Description of Half Day Tutorials
TA01 TP01 TA02 TP02 TA03 TP03 TA04 TP04 TA05 TP05 TA06 TP06 TA07 TP07
TA01: Architecting and Developing Message-Oriented Web Services
Savas Parastatidis, University of Newcastle upon Tyne,UK
Best practice in Web Services architecture and development has moved on since the days where the technology was used as a platform-agnostic RPC mechanism. Best practice in modern Web Services architecture is to consider problems in terms of explicit message passing between autonomous computational entities, in arbitrary message exchange patterns over arbitrary protocols. Over time Web Services middleware has evolved to support this notion and now both .Net and Java provides a message-oriented abstraction for composing Web Services into applications and security and policy infrastructure.
This tutorial is derived from an extended version of a talk given for the ACS Web Services SIG and an abbreviated version of a commercial training course, and supported by a chapter in the forthcoming MIT Press book, “Readings in Service Orientation: The Web Services Phenomenon.” It will last for approximately 3 hours and introduce the message-oriented aspects of common Web Services middleware and show how to apply its features to building Web Services with interesting transport-neutral message exchange patterns and security requirements.
The tutorial will be code-focussed and will take the audience through the design and implementation of a fully-featured Web Services application based on the classic noughts-and-crosses (or tic-tac-toe) game. Once the implementation has progressed to the point where the game can be played between two remote players, it will be used as a test bed to illustrate various aspects of WS-Security (via WS-Policy) to show how messages exchanges can be made robust against tampering, non-repudiable, and private.
TP01: Basis for Automatic Web Service Composition
Daniela Berardi, Universitá di Roma "La Sapienza"
Massimo Mecella, Universitá di Roma "La Sapienza"
Time: 14:00 — 17:30
The tutorial aims at providing a deep comprehension of Web Service Composition problem and automated techniques to tackle it. Web Service Composition is currently one the most hyped and addressed issue in the Service Oriented Computing. Starting from an analysis of current technologies and standards for Web Service composition, the tutorial will lead the attendees to consider formal models at the base of current proposals, and techniques that can be fruitfully considered to address automatic composition synthesis in each of them.
TA02: Emerging Infrastructures for Information Dissemination - XML Aware Networks
K. K. Ramakrishnan, AT&T Labs. Research
As networking technology matures, it is being used mostly for information exchange and dissemination. The Web has become a dominant source of information for the masses, worldwide. Larger and larger numbers of sources of information covering a tremendous variety of topics are being created, and it has become important to manage the information exchange between the information sources and interested receipients.
Information is increasingly being created, exchanged and stored in the Extensible Markup Language (XML). XML is suitable for this purpose because of its flexibility and self-describing nature: it is human readable, while at the same time it is convenient for machine processing. Examples of Information on the Web that is XML-based include technical journals, bibliographical databases, information on Yahoo, EBay, the Department of Defense, the US Congress and the Justice Department, to name a few. Even network management is moving to an XML based environment because of its flexibility.
Communicating XML-based information presents opportunities and challenges from a networking perspective. A network that can efficiently forward XML data can be advantageous, as it can offload the filtering of tremendous volumes of data from receivers. Or, it can reduce the load on the source and the network, by avoiding sources having to dissemminate information to individual receivers. Using network layer multicast, while alleviating the problem somewhat, is not effiecient enough. A semantic-aware switching infrastructure can overcome these limitations. However, XML aware forwarding by switches and routers requires deep packet lookup for forwarding based on the XML content. Further complicating this is that XML content can in fact span multiple packets at the lower layers. If the overlay switching and routing functionality has the appropriate acceleration for forwarding XML content at high speeds, it would be a significant advantage. There is a growing industry effort to provide such accelerators as part of XML Switches and Routers.
This tutorial introduces the area, and describes the challenges in providing an XML Aware Network.
This tutorial is a result of the continuing collaboration of the authors with Divesh Srivatsava, Nick Koudas, Yin Zhang and Bill Fenner.
