Unicom Seminar on Implementing Agents March 23, 1999

Software Agents and

Agent Communication
Languages
Tim Finin tell
ask-all
University of Maryland recommend
register
subscribe advertise
Baltimore County
[email protected]
March 23, 1999

http://umbc.edu/~finin/talks/unicom99
1

Overview and objectives

•The agent paradigm -- what is it and what’s

important about it (IMHO)
•The centrality of communication for agents
•Some historical context and future directions
•Basic concepts of agent communication and their
realization in FIPA
•Some examples of applications

Copyright 1999 Tim Finin 1

Unicom Seminar on Implementing Agents March 23, 1999

What is a software agent?

• No consensus yet, but several key
concepts are important to this emerging
paradigm.
•A software agent is a program that assists
people and acts on their behalf. People
can delegate tasks and work to agents.
An agent:
– is an autonomous, goal-directed process
– is situated in, is aware of, reacts to and adapts to its
environment
– cooperates with other agents (software or human) to
accomplish its tasks
3

Agent Characteristic:
Mobility?
A mobile agent is an executing program that
migrates from machine to machine in a
heterogeneous network under its own control.

• Examples: programs in Telescript, Agent-Tcl, Voyager, etc.

and, to a limited degree, Java Applets.
• Note -- this definition implies some agent attributes, e.g.
autonomy, persistence, ...
• Mobile agents offer some very interesting advantages as well as
some disadvantages.
• This is an important technology for distributed systems but is
largely orthogonal to other “agent” issues.
4

Copyright 1999 Tim Finin 2

Unicom Seminar on Implementing Agents March 23, 1999

Agent Characteristic:
Intelligence?
Q: What makes an agent an “intelligent agent”?
A: The size of the price tag.
More seriously…
– The paradigm covers agents of varying degrees of intelligence
– Intelligent agents will tend to
• know and apply more sophisticated domain knowledge
• recognizing underlying goals and intentions
• react to unexpected situations in a robust manner
• better NLP skills
• etc.
Much of what we will be saying applies to agents of little or no
intelligence.

Some key ideas

•Software agents offer a new paradigm for very large
scale distributed heterogeneous applications.
•The paradigm focuses on the interactions of
autonomous, cooperating processes which can adapt
to humans and other agents.
•Mobility is an orthogonal characteristic which many,
but not all, consider important.
•Intelligence is always a desirable characteristic but is
not strictly required by the paradigm.
•The paradigm is still forming.

Copyright 1999 Tim Finin 3

Unicom Seminar on Implementing Agents March 23, 1999

Why is communication important?

•Most, but not all, would agree that communication is
a requirement for cooperation.
•Societies can do things that no individual (agent)
can
•Diversity introduces heterogeneity.
•Autonomy encourages disregard for other agents’
internal structure -- communicating agents need only
care about understanding a “common language”.
For social agents (and animals), the communicative
actions are are among the most important actions to
master. 7

Agent Communication
• Agent-to-agent communication is key to realizing the
potential of the agent paradigm, just as the development of
human language was key to the development of human
intelligence and societies.
• Agents use an Agent Communication Language or ACL to
communication information and knowledge.
– Genesereth (CACM, 1992) defined a software agent as any system
which uses an ACL to exchange information.
• Understanding a “common language” means:
– understanding its vocabulary, i.e., the meaning of its tokens
– knowing how to effectively use the vocabulary to perform tasks,
achieve goals, effect one’s environment, etc.
• For ACLs we’re primarily concerned with the vocabulary
8

Copyright 1999 Tim Finin 4

Unicom Seminar on Implementing Agents March 23, 1999

Some ACLs
•Is CORBA an ACL?
•Knowledge sharing approach
– KQML, KIF, Ontologies
•FIPA Shared
•Ad hock languages ? experiences
and strategies
– e.g., SGI’s OAA
e.g., ?
Intentional Shared beliefs, plans, goals,
Sharing
and intentions
e.g., KQML, FIPA,
KIF, Aglets
Knowledge Shared facts, rules, constraints,
Sharing
procedures and knowledge

Object
e.g., CORBA, Sharing
Shared objects, procedure calls
RPC, RMI and data structures

