Tutorials Program


Session Schedule:

SAC 2004 offers the following tutorials on Sunday March 14, 2004
T4, all other tutorials are half-day tutorials.

9:00AM - 1:00PM
2:00PM - 6:00PM
10:00PM - 5:00PM (Full Day)

T1: Developing Enterprise Web Services and Applications.

By: Sandeep Chatterjee
Chief Technology Officer
FoundationalNet, Inc.,

T2: Critical Systems Development with UML: Methods and Tools.

This Tutorial has been canceled.

T4: Agent-Oriented Software Engineering.

This Tutorial has been canceled.

T3: Bioinformatics and Machine Learning Methods.

By: Chris Ding
Staff Computer Scientist
Lawrence Berkeley National Lab

T5: Data Mining and Agent Technology: Tools and techniques for dynamic infusion of intelligence.

By: Andreas Symeonidis and Pericles A. Mitkas
Department of Electrical and Computer Engineering,
Aristotle University of Thessaloniki


T6: Mobile Commerce Basics and Techniques

This Tutorial has been canceled.




Registration Fee:

The Tutorial registration fees and deadlines are as follows.
You can add Tutorials to your online registration from the "Additional Items" tab of your registration record.

Registration Type
Half-Day Fee
Full-Day Fee
Member of ACM or SIGAPP
By 2/8/2004
Member of ACM or SIGAPP.
Proof of valid membership is required at Check-In.
By 2/8/2004
Member of ACM or SIGAPP
After 2/8/2004
Member of ACM or SIGAPP.
Proof of valid membership is required at Check-In.
Attendee - Non-Member
After 2/8/2004
Any Time
Full-time student. Student ID is required at Check-In.





Tutorials Abstracts:

T1: Developing Enterprise Web Services and Applications (half-day)

Sandeep Chatterjee
Chief Technology Officer
FoundationalNet, Inc.
2008 Virginia Street
Berkeley, CA 94707

The next stage in the evolution of enterprise applications will be based on Web services. Web services are pieces of application functionality that are
exported through a set of standard application programming interfaces (APIs), and allow applications to be constructed by locating and binding to the exported
functionality. More interestingly, multiple Web services can be coordinated together in unique combinations in an Internet application to implement value-added
services for users. In this tutorial, we describe the design, development, deployment, and maintenance of Internet applications based on Web services. We also
describe the emerging mobile Internet environment, the unique issues inherent to these environments, and the challenges in developing mobile applications based
on loosely coupled Web services. In addition to a broad coverage of the fundamental topics, industry standards, and technologies (e.g., Java, J2EE, application
servers, XML, SOAP, WSDL, UDDI) underlying the development of Web services and applications based on Web services, the tutorial will provide practical,
step-by-step instruction for the development and deployment of enterprise-class Web services and applications based on standard Java 2 Enterprise
Edition (J2EE) application servers and SOAP servers. We also touch on .Net technologies in support of enterprise Web services.


               I.      Introduction

a.     Goals of the tutorial

b.     Agenda

             II.      Application Development for the Internet and corporate intranets

a.     What are the requirements and issues in developing applications for the Web, the general Internet, and for corporate intranets.

            III.      Platform Independent Software Development with Java

a.     Increasing time-to-market pressures coupled with customer demands for personalization and customization makes a “write once, run anywhere” technology such as Java attractive.

b.     Java technology allows programmers to focus on the unique functionality of their application, without having to waste time thinking about low-level portability and compatibility issues.

c.      More efficient Java compilers and runtime environments in conjunction with faster and faster hardware platforms makes Java a very competitive choice for application development.

         IV.      Scalable, Reliable Enterprise-class Application Development with J2EE Application Servers

a.     J2EE application servers provide a high-level abstraction for programmers to build scalable, fault-tolerant, enterprise-class applications.

b.     J2EE application servers implement “contractual agreements” between the application and the supporting server. As long as the “terms of the contract” are adhered to by the application (e.g., through the container API), the application, as viewed by a client, is guaranteed to exhibit certain characteristics, e.g., fault tolerance.

