WASP -- Web Agent Support Program
B. Proposal
8. Content
In this section we describe the underlying scientific
problem our project is concerned with, and
we will further detail the method(s) we will employ,
and the expected results. In addition we will explain
the relevance of our contribution for the scientific
community and we will assess our project with respect to
its scientific merits.
a. Scientific Problem
The notion of agents has become a powerful metaphor
for designating quasi-autonomous intelligent processes
that aid the user in managing the complexity
of a variety of tasks, including information retrieval,
activity scheduling and process monitoring.
Our project aims at providing insight in and solutions
for
- modeling the behavior of cooperating agents
- generic means for realizing actual agents in a Web-aware context
- architectural support for programming agent-based systems
In effect, agents have been characterized
as the dominant human-computer interface of the future [Negro].
The aspects of our research as indicated above
address the problems involved in defining and realizing
the potential of the agent metaphor as a
human-computer interface in the distributed information
system domain, in particular the Web.
Modeling agent behavior
The ability to model the behavior of solipsistic
and cooperative agents is a prerequisite for
(a) managing the complexity of defining agent behavior,
and (b) for adapting the behavior of agents to
user preferences or circumstantial conditions.
Realizing agents
We chose for aiming our project at realizing
agents in a Web-aware context
(a) because the Web provides a rich reservoir
of information,
and (b) the Web offers a suitable infrastructure
for defining an actual graphical user interface
for employing agents.
In particular, agents may be integrated with
standard means for accessing the Web by means
of applets.
In particular we wish to provide 'standard' components
for creating agent-based systems
(including support for graphical user interfaces,
access to the Web and cooperation).
Architectural support
The architectural requirements for realizing agents
in a Web-aware context consists of
(a) a high-level distributed language or system for
programming the behavior of agents,
and (b) access to the Web both in terms
of server-side and client-side computation.
The cooperation between agents and other
components of an application requires advanced
means for distributed object creation and interaction.
Discussion
Currently agents are in the focus of interest
of numerous research groups
[Internet,Softbots,Software,Architecture,Survey].
However, one may observe a divergence between
purely formal approaches and purely pragmatical
approaches, that provide an operational realization
of agents lacking a solid foundation.
Although many of the ideas and concepts
underlying agents have been a topic of
research for decades, notably issues
concerning distributed systems and
programming language design, the notion of agents
itself hase become a major research topic
only recently.
A survey of agent theories, architectures
and languages is given in [Survey].
As concerns agent theories, the main focus of
attention lies on a logical analysis
of intentional notions underlying the behavior
of single agents, whereas for agent architectures
issues of communication and cooperation between agents seem
to be more important.
In [Survey], concurrent logic programming languages
are considered to be the ancestors of agent languages.
Of immediate relevance for our own work is the
Oasis programming language described in [Oasis],
which bears a close resemblance to the
distributed logic programming language DLP [DLP].
Recently, commercial agent systems have been developed for the Web.
For example, General Magic has launched its
Telescript Agent System, a system that employs
mobile agents defined in General Magic's Telescript language.
Also in [Internet], a number of agent-like programs
are described.
We are reluctant to qualify these systems as 'agents',
however, since they are defined in a purely operational way.
We find, for complex agents anyway, a more explicit
representation of the intentional aspects desirable.
In a recent issue of the CACM( 37(7), July 1994),
some current trends and applications of agent research
are presented.
From our perspective we consider as important representatives
of ongoing research:
[Software],
that describes the agent communication language
KQML (Knowledge Query and Manipulation Language);
[Architecture], which describes the complexity of of integrating
collections of agents;
and [Softbots], which describes a number of simple agents
for facilitating access to the Web.
b. Method
For each of the levels or facets indicated,
that is modeling, realization
and architecture, we will briefly describe
the way we plan to tackle the problems involved.
Modeling the behavior of cooperative agents
Our approach will be based on the work already
done in the context of DESIRE [Desire].
In effect, DESIRE provides a logic-based
formalism in which the cooperation between
multiple components can be expressed in a declarative,
compositional
way. In particular, DESIRE has been
employed for describing cooperative multi-agent systems
for monitoring processes [Modeling,ModelAge,Cooperation].
Research in this area will be focussed on
defining cooperation in information-retrieval tasks
and on expressing user-preferences and information-needs
in an adequate way.
Other topics of interest include modeling users as agents
and the verification of agent specifications.
Realizing Web-aware agents
The development of Web-aware applications
has been addressed in the DejaVu project [Applications,Music].
The approach taken there was to employ
object-oriented design technology
for realizing a component-based approach
to integrating multimedia in
Web-based hypermedia applications.