Agent Communication,
at the technical level
•Messages are transported using some lower-level
transport protocol (SMTP,TCP/IP, HTTP, IIOP, etc.)
•An Agent Communication Language (ACL) defines
the types of messages (and their meaning) that agents
may exchange.
•Over time, agents engage in “conversations.” Such
interaction protocols (negotiation, auction, etc.),
defines task-oriented, shared sequences of messages.
•Some higher-level conceptualization of an agent’s
goals and strategies drives the agent’s communicative
(and non-communicative) behavior.
10

Copyright 1999 Tim Finin 5

Unicom Seminar on Implementing Agents March 23, 1999

Agent Communication,
at the theoretical level
• ACL have message types that
are usually modeled after B + D => I B + D => I
I => A
I => A
speech acts, which are
understood in terms of an
intentional-level description of
an agent
• An intentional description makes references to beliefs,
desires, intentions (BDI) and other mental states.
• BDI frameworks have the power to describe an agents’
behavior, including communicative behavior
• Describing behavior at this level is an important contribution
of the agent-based approach.
11

The intentional stance

•Agents have “propositional attitudes”
•Propositional attitudes are three-part relationship
between
– an agent,
– a content-bearing proposition (e.g., “it is
raining”), and
–a finite set of propositional attitudes an agent
might have w.r.t. the proposition (e.g., believing,
asserting, fearing, wondering, hoping, etc.)
•<a, fear, raining(tnow) >

12

Copyright 1999 Tim Finin 6

Unicom Seminar on Implementing Agents March 23, 1999

On ascribing mental
qualities to machines
•The issue is not whether a system is really intentional
but whether we can coherently view it as such (Daniel
Dennett)
•Ascribing mental qualities to machines (John
McCarthy):
– legitimacy: the ascription expresses the same information about a machine
that it expresses about a person
– usefulness: the ascription helps us understand the structure of the machine,
its past or future behavior, or how to repair it or improve it.
•As MAS get more complex, we will find it useful to
ascribe mental qualities to them, just as we do for other
animals.
13

BDI Agents, Theories and

Architectures
•BDI architectures describe the internal state of an
agent by the mental states of beliefs, goals and
intentions
•BDI theories provide a conceptual model of the
knowledge, goals, and commitments of an agent
•BDI agents have some (implicit or explicit)
representations of the corresponding attitudes

14

Copyright 1999 Tim Finin 7

Unicom Seminar on Implementing Agents March 23, 1999

BDI Model and Communication

B + D => I B + D => I
I => A I => A

• Communicative actions attempt to (1) inform others about our BDI

state, and (2) effect the BDI state of others.
• Note the recursion: an agent has beliefs about the world, beliefs about
other agents, beliefs about the beliefs of other agents, beliefs about the
beliefs another agent has about it, ... 15

Criticism of BDI theories

•The necessity of having all three modalities is
questioned from both ends:
–too few
–too many
•System builders question their relevance in practice:
–multi-modal BDI logics do not have complete
axiomatizations
–they are not efficiently computable
There is a gap between theory and practice

16

Copyright 1999 Tim Finin 8

Unicom Seminar on Implementing Agents March 23, 1999

Speech Act Theory

High level framework to account for
human communication, Language as Action
(Austin)
• Speakers do not just utter true or false sentences
• Speakers perform speech acts:
requests, suggestions, promises, threats, etc.
• Every utterance is a speech act
• The intended underlying speech act may need to
be inferred
– “Pass the salt”, “Please pass me the salt”, vs. “Can
you pass the salt” 17

Agents and agencies

•Groups of agents can form a team to cooperate and
act as one super-agent.
•Opening up an agent we may find it useful to
describe its internal architecture as a collection of
sub-agents.
•What’s going on here? Is it agents all the way
down?
•My take -- a group of agents which can be modeled
as having collective “mental states” (e.g., beliefs,
desires, intentions) and can take collective actions
can be usefully described as an agent.
18

Copyright 1999 Tim Finin 9

Unicom Seminar on Implementing Agents March 23, 1999

Dividing up
the problem
19

Historical Note:
Knowledge Sharing Effort
•Initiated by DARPA circa 1990
•Sponsored by DARPA, NSF, AFOSR, etc.
•Participation by dozens of researchers in academia
and industry.
•Developing techniques, methodologies and
software tools for knowledge sharing and
knowledge reuse.
•Sharing and reuse can occur at design,
implementation or execution time.