           V.      XML As A Vehicle For Information Exchange

a.     XML is a simple and well-defined technology for the exchange of self-describing information.

         VI.      Web Services

a.     Web services are pieces of application code that are exported through a well-defined API.

b.     Web services are fundamentally different from purchased intellectual property (IP) cores and software objects, such as Java classes. Since Web services are owned and maintained by third parties, traditional application development techniques based on integrating software objects is inadequate.

        VII.      SOAP as a Lightweight Protocol for Web Services-based Application Development

a.     Applications are based on the composition and coordination of Web services that are dynamically discovered and bound to at runtime.

b.     Other technologies for distributed computing, such as CORBA and Java RMI, are based on “tight coupling” that requires a strict understanding between the various components. These architectures have emerged to be too restrictive for ubiquitous computing needs.

c.      SOAP builds upon XML and provides a simple, lightweight mechanism for the communication and exchange of messages between distributed components.

      VIII.      Example: Conference Rom Scheduling and Concierge Services Application

Biographical sketch of the presenter:
Dr. Sandeep Chatterjee is a seasoned technology expert and business professional with over a decade of contributions as a thought leader, technologist,
consultant, entrepreneur, and author. Dr. Chatterjee is chief technology officer of a Web services delivery and management startup, where he is responsible
for the development and strategic positioning of the company's flagship enterprise Web services runtime platform. He also serves as chief technology
consultant for Fortune-100 and major not-for-profit organizations including Hewlett-Packard and ACCION International.

Dr. Chatterjee is co-author of "Developing Enterprise Web Services: An Architect's Guide", a book by Prentice Hall, and has served on the Expert Group
that specified the worldwide standard for mobile Web services. He also sits on the Board of companies developing mobile and Web services technologies,
including Clickmarks, LeadIron Technologies, and Foundationalnet.

Previously, Dr. Chatterjee was the lead and chief architect of Hewlett-Packard's Web Services Mediation Platform. He was also Entrepreneur-in-Residence
at FidelityCAPITAL, the venture capital arm of Fidelity Investments, and was Founder & Chief Technology Officer of Satora Networks, which developed
tools and technologies for developing appliances and services for the mobile and pervasive Internet.

Chatterjee holds a Ph.D. in Computer Science from the Massachusetts Institute of Technology, where his research in networked client architectures and
systems was selected as one of the top thirty-five inventions in the thirty-five year history of MIT's Laboratory for Computer Science, and his invention is
showcased in a time capsule at the Museum of Science in Boston, Massachusetts.

More information about Dr. Chatterjee can be found at: http://www.csg.lcs.mit.edu/~sandeep


T3: Bioinformatics and Machine Learning Methods

Chris Ding
Staff Computer Scientist
Lawrence Berkeley National Lab

The fast evolving trends in bioinformatics and computational genomics are to use machine learning methods to computationally determine functions, structures,
interactions, among DNAs and proteins with biological significance. For example, using classification methods, one can predict protein 3D structures, RNA
coding regions, binding /non-bind active sites, etc. In this tutorial, I will cover several areas where machine learning methods are most widely and fruitfully adopted.


(1) Genomic Basics: DNA, Proteins, annotation, sequence alignment, hidden Markov models
(2) Function and Structure predictions using classification methods. This is the most widely adopted in current research and applications. 
Protein fold recognition. RNA coding region detection. Binding site detection. Cancer type detection.
(3) Feature extraction and selection. This is necessary for effective application of classification methods. More important, most relevant 
features often have direct biological significance regarding to the biological system in question. Filter methods. Wrapper methods.
(4) Unsupervised class, feature, phenotype discovery through data clustering. This supplements biological expertise and help to achieve new 
and deeper insights.
(5) Biological networks: gene regulation networks, protein interaction networks. The fastest growing research area. Network structure  
Bayesian networks. Linear models.  Multi-protein complexes identification. Connectivity analysis and clustering.