The development of generic components
for realizing agent-based systems will
be addressed in the same fashion,
in order to allow for the integration
of adequate multimedia and graphical
user interface support.
Architectural support
In the DejaVu project we have already gained
expertise in the area
of language embedding and integration [DLP,Hush,VE95].
It is our intention to employ the distributed
logic programming language DLP [DLP] as a high-level
vehicle for programming the behavior of agents.
We strive on the one hand for a close integration
of the DESIRE framework and DLP and on the
other hand to extend DLP with the primitives needed
to access the Web and to create adequate graphical
and multimedia user interfaces.
Also, we have gained some expertise in
supporting cooperative work [Chatting].
This work needs to be extended to allow for
agents to join a session in which they may
communicate and cooperate.
In addition, to allow for managing
the information retrieval and management tasks
involved in dealing with the Web,
helper applications must be developed which
employ the native HTTP protocol of the Web.
An information retrieval tool [Bra,Fish,Search]
has been developed at the TUE and integrated into a WWW browser.
This tool is also available as a server-side tool.
The latest version uses multiple search-agents operating in parallel.
Current Web developments make it possible to integrate agents into
browsers in a more flexible way, by using applets or plug-ins.
A method for improving the use of the internal structure of documents
has been defined by means of object-oriented databases.
An intelligent retrieval helper application needs to be aware of the
internal structure of documents, and needs to be able to store the
information it finds in a similar structure. A Web-server based on an
object-oriented database [OODB]
has been developed for this purpose.
Discussion
Needless to say that we wish to maintain
a close integration between the efforts involved
in the research concerning the aspects mentioned.
In particular, we intend to develop a prototype agent
application Pamela (Personal Assistent for Managing
Electronic Archives) as a point of reference.
c. Results
The WASP project is aimed to result in
a framework (in its extended meaning) for the
development of agent-based Web-aware applications.
Such a framework includes:
- a methodology for developing agent-based applications, as well as
- a logical foundation for modeling agent behavior; and in addition
- guidelines for realizing actual agent applications, and
- software components that can be used as building blocks, including
- a language for programming agent behavior.
In addition we wish our work to result in a generic
agent application Pamela that demonstrates the
capabilities of our approach.
The intended results, divided over the various aspects
of our project, are summarized below.
Modeling the behavior of cooperative agents
- a formalism for modeling cooperative agents
- logical foundations of agent behavior
- modeling user preferences in information retrieval
- modeling circumstantial awareness and negotiation
Realizing Web-aware agents
- generic agent software components
- user interface components for managing agents
- communication primitives for agent communication
- guidelines for developing agent systems
Architectural support
- an agent-based programming language
- a high-level API for Web-aware applications
- an object-based framework for distributed agent applications
- tools for agent-based information retrieval and management
d. Contribution
Scientifically, our work is of significance for the
areas of
- information and knowledge-based systems,
- user-interface development, and
- object-oriented software engineering.
In particular, because we propose to take
an integrated approach to the problem of providing
agent support for Web-aware applications,
the efforts coming from the different areas
of expertise of the participants
results, hopefully, in a more encompassing
theoretical and practical framework
than would have been possible for each
of the participants individually, or more in general
in a more mono-disciplinary approach.
3. Merits
Clearly our project covers both fundamental
and pragmatic aspects.
It is application-oriented in the sense that
it is aimed at satisfying the conditions
for developing actual agent-based applications.
However, it is fundamental to the extent
that it is based on well-founded
principles of modeling and software engineering.
In the area of information retrieval, moreover,
we hope to come up with a substantially
new approach for the discovery of new (recent) interesting
and relevant information in a large distributed hypermedia
information system such as the Web.
12. Plans for implementation
Participants of the project are
- SE/VU -- Software Engineering (VU)
- AI/VU -- Artificial Intelligence (VU)
- IS/TUE -- Information Systems (TUE)
a. Activities
The first year of the project we plan to have
two OIO's working at a number of the technical issues
concerning the software architecture needed
for realizing Web-aware agents for assisting a Web user
in finding and maintaining information.
One OIO, working in the SE/VU group, will concentrate on the
agent and communication techniques, while another OIO, working
in the IS/TUE group, will concentrate on the information
retrieval and discovery aspects.
In the second year and third year, we plan to have
a post-doc stationed at the AI/VU group to coordinate the two
OIO's and to work on the theoretical issues involved in modeling
the behavior of agents and intelligent information retrieval.
The remainder of the time, after the end of the post-doc's contract,
is devoted to publications and writing and defending the doctoral
theses.
The activities planned over four years, as divided
between the participating groups, are summarized below.