TP02: Application Issues of XML Schema Languages
XML Schema plays a central role in today's family of XML technologies, most notably in the upcoming XQuery family of specifications. This tutorial explains the role of XML Schema and demonstrates a number of real-life problems with XML-based architectures where today's XML technologies provide no immediate and simple solution, such as schema evolution (open and extensible schemas), loosely coupled Web Services, and the management of different schema versions and the associated code. It is shown how existing XML technologies can be employed to tackle these problems, and how XML designers and developers can make sure that their XML architecture is as robust and future-proof as possible.
Tutorial attendees will gain a solid understanding of how XML Schema is located within and interacts with numerous other XML technologies, and how these interactions directly influence the way how XML should be used today.
TA03: Standards-Based Design
Time: 9:00 — 12:30
The overall goal of this tutorial is to make attendees familiar with the current state of standards-oriented design and to improve their skills in this area. It will not spend time on 'selling' the benefits of such an approach, but will instead focus on how designers can more easily attain those benefits in the real world.
The tutorial will be split into four subtopics, each taking up about a quarter of the time available. The subtopics are: creating a development environment for free; the pros and cons of specific CSS design techniques; recent advances that improve standards support and counter CSS limitations; and current trends in standards-oriented design. The session will be interactive, with audience questions and observations very much encouraged.
TP03: Speech enabling the Web using ECMA/ISO standard (ECMA-269/ISO 18051)
Computer Supported Telecommunications Applications (CSTA) is an ECMA/ISO standard that governs the interoperability of telecommunication devices, ranging from mobile phones, video conference equipments to local or long distance telephone switches. Historically rooted in facilitating human-to-human communications, CSTA has evolved to cover Web-based human-to-computer interactions as more and more telecommunication services are offered through automated software agents.
This tutorial will describe in detail the general concepts of CSTA and its speech service aspects that include speech recognition, speech synthesis, speaker verification and identification.
The tutorial will demonstrate how a Web application can be speech enabled through XML programming in a Web Service or a Voice over Internet Protocol (VoIP) environment, objected oriented programming using a Common Language Infrastructure (CLI, ECMA-335) compliant programming language (e.g., C#, Java, ECMAScript), and markup language programming using speech extensions with HTML.
TA04: RDF and Topic Maps Interoperability in Practice
This tutorial will be done in Japanese.
RDF is a key component of the "Semantic Web" by the World Wide Web consortium. Topic Maps is an ISO standard that has much in common with RDF and is expected to play a major role in the wider (lower-case) "semantic web". Because of the similarities and overlap between RDF and Topic Maps, many people have seen them as being competing "standards". The authors of this tutorial take a different position. They view Topic Maps and RDF as being complementary and have for several years advocated looking for synergies.
During the last year they have developed and implemented an approach to interoperability between RDF data and Topic Maps data that really demonstrates such synergies in practice and has been utilized in Real World applications. This tutorial will explain and justify the approach that has been taken and teach how to apply it in practice. It is hoped that this will enable the community to move beyond the current tendency to discuss which of the two is "best", to a situation where a fuller understanding of their differences leads to synergies on both sides.
TP04: MDA Standards for Ontology Development
Dragan Djurić, University of Belgrade, Serbia and Montenegro
Vladan Devedžić, University of Belgrade, Serbia and Montenegro
The Semantic Web is the main direction of the future Web development. Domain ontologies are the most important part of Semantic Web applications. Artificial intelligence techniques are used for ontology creation, but those techniques are more related to research laboratories. Recently, there are many proposals to use software engineering techniques, especially the UML since it is the most accepted software engineering standard, in order to bring ontology development process closer to wider practitioners' population. However, UML is based on object oriented paradigm, and has some limitation regarding ontology development. These limitations can be overcome using UML'fs extensions (i.e. UML profiles), as well as other OMG's standards (i.e. Model Driven Architecture — MDA). Currently, there is an initiative (i.e. RFP) within the OMG aiming to define a suitable language for modeling Semantic Web ontology languages in the context of the MDA.
