has more than 30 years of progressively responsible experience in Electrical and Electronics Engineering and Technology/Resource management. Prior to joining Johnson Controls, Jeff held various Engineering and Technical Leadership positions for Visteon, Ford Motor, and General Dynamics Corporations. Jeff holds a Bachelor of Science Electrical Engineering degree from Lawrence Technological University and a Master of Science Electrical Engineering degree from Wayne State University.