SE/VU
Year 1:
- The extension of DLP to an agent-based programming language
- The definition of high-level components
for developing Web-aware applications
Year 2:
- The definition of user interface components for managing agents
- Exploration of distributed object architectures
Year 3:
- An object-based framework for distributed agent applications
Year 4:
- Completion of Pamela, publications and thesis
AI/VU
Year 2 and 3:
- A modeling framework for cooperative agents
- Logical foundations of agent behavior
- The development of domain-specific generic agent models
- Verification and validation of agent specifications
IS/TUE
Year 1:
- Study of information retrieval tools
- The definition of a suitable data structure for storing found information
Year 2:
- Development of tools for agent-based information management
- Optimization of search strategies for agent-based retrieval engines
Year 3:
- Definition of an object-oriented data structure (and database) for
retrieved information
- Development of search tools for extracting information from the database
- Development of a communication protocol for combining the knowledge
acquired by different agents.
Year 4:
- Completion of Pamela, publications and thesis
Joint work on
- A methodology for developing agent-based systems
- Pamela - a generic agent-based application
b. Personel
Funding is requested for two OIO's for four years,
who will be placed at the SE/VU group and the IS/TUE group,
and one post-doc (for two years)
who will be placed in the
the Artificial Intelligence section of the VU (AI/VU).
c. Environment.
SE/VU
The software engineering group of
the VU involved in the DejaVu project consists of
one coordinator (A. Eliëns) two AIOs, that are working
in the area of time-based hypermedia systems (J.R. van Ossenbruggen)
and the integration of simulation and hypermedia
(S.P.C. Sch\"onhage, who is paid by the USF funded SINS project),
and one programmer (C.T. Visser) who is responsible for
the implementation of DLP and the technical aspects of
distributed object systems.
AI/VU
In the AI Section at the VU the research of a number of members of the
group focusses on multi-agent
systems: Dr. F.M.T. Brazier, drs. P.A.T. van Eck, dr. C.M. Jonker, prof. dr.
J. Treur, dr. L.C. Verbrugge. The topics are: compositional formal
modeling of multi-agent systems, models for specific types of intelligent
agents, cooperation models, information retrieval, human-computer
interaction, reflective agents, applications of multi-agent systems
(e.g., coordination of design projects), formal semantics of multi-agent
systems, verification and validation, defeasible reasoning, task analysis.
Some of these topics are addressed in cooperation with researchers from
other institutes, with whom joint papers have been and are being prepared;
for example, dr. N.R. Jennings (Queen Mary & Westfield), dr H. Johnson,
prof dr P. Johnson (Queen Mary & Westfield), prof. dr V.W. Marek and
dr. M. Truszczynski (University of Kentucky), prof. dr. H. Herre
(University of Leipzig), dr. B.M. Dunin-Keplicz (University of Warsaw).
IS/TUE
The database and hypermedia research group within the Information
Systems section of the TUE consists of one associate professor
(dr.P. De Bra), also the head of the entire Information Systems section,
three UDs (dr. A. Aerts, dr. F. Dignum and dr.ir. G.J. Houben),
one part-time professor (dr. J. Paredaens),
drs. R. Post, ir. J. Hidders, ir. C. Hoskens (NWO) and two
part-time aio's (drs. R. Selj\'ee and drs. L. Matthijssen).
A full-time programmer (P. Lemmens) is available to aid in the
implementation aspects of Pamela.
Dr. De Bra also holds a part-time position at CWI, initiating research
efforts on Databases and Internet.
d. Finances
The ASZ GAK group, Object+
and Rabofacet
have expressed interest in taking part in the project.
So they may be regarded as potential co-financers.
(See the appendix for a brief description.)
>document>
e. Output
We plan to produce the deliverables
concerning the following subjects:
- An agent-based programming language
- User interface components for managing agents
- Logical foundations of (cooperative) agent behavior
- An object-oriented framework for distributed agent applications
- A modeling framework for cooperative agents
- A methodology for developing agent-based systems
- Pamela - a generic agent-based application
mailto:eliens@cs.vu.nl
(
Thu Sep 26 14:04:29 MET DST 1996
)
WASP
WASP -- Web Agent Support Program
9. Societal Usefulness
a. Relevance
Our society becomes more and more dependent
on electronically available information.
Due to the increasing amount and complexity
of the information provided,
intelligent agents seem to become the dominant
user interface metaphor for the average user
of information systems in general,
and Web-based applications in particular.
b. Contribution
Our project aims at satisfying the conditions
for developing agent-based Web-aware applications,
which allow the average user to concentrate
on the task at hand rather than the tools
and mechanisms needed to access information.
c. Transfer
The transfer of our expertise will be through
education at our institutes and via contacts
with business and industry.