20

Copyright 1999 Tim Finin 10

Unicom Seminar on Implementing Agents March 23, 1999

Knowledge Sharing Effort

•Knowledge sharing requires a communication
which requires a common language
•We can divide a language into syntax,
semantics, and pragmatics
•Some existing components that can be used
independently or together:
–KIF - knowledge interchange
format (syntax) Propositional
–Ontolingua - a language for defining
sharable ontologies (semantics)
–KQML - a high-level interaction Propositional
language (pragmatics) attitudes
21

Knowledge Sharing Effort => FIPA

•Knowledge sharing requires a communication
which requires a common language
•We can divide a language into syntax,
semantics, and pragmatics
•Some existing components that can be used
independently or together:
–KIF - knowledge interchange => SL
format (syntax) Propositional
–Ontolingua - a language for defining
sharable ontologies (semantics)
–KQML - a high-level interaction => ACL Propositional
language (pragmatics) attitudes
22

Copyright 1999 Tim Finin 11

Unicom Seminar on Implementing Agents March 23, 1999

Knowledge Interchange Format

Sys 3 Know. Base
in
Know. Base
Lang3 in KIF

• KIF ~ First order logic KIF <-> Lang3 Translator with

Library
set theory KIF
• An interlingua for encoded KIF <-> Lang1 Translator KIF <-> Lang2 Translator

declarative knowledge Know. Base

in
Know. Base
in
Lang1 Sys 1 Lang2 Sys 2
– Takes translation among n
systems from O(n2) to O(n)
• Common language for reusable knowledge
– Implementation independent semantics
– Highly expressive - can represent knowledge in typical application KBs.
– Translatable - into and out of typical application languages
– Human readable - good for publishing reference models and ontologies.
• Current specification at http://logic.stanford.edu/
• FIPA’s SL ≈KIF + modal operators - default reasoning
23

KIF Syntax and Semantics

• Extended version of first order predicate logic
• Model-theoretic semantics
• Simple list-based linear ASCII syntax, e.g.,
(forall ?X (=> (P ?X) (OR (Q ?X) (R ?X)))
(exisits ?person (mother mary ?person))
(=> (apple ?x) (red ?x))
(<<= (father ?x ?y) (and (child ?x ?y) (male ?x))
• KIF includes an axiomatic specification of large
function and relation vocabulary and a vocabulary for
numbers, sets, and lists

24

Copyright 1999 Tim Finin 12

Unicom Seminar on Implementing Agents March 23, 1999

Common Semantics
Shared Ontologies and Ontolingua
• Ontology: A common vocabulary and agreed upon
meanings to describe a subject domain.
• Ontolingua is a language for building, publishing, and
sharing ontologies.
– A web-based interface to a browser/editor server.
– Ontologies can be automatically translated into
other content languages, including KIF, SL,
LOOM, Prolog, etc.
– The language includes primitives for combining
ontologies.
25

Common Pragmatics
Knowledge Query and Manipulation Language
• KQML is a high-level, message-oriented, communication
language and protocol for information exchange
independent of content syntax and ontology.
• KQML is also independent of
– transport mechanism, e.g., tcp/ip, email, corba, IIOP, ...
– High level protocols, e.g., Contract Net, Auctions, …
• Each KQML message represents a single speech act (e.g.,
ask, tell, achieve, … ) with an associated semantics and
protocol.
• KQML includes primitive message types of particular
interest to building interesting agent architectures (e.g., for
mediators, sharing intentions, etc.)
26

Copyright 1999 Tim Finin 13

Unicom Seminar on Implementing Agents March 23, 1999

KQML
Knowledge Query and Manipulation Language
• KQML is a high-level, message-oriented, communication
language and protocol for information exchange independent
of content syntax and ontology.
• KQML is independent of
– the transport mechanism (e.g., tcp/ip, email, corba objects, IIOP, etc.)
– Independent of content language (e.g., KIF, SQL, STEP, Prolog, etc.)
– Independent of the ontology assumed by the content.
• KQML includes primitive message types of particular interest
to building interesting agent architectures (e.g., for mediators,
sharing intentions, etc.)