The main goal of this tutorial is to present comprehensive introduction into MDA-based ontology development. It will provide an introduction to the field of the Semantic Web and ontology engineering, a description of several UML- and metamodeling- based solutions and tools for ontology development, an overview of the OMG's MDA effort and related standards (Meta-Object Facility — MOF, UML, XML Metadata Interchange — XMI), a detail overview of the OMG's proposal for Ontology Definition Metamodel (http://ontology.omg.org). Finally, we will describe our experiences in developing and employing an MDA-based infrastructure for ontology engineering we defined using the OMG's recommendations.
TA05: Introduction to RDF Query with SPARQL
Steve Harris, University of Southampton
Eric Prud'hommeaux, W3C
Andy Seaborne, Hewlett-Packard Laboratories, Bristol
Time: 9:00 — 12:30
SPARQL is the query language and protocol for RDF being designed by the W3C. Around May 2005 the plan is that the work will be in its final stages (at Last Call stage) and that several compatible implementations will be shipping products supporting it.
The purposes of this tutorial are to introduce SPARQL, to explain its benefits for querying RDF over other approaches to enable easy access to manipulating RDF data.
We will demonstrate how SPARQL can be used to significantly simplify the development of semantic web applications enabling easy reuse of existing RDF data as well as building new RDF data services.
The tutorial is divided into two sections. In the first section, we give a overview of SPARQL's key features in accessing RDF, constraining it and producing result formats. By the end of the first section, the attendees should be able to write simple queries for extracting RDF data. In the second section, we propose to focus on applying SPARQL in the development of a small Semantic Web application, taking RDF from a number of data services, matching and transforming and using it to generate a variety of outputs including more RDF, XML and via XML transformations, HTML.
The tutorial will be delivered as slides with demonstrations of SPARQL queries done using web forms in the browser and on the command line. These will talk to servers providing SPARQL query over example data. Some software API work may be shown in Python, Java and possibly C depending on available of implementations in May 2005.
This tutorial will be presented at a level accessible to Web programmers, advanced developers and experienced students.
TP05: Web-based Interactive Collaboration using Semantic Web Technology — Introduction of the Annotea and its Comparison to the Blog —
This tutorial will be done in English. The material will be provided in both English and Japanese.
Recently, the Blog is getting popular in Japan as the most typical Web-based interactive collaboration. But on the other hand, though the Annotea is also the Web-based interactive collaboration system which is using Semantic Web technology, there are very few Japanese materials about Annotea and it is not so popular in Japan.
In this tutorial, I am going to interpret the architecture and mechanism of the Annotea and argue about the strong point and the weak point of the Annotea from the aspect of using Semantic Web technology by comparing the Annotea and other Web-based interactive collaboration systems such as the Blog.
TA06: NETWORKED ARTS — Methods and Tools to create and maintain Virtual Museums —
Please refer the following web page for the details.
In the last years computer graphics, hypermedia and telecommunications were applied in exploitation of museums, art galleries, architecture and other kinds of works of art. Network access to museums and galleries seems to offer both easier access to cultural heritage and new revenue for its preservation and display. Many relevant players both in the Museums and ICT communities invested time and resources creating pilot projects and applications ranging between 3D reconstruction, image based rendering and virtual museums.
We are now in a position to consider if such investments are effectively useful and really increase and diffuse knowledge in the arts, sciences and history and if they satisfy users' requirements. Do virtual museums really provide added value to end-users? Museums, content providers and users are ready and willing new technologies for cultural heritage? The tutorial provides a comprehensive introduction to the state of the art technologies, methods and standards in the field of online cultural content (museums, collections, memory banks).
The tutorial aims to address the needs of new emerging market including new mixed-economy models for exploitation, repackaging and re-use.
