|
|
|
Web Service Management System (WSMS) Over the years, numerous applications have been developed for effcient data access and reuse. These data-centric solutions were based on the premise of an interactive Web where users involvement was necessary for virtually any Web-based transaction. In recent years, Web-related research has shifted from how to access such Web content to exploring new forms of what can be published and accessed in the Web. Application reusability is key to enabling this paradigm shift. TheWeb that had been essentially a repository of static content has started a steady evolution to become a vibrant environment where applications can be automatically invoked by other Web clients. These advances have brought revolutionary changes in building distributed applications with the introduction of the service oriented computing (SOC) paradigm. The service-oriented Web represents an attractive paradigm for tomorrow's interactions spanning a wide range of domains from e-economy to e-science and e-government. Enabling the service Web requires a comprehensive infrastructure to develop, deploy, publish, discover, compose, monitor, and optimize access to Web services. Simply put, materializing the service Web necessitates a Web Service Management System (WSMS) that provides for Web services what DBMSs has provided for data. In such a system, Web services would be treated as first-class objects that can be manipulated as if they were pieces of data. Services are inherently different from data. First, data in DBMSs are passive objects with a set of known properties, e.g., structure, value, functional dependencies, integrity constraints. Web services are active and autonomous entities that have a function rather than a value. Moreover, unlike data in DBMSs, Web services have a run-time behavior that may not be precisely known when they are invoked. Second, accessing a service on the Web is potentially far more complex than retrieving data from a DBMS. To access a Web service, the service requester must search in one or more service registries, understand the different services' syntactic and semantic descriptions, select the services providing the requested functionality, understand their communication protocols, and finally engage in a sequence of message exchange with the selected services. In more complex scenarios, requests for services may require the composition of several Web services that may span different administrative domains. A WSMS must hide these complexities and enable users to access services transparently as they would access data in a DBMS. My research group at the ECEG Research Lab is focused on defining and developing a comprehensive WSMS. Current Research Topics
Service optimization research is geared towards building a formal framework for enabling Web service query optimization. The objective is to provide a solid foundation upon which reasoning about service queries can be performed, allowing the development of (so far lacking) theories and algorithms to efficiently query Web services. Note that the focus of our work is not on the design of service query languages. Instead, we provide the fundamental support for enabling the definition of such languages. In this regard, we are working to address some of the foundational issues regarding optimization in the emerging field of service science.
Web service composition refers to the process of combining several Web services to provide a value-added service. It is emerging as the technology of choice for building cross-organizational applications on the Web. Web service composition has recently taken a central stage as an emerging research area. Standardization efforts are under way for supporting Web service composition (e.g., BPEL4WS). However, these techniques and standards provide little or no support for the semantics of participant services, their messages, and interactions. Additionally, they generally require dealing with low level programming details which may lead to unexpected failures at run-time. We are working on presenting novel approaches for the automatic composition of Web services on the envisioned service Web.
In the post-genomic age, there are currently two main challenges facing the bioinformatics research community, namely functional discovery and data integration. Time consuming development of customized interfaces is needed so that different models can interoperate. The integration of models through software applications will ultimately allow us to establish dynamic models that can be used for various medical experiments. While such applications have been investigated, none has been based on modeling biological entities using Web services. We are working on using a Web service mining framework so that models can be mined for the purpose of discovering biological and molecular interactions and reactions.
A plethora of Web services competing in offering similar functionalities are expected for the new ``service Web". A key requirement then is to provide mechanisms for the quality access and retrieval of services. In essence, the major challenge lies in providing a trust framework for enabling the selection and composition of Web services based on trust parameters. The fundamental rationale behind the need for trust is the necessity to interact with unknown entities. The lack of a global monitoring system for the service-oriented environment exacerbates the problem of trust. By definition, Web services are autonomous (i.e., provided by independent service providers), highly volatile (i.e., low reliability), and a priori unknown (i.e., new or no prior history). Web services can have no particular behavior mandated on them and are highly volatile as they are subject to frequent fluctuations during their lifetime (e.g., unavailability, changes in quality and performance). Hence, resolving trust issues is a determining key to the selection of services. We are working on developing a comprehensive trust framework for service-oriented environments.
Service-Oriented Enterprises (SOEs) are essentially composed Web services that are outsourced from autonomous Web service providers. Changes may be very common and frequent due to the highly dynamic business environment involving different service providers. In this regard, change management is a paramount research issue in enabling the deployment of service oriented enterprises. There are two types of changes, including top-down changes (initiated at the SOE level) and bottom-up changes (initiated at the Web service level). Manually handling these changes demands intensive manual work. We are working to automate the process of detecting, propagating, reacting, and validating changes in a SOE.
Wireless communication has several advantages over traditional wireline communication. In wireless environments, data is transmitted on the air, which does not require physical media be built between any two users before communication takes place. Furthermore, mobile users in wireless environments can access information while on the move. At Eceg, we are working on improving wireless data access efficiency through different access methods. The work involves investigating uniform and nonuniform access patterns to assess the performance of the proposed access methods. Moreover, exploration of the use of range queries, varied-size data, and other data distributions are also being studied.
Research Projects E-Government: WebSenior The healthcare needs of senior citizens are expected to surge as baby-boomers start retiring. As a result, this trend will put tremendous stress on the already stretched and cost spiraling healthcare systems. Of special interest in this project is senior citizens healthcare. Senior citizens are challenged in ways fundamentally different from the general population. Receiving services in a convenient way is not a matter of choice for them, but of necessity. Failing health accompanied with limited mobility may force senior citizens into unhealthy choices. Recent advances in Web standards and technologies offer a cost effective leverage to maintain and use large healthcare information systems. New studies suggest that senior citizens are now the fastest growing group of World Wide Web users. This pilot project, a collaborative effort between Virginia Tech and the Virginia Department for the Aging (VDA), focuses on the development of a Web service middleware infrastructure. The infrastructure will enable the seamless interoperation of the VDA's autonomous information systems based on new Web service standards. E-Commerce: WebBIS This project looks at the database and Web services support for electronic commerce. As proof of concept, we have used Web services composition in defining virtual enterprises. We have also deployed a novel approach for the preservation of trade secrets in a Business-to-Business (B2B) environment that involves trade among competitors. In B2B collaborations information may be exchanged of a sensitive nature, forming a business trade secret. We use a Peer-to-Peer (P2P) approach for preserving trade secrets in B2B interactions. We introduce a set of techniques based on data perturbation for preserving data privacy. |
|
|
|
|
|
|
|
Publications
(DBLP
Entry) Book (Authored)
Book (Edited)
Book Chapters
Refereed Journal Publications
Short Articles
Refereed Conference Publications
Other
Publications Theses
Non-refereed Publications
|
|
Current PhD Students
Former PhD Students
Master Students
|
|
Editorial Boards
Guest of Special Issues
Conference Technical Chair
Conference Panel Member
Tutorial Chair
Conference Program Coordination
Conference Program Session Chairing
Conference Program Committee Membership
Pictures from
the ECEG Lab
|