27

A KQML Message
Represents a single speech act or performative
ask, tell, reply, subscribe, achieve, monitor, ...
with an associated semantics and protocol
tell( i,j, Biφ) = fp[Bi Biφ∧ ¬ Bi( Bif j Biφ∨ Uif j Bi φ)] ∧ re[Bj Biφ] ...
and a list of attribute/value pairs
:content, :language, :from, :in-reply-to
(tell :sender bhkAgent
:receiver fininBot
performative
:in-reply-to id7.24.97.45391
parameter :ontology ecbk12
value :language Prolog
:content “price(ISBN3294,24.95)”)

28

Copyright 1999 Tim Finin 14

Unicom Seminar on Implementing Agents March 23, 1999

KQML Reserved Parameter Keywords

1997
:sender the actual sender of the performative
:receiver the actual receiver of the performative
:from the origin of the performative in :content when forward is used
:to the final destination of the performative in :content when
forward is used
:in-reply-to the expected label in a response to a previous message (same
as the :reply-with value of the previous message)
:reply-with the expected label in a response to the current message
:language the name of the representation language of the :content
:ontology the name of the ontology (e.g., set of term definitions)
assumed in the :content parameter
:content the information about which the performative expresses an
attitude

29

Performatives (1997)
Insert
Tell
Uninsert Inform Untell
Delete-one
Delete-all DB Basic Broadcast
Undelete Forward
Ask-if Inform
Ask-one Network
Ask-all
Basic Achieve
Unachieve

Stream Stream Query Request Goal

Eos
KQML
Performatives Facilitation
StandbyCursor Broker-one
Ready Recommend-one
Next Reply Recruit-one
Rest Stream
Promise Broker-all
Discard Eos Recommend-all
Advertise
Deny Unadvertise Recruit-all
Meta
Subscribe
30

Copyright 1999 Tim Finin 15

Unicom Seminar on Implementing Agents March 23, 1999

Facilitation Services
Facilitators are a class of agents who
• traffic in meta-knowledge about other agents.
• provide communication services such as:
– message forwarding and broadcasting
– resource discovery
– matchmaking
– content-based routing
– meta-knowledge queries
• Performatives of special interest to facilitators are
– advertise, broker, recruit, recommend, forward, broadcast, etc.
• Brokers are generally considered to focus on matchmaking
• Facilitators can be intelligent or not
– Intelligent facilitators use domain knowledge in matching services needs
and offers.
31

If KQML is so great, why do

we need the FIPA ACL?
• There are two KQML specification documents and many
dialects and “extended” versions of KQML plus many
important concepts not yet addressed (e.g., security).
• The FIPA ACL has been specified along with other critical
aspects of agent systems (e.g., agent management).
• The FIPA ACL has the support of a formal standardization
process and organization.
• FIPA provides us with an opportunity to revisit and
improve on many of the design decisions made in KQML
– e.g., primitive CAs are defined in such a way as to allow them to be
composed, creating new CAs

32

Copyright 1999 Tim Finin 16

Unicom Seminar on Implementing Agents March 23, 1999

The
FIPA
ACL
33

Major Features of FIPA ACL

compared to KQML
•Management and facilitation primitives (register,
broker, recruit, etc.) are not part of the ACL
•Primitives can be defined compositionally from “core”
primitives
•Use of a powerful language to define agents’states
(Semantic Language, or SL)
•Semantics based on mental attitudes (belief, intention,
etc.)
•The meaning of primitives is given in terms of
Feasibility Preconditions (FPs) and Rational Effect (RE)
34

Copyright 1999 Tim Finin 17

Unicom Seminar on Implementing Agents March 23, 1999

Outline of FIPA ACL Semantics

•A primitive’s meaning is defined in terms of FPs

and REs
•The Feasibility Preconditions of a CA define the
conditions that ought to be true before an agent
may plan to execute the CA
•The Rational Effect is the effect that an agent
hopes to bring about by performing an action (but
with no guarantee that the effect will be achieved)
•The FPs and the REs involve agents state
descriptions that are given in SL
35