TP06: Matching Words and Pictures: Problems, Applications and Progress
Time: 14:00 — 17:30
The development of technology generates huge amounts of non-textual information, such as images. An efficient image annotation and retrieval system is highly desired. Clustering algorithms make it possible to represent visual features of images with finite symbols. Based on this, many statistical models, which analyze correspondence between visual features and words and discover hidden semantics, have been published. These models improve the annotation and retrieval of large image databases. However, image data usually have a large number of dimensions. Traditional clustering algorithms assign equal weights to these dimensions, and become confounded in the process of dealing with these dimensions.
In this tutorial, first, we will present current state of the art and their shortcomings. Second, we will present weighted feature selection algorithm as a solution to the existing problem. For a given cluster, we determine relevant features based on histogram analysis and assign greater weight to relevant features as compared to less relevant features. Third, we will exploit spatial correlation to disambiguate visual features, and spatial relationship will be constructed by spatial association rule mining. Fourth, we will demonstrate various models including current state of the art to link visual tokens with keywords based on the clustering results of K-means algorithm with weighted feature selection and without feature selection, and will evaluate performance using precision, recall and correspondence accuracy using benchmark dataset. Fifth, we will show that weighted feature selection is better than traditional ones for automatic image annotation and retrieval. Finally, we will discuss open problems and future directions in the domain of image and video.
TA07: Web Content Mining
Time: 9:00 — 12:30
Web mining aims to develop a new generation of tools and techniques to effectively extract and/or mine useful information or knowledge from the Web. It consists of Web usage mining, Web structure mining, and Web content mining. Web usage mining refers to the discovery of user access patterns from Web usage logs. Web structure mining tries to discover useful knowledge from the structure of hyperlinks. Web content mining aims to extract/mine useful information or knowledge from web page contents.
In this tutorial, we focus on Web content mining. In the past few years, there was a rapid expansion of activities in the Web content mining area. This tutorial will introduce the main mining tasks/problems and state of-the-art existing techniques for solving these problems. Topics include: data/information extraction, mining the Web to build concept hierarchies or ontology, mining for Web information integration, segmenting Web pages and detecting noise, mining online opinion sources such as reviews and forums, etc. All these tasks and their associated techniques have immediate applications in the real world.
The tutorial will have many examples to help participants to better understand the concepts and techniques, and also to illustrate how they can be deployed in practice to help businesses. All parts of the tutorial will have a mix of research and industry flavor, addressing seminal research concepts and also looking at the technology from an industry point of view. Thus, apart from researchers and graduate students, we particularly encourage practitioners from industry to participate.
TP07: Location-based Services in Mobile Information Systems: Architectures, Description, and Systems
Time: 14:00 — 17:30
With the growing market of sensing and positioning technologies and the growing popularity and availability of mobile communications, location-based information management has become an important problem in mobile computing systems. Furthermore, the computational capabilities in mobile devices, ranging from navigational systems in cars to hand-held devices and cell phones, continue to rise, making mobile devices increasingly accessible. However, significant research efforts to date have been devoted to location management techniques and location-based services in centralized location monitoring systems. Very few have studied the distributed approach to real-time location monitoring. We argue that for mobile applications that need to manage a large and growing number of mobile objects, the centralized approaches do not scale well in terms of server load and network bandwidth, and are vulnerable to single point of failure.
This tutorial presents the necessary concepts, architectures, techniques, and infrastructure to understand Location-based Services in mobile information systems. The tutorial is designed to be self-contained, and gives the essential background for anyone planning to learn about and contribute to the principles and applications of location-based services in mobile commerce and geographical information systems. It guides practitioners by highlighting best practices in location based information monitoring and introduces students and advanced developers to design and engineering issues in building scalable and privacy-aware distributed location based services, including the key trade-offs, as well as the limitations of current approaches. This tutorial is presented at a senior or beginning graduate student level. It is accessible to Web programmers, advanced application developers, and graduate students.