As interested contacts we mention:
ASZ (GAK group, S.C. Chang, Manager R&D)
Object+ (Peter de Weerdt, Commercial Manager),
Rabo Nederland (Mieke Cornelisse, Rabofacet).
These parties have showed interest in the WASP
project in informal contacts.
mailto:eliens@cs.vu.nl
(
Thu Sep 26 14:04:29 MET DST 1996
)
WASP
WASP -- Web Agent Support Program
10. Collaboration of Subdisciplines in Computer science
This project is a collaboration between the
Software Engineering section of the VU
(SE/VU), the Artificial Intelligence section of the VU (AI/VU)
and the Information Systems section of the TUE (IS/TUE).
All three participants are member of the SIKS school
of research.
As member of SIKS,
the participating sections share common research interests.
The WASP project provides an opportunity
for us to collaborate on a theme
of mutual interest and allows us to profit
from the wide range of non-overlapping expertise available.
Apart from pure computer science the WASP project involves research in
human-computer interaction, in which a lot of expertise is available
at the IPO institute at the TUE.
The selection of relevant information is also related to the field of
Library and Information Sciences. Dr. P.M.E. De Bra is a part-time
professor in the Library and Information Sciences group at the University
of Antwerp, and can draw expertise from that group into the project as
needed.
mailto:eliens@cs.vu.nl
(
Thu Sep 26 14:04:30 MET DST 1996
)
WASP
WASP -- Web Agent Support Program
11. National and International Initiatives or Activities
SE/VU
- Member of ERCIM W4G group
- Workshop Chair: Programming the Web -- a search for APIs,
Fifth Int. World Wide Web Conference
- Contacts with W3C/INRIA -- Patrick Duvall
AI/VU
- REVISE (Redesign in Knowledge-Based Systems)
This project, funded by NWO-SION and coordinated by VU/AI, supports
cooperation between four research groups in the Netherlands: VU/AI,
University of Amsterdam/SWI, University of Twente/Knowledge-Based
Systems Group and the Energy Research Center ECN. For each of the
groups a Ph.D. student is funded by the project. This project
focusses on the design and development of (knowledge-based) support
systems
for redesign.
- dr. N.R. Jennings, Imperial College, London.
Joint research on modeling techniques for multi-agent systems,
together
with dr. B. Dunin-Keplicz has resulted in three joint publications
on modeling complex systems.
- prof. dr R. Gustafson, Technological University of Ronneby.
The Technical University of Ronneby group has a strong tradition of
cooperation with industry in Sweden, in particular, with respect to
research on Multi-Agent Systems supporting management of energy supply
and
distribution. The VU/AI group is currently part of this cooperation.
- prof.dr. J. Fox, Imperial Cancer Research Center, London.
Together with this group safety critical applications of
Knowledge-Based
Systems and Multi-Agent Systems are studied. Each of the two
particpating
groups is supported for this project by national funds.
- dr. H. Johnson and prof dr P. Johnson, Queen Mary and Westfield
College,
University of London.
The merits and weaknesses of task analysis approaches within HCI and
AI
have been compared and analysed in close cooperation with these
experts
in the field of HCI. Current joint research focusses on the analysis
of complex tasks and the role of reflection.
IS/TUE
- Workshop Chair: Object Based Technologies for WWW Services delivery,
Fifth Int. World Wide Web Conference
- Lecturer for the tutorial on Robots at the
Fifth Int. World Wide Web Conference
- Contacts with W3C/INRIA/ERCIM through a part-time position in the
World Wide Web subgroup of the database group at CWI.
- Member of an INCO/Copernicus initiative on World Wide Web caching,
with City University of London and IQ Soft of Hungary.
- Member of the Intide Tourism for All European initiative for making
tourist information (for disabled) available through the Web.
mailto:eliens@cs.vu.nl
(
Thu Sep 26 14:04:30 MET DST 1996
)
WASP
WASP -- Web Agent Support Program
13. Available Expertise
SE/VU
AI/VU
IS/TUE
- Web-based information retrieval and management [Fish,Dress]
- Web-based authoring and serving [OODB]
- Hypermedia data modeling and browsing [Tower,Browsing]
As concerns the various aspects of the WASP project,
our expertise encompasses
wrt. modeling the behavior of cooperative agents:
object-orientation (SE/VU) and
logical foundations and modeling (AI/VU);
wrt. the realization of Web-aware agents:
the integration of applications and the Web (SE/VU),
knowledge-based system development (AI/VU) and
client- and server-side information retrieval and search (IS/TUE);
and for developing architectural support:
distributed logic programming and object technology (SE/VU)
and tools for Web-based information management (IS/TUE).
mailto:eliens@cs.vu.nl
(
Thu Sep 26 14:04:31 MET DST 1996
)