Semantic Language (SL)

•SL is the formal language used to define the semantics

of FIPA ACL
•In SL, logical propositions are expressed in a logic of
mental attitudes and actions
•The logical framework is a first order modal language
with identity (similar to Cohen & Levesque)
•SL provides formalizations for three primitive mental
attitudes: Belief, Uncertainty and Choice (or Goal);
Intention is defined as a Persistent Goal
•SL can express propositions, objects and actions
36

Copyright 1999 Tim Finin 18

Unicom Seminar on Implementing Agents March 23, 1999

An example of FIPA ACL semantics

(inform)

<i, inform( j, φ)>

FP: Biφ∧ ¬ Bi( Bif j φ∨ Uif jφ)
RE: Bjφ

Agent i informs agent j that (it is true that) it is raining today:

(inform
:sender i
:receiver j
:content "weather(today,raining)"
:language Prolog
:ontology weather42)

37

Another example of FIPA ACL

semantics (request)

<i, request( j, a )>

FP: FP(a) [i\j] ∧ Bi Agent( j, a ) ∧ ¬ Bi Ij Done(a)
RE: Done(a)

Agent i requests j to open a file:

(request
:sender i
:receiver j
:content "open \"db.txt\" for input"
:language vb)

38

Copyright 1999 Tim Finin 19

Unicom Seminar on Implementing Agents March 23, 1999

Composing new primitives

<i, query-if(j, φ) ≡ <i, request(j, <j, inform-if(i, φ)>)>
FP: ¬ Bif iφ∧ ¬ Uif iφ∧ ¬ Bi Ij Done(<j, inform-if(i, φ)>)
RE: Done(<j, inform(i, φ)>|<j, inform(i, ¬ φ)>)
Agent i asks agent j if j is registered with domain server d1:
(query-if
:sender i
:receiver j
:content (registered (server d1) (agent j))
:reply-with r09
…)
Agent j replies that it is not:
(inform
:sender j
:receiver i
:content (not (registered (server d1) (agent j)))
:in-reply-to r09)
39

Ontologies
40

Copyright 1999 Tim Finin 20

Unicom Seminar on Implementing Agents March 23, 1999

Overview
• What is an ontology?
• Tools for building, using and
maintaining ontologies
• Existing ontologies of
general interest
• FIPA's view on agents and
ontologies

41

Common Semantics
Shared Ontologies and Ontolingua

Ontology : A common vocabulary and agreed upon

meanings to describe a subject domain.
On*tol"o*gy (?), n. [Gr. the things which exist (pl.neut. of , , being, p.pr. of
to be) + -logy: cf.F. ontologie.]
That department of the science of metaphysics which investigates
and explains the nature and essential properties and relations of all
beings, as such, or the principles and causes of being.
Webster's Revised Unabridged Dictionary (G & C. Merriam Co., 1913, edited by Noah Porter)

This is not a profoundly new idea …

–Vocabulary specification
–Domain theory
–Conceptual schema (for a data base)
–Class-subclass taxonomy
–Object schema
42

Copyright 1999 Tim Finin 21

Unicom Seminar on Implementing Agents March 23, 1999

Conceptual Schemas
A conceptual schema specifies the intended meaning of
concepts used in a data base
139 74.50
Data Base: 140 77.60
… …

Data Base Schema: Table: price

*stockNo: integer; cost: float
Auto
Product
Ontology
price(x, y) =>
∃(x’, y’) [auto_part(x’) Product
Conceptual Schema: & part_no(x’) = x Ontology
& retail_price(x’, y’, Value-Inc)
& magnitude(y’, US_dollars) = y]

Units &
Measures
Ontology
43

Implicit vs. Explicit Ontologies

•Systems which communicate and work together

must share an ontology.
•The shared ontology can be implicit or explicit.
•Implicit ontology are typically represented only by
procedures
•Explicit ontologies are (ideally) given a declarative
representation in a well defined knowledge
representation language.

44

Copyright 1999 Tim Finin 22

Unicom Seminar on Implementing Agents March 23, 1999

Conceptualizations, Vocabularies
and Axiomitization
•Three important aspects to explicit ontologies
– Conceptualization involