ASIAN'07,
12th annual Asian Computing Science Conference, Dec 9 – 11, 2007, Doha,
Qatar.
The ASIAN conference series provides a forum for researchers throughout Asia
to present cutting-edge results in yearly-themed areas of Computer Science,
to discuss advances in these fields, and to interact with researchers from
other continents. The 2007 edition focuses on computer and network security.
Networked computer systems stand at the core of today's information society:
by providing fast access to huge quantities of information, they enable higher
productivity at a much higher convenience than ever in the past. They also
provide a substantial opportunity for abuse. The closely related areas of
Network and Computer Security are concerned with mitigating this risk and
preventing disruption, fraud, disclosure, etc. The former tends to concentrate
on the infrastructure, while the latter deals with everything else. In recent
years, they have made enormous progress both in our theoretical understanding
of what security is and in the development of practical protection mechanisms.
Yet, they are engaged in a continuing arms race with an underground army of
ever more sophisticated attackers intent to bypass whatever shield protects
a system. This makes network and computer security one of the most active
fields in today's scientific landscape.
AINTEC'07,
Asian Internet Engineering Conference, Nov. 27 – 29, 2007, Phuket, Thailand.
The 3rd Asian Internet Engineering Conference (AINTEC) follows the first two
successful editions held in Bangkok, Thailand, and focuses on developing synergies
between researchers in Asia, cooperations at the international level and target
challenges specific to the Asian environment. AINTEC 2007 will be a major
opportunity for presentations and discussions around these objectives. In
particular, it aims at addressing issues pertinent to the Asian region with
vast diversities of socio-economic and networking conditions while inviting
high quality and recent research results from the global international research
community to be presented. The conference is single-track and features a technical
program with significant opportunities for individual and small-group among
a diverse set of participants.
Service Web 3.0, by Dieter Fensel, University
Innsbruck, Austria.
Tuesday, Dec. 18, 2008, 13:30, Milton Bender Auditorium.
Computer science is on the edge of an important new period of abstraction.
A generation ago we learnt to abstract from hardware to software and currently
we learn to abstract from software in terms of Service Oriented Architectures
(SOA). The Semantic Web has opened new perspectives in terms of data and information
management, and aims to lay the foundations of a new generation of semantically-enabled
service infrastructures. This presentation explores the growing field of Serviceware
and the potential of its combination with Semantics to enable the Service
Web. It also discusses the challenges facing the scalability of an open, heterogeneous
and distributed web of semantically enabled services. Techniques like intelligent
information retrieval and approximate reasoning combined with a heuristic
and goal driven approach will define computer science of the 21st century.
Sequential Voting in Multi-Issue Domains, by Jérôme
Lang, CNRS, Toulouse, France.
Monday, Dec. 17, 2007, 11:00, CSIM #209.
In many real-world group decision making problems, the set of alternatives
is a Cartesian product of finite value domains for each of a given set of
variables (or issues). Dealing with such domains leads to the following well-known
dilemma: either ask the voters to vote separately on each issue, which may
lead to so-called multiple election paradoxes as soon as voters' preferences
are not separable; or allow voters to express their full preferences on the
set of all combinations of values, which is practically impossible as soon
as the number of issues and/or the size of the domains are more than a few
units. We try to reconcile both views and find a middle way, by relaxing the
extremely demanding separability restriction into this much more reasonable
one: there exists a linear order x_1 > ... x_p on the set of issues such
that for each voter, every issue x_i is preferentially independent of x_{i+1},...,
x_p given x_1,..., x_{i-1}.
This leads us to define a family of sequential voting rules, defined as the
sequential composition of local voting rules. These rules relate to the setting
of conditional preference networks (CP-nets) recently developed in the Artificial
Intelligence literature. Lastly, we study in detail how these sequential rules
inherit, or do not inherit, the properties of their local components.
Effective
Software Testing, by Chris Nelson and Dr. Akmal Chaudhri,
IBM.
Monday May 14, 2007, 13:00, Milton Bender auditorium.
Shortening the time to market of new software systems without compromising
their quality is a challenge that every company faces. Time-to-market pressures,
building new Service-Oriented Architectures, complying with regulatory requirements,
and working with geographically distributed teams add additional complexities
that can derail software testing projects that aren't ready for these challenges.
To meet these challenges, successful IT organizations are taking a strategic
approach to quality assurance, applying the practices of business governance
to achieve a competitive advantage through better software. Whether you are
planning to build expertise in your IT organization, or looking for partners
with testing expertise, Rational-software quality solutions and best practices
can provide you with the framework and tools needed to use excellence in software
testing as a strategic business advantage.
Building semantic web applications as information/knowledge
sharing systems, by Prof. Hideaki
Takeda, National Institute of Informatics, Tokyo, Japan.
Monday, Feb. 26, 2007, 9:00, CSIM #209.
In this talk, I discuss the methodology to design Semantic Web applications
that can be acceptable widely by ordinary people. I first analyze miracle
of web as an information sharing tool that is basically difficult for people
to accept. In order to overcome this point, Semantic Web applications should
have two types of gratification simultaneously, i.e., instant gratification
that can be obtained even without information/knowledge sharing, and delayed
gratification that
can be obtained through information/knowledge sharing. The gap between two
types of gratification can be bridged by the translucence strategy that lures
people into information/knowledge sharing by showing delayed gratification
within kissing distance. I then show our experience to build information/knowledge
sharing tools with the above methodology. One is Community Navigator that
helps participants for a conference to share knowledge like topics and related
people. The other is Semblog systems that helps weblog people to exhibit and
exchange their information more.
ProCLAIM: An Argument-Based Model for Deliberating Over the Appropriateness
of a Proposed Action, by Pancho Tolchinsky, Technical University
of Catalonia, Spain.
Wednesday, Jan. 31, 2007, 16:00, CSIM #209.
In many domains actions are taken following established guidelines that guarantee
their success and/or safety in the case of safety-critical domains. However,
there are circumstances in which actions that deviate from the guidelines
are justified. In this presentation we describe the argument-based model ProCLAIM
that provides a setting for agents to deliberate over the appropriateness
of a proposed action. Broadly construed, the ProCLAIM model consist of a mediator
agent, MA, directing proponent agents in an argument based collaborative decision
making dialog, in which the final decision (course of action) must comply
with certain domain dependent guidelines. However, the arguments submitted
by the proponent agents may persuade the MA to accept decisions that deviate
from the guidelines. For example, the MA may be able to reason that the submitted
arguments supporting an alternative course of action have
proven to be successful in previous similar deliberations. We believe that
ProCLAIM is of particular value in safety-critical domains (although the scope
of domain may well be wider) where the consequences ensuing from wrong actions
may be catastrophic. Guidelines in such sensitive environments usually exist
and are created in an attempt to
minimize hazardous decisions. Nonetheless, there are circumstances in which
a course of action is appropriate despite violating the basic guidelines.
ProCLAIM defines three main tasks for the MA: 1) Inform the proponent
agents as to what are their dialectical possible moves at each stage of the
deliberation; 2) Ensure that the submitted arguments are relevant (i.e., comply
with the guidelines), and 3) Evaluate the submitted arguments in order to
identify the winning arguments and thus determine whether a proposed action
is appropriate. This last task may require the assignment of strengths to
the given arguments and possibly submission of additional arguments.
The practical value of ProCLAIM is illustrated in a health-care scenario,
in which proponent agents will deliberate over the appropriateness of the
transplantation of an available organ to a potential recipient.
2007 Conferences