Biographical sketch of the presenter:
Chris Ding is a staff computer scientist at Lawrence Berkeley National Laboratory. He obtained a Ph.D. from Columbia University and worked previously in
California Institute of Technology and Jet Propulsion Laboratory. He started work on biomolecule simulations in 1992 and computational genomics research in
1998. He is the first to use Support Vector Machines for protein 3D structure prediction. He's written 8 papers on computational biology, and also published
extensively on data mining, text and Web link analysis. He's given invited seminars at Stanford and UC Berkeley, and many conferences, workshops and panel
discussions. The tutorial grows out of his direct research experiences in the area.


T5: Data Mining and Agent Technology: Tools and techniques for dynamic infusion of intelligence (half-day)

Andreas L. Symeonidis and Pericles A. Mitkas
Department of Electrical and Computer Engineering
Aristotle University of Thessaloniki
54124 Thessaloniki, Greece
E-mail: asymeon@iti.gr, mitkas@auth.gr

The tutorial will present methods and tools that a software developer may use to build applications with intelligent agents. The agent intelligence can be extracted by
performing data mining on historical data. Our major standing point is that inductive reasoning (data mining and knowledge extraction) constitutes a powerful means
for enhancing intelligent agent systems. The tutorial will provide an overview of the most popular data mining techniques and delineate their added value for the
development of Multi-Agent Systems (MAS) with domain-specific knowledge. The same techniques can be used on agent behavior data to retrain the already
deployed agents. This tutorial reviews a number of approaches to the problem and indicates promising solutions. One integrated approach is discussed in more
detail, both from the data mining as well as the intelligent agent perspective. A number of test cases will be also presented. .


1. Basic primitives of data mining technology

A brief introduction on data mining, outlining the key points of the technology and presenting the techniques that can be used in order to induce
logic into
agent systems.

2. Data mining and semantics In order to incorporate extracted data mining knowledge into agents, semantical and domain knowledge considerations have to be taken into account. These considerations are also discussed within the context of the tutorial.
3. Embedding domain knowledge into data mining algorithms

New, agent-oriented, data mining algorithms are presented. These algorithms attempt to extract knowledge from agent behaviors.

4. Software agents

A brief introduction on software agents and agent-based systems

5. Agent Intelligence infusion: tools and methodologies for embedding intelligence into agents, certain research initiatives that attempt to take advantage of the benefits of inductive logic are presented. Detailed descriptions of some of them are given.
6. MAS exploiting data mining extracted intelligence

Four prototypes of multi-agent systems that employ agents with data mining extracted intelligence will be presented:

a. An ERP add-on, providing intelligent policy recommendations on customer and supplier management.
b. An intelligent environmental monitoring system.
c. A personalized, decentralized maintenance management system.
d. An agent-based, e-auction system.

7. Open issues
Finally, the open issues for the coupling of the two technologies (data mining and intelligent agents) are identified and discussed.

Biographical sketch of the presenter:
Mr. Andreas L. Symeonidis is an Electrical and Computer Engineer, currently working towards his Ph.D. on Data Mining and Intelligent Agents, at the Aristotle
University of Thessaloniki, Greece. He is also a Research Associate at the Informatics and Telematics Institute in Thessaloniki, Greece. His area of expertise is data
mining and the exploitation of domain knowledge in intelligent agent applications.

Dr Pericles A. Mitkas is an Associate Professor with the Department of Electrical and Computer engineering at the Aristotle University of Thessaloniki and Deputy
Director of the Informatics and Telematics Institute in Thessaloniki, Greece. His research interests include parallel architectures for large multimedia databases,
data mining and data warehousing, intelligent information agents, and bioinformatics. Between 1990 and 1999 he was an Assistant and Associate Professor with the
Department of Electrical and Computer Engineering at Colorado State University where he performed research on parallel storage and processing systems for large
databases. Dr. Mitkas' work has been reported in over 100 journal and conference publications, several invited lectures and five book chapters. He has also
organized four conferences. DR Mitkas is a senior member of IEEE and the Computer Society.

Both Prof. Mitkas and Mr. Symeonidis have been heavily involved in Agent Academy, an IST-funded project for developing a Data Mining framework for
Training Intelligent Agents (IST-2000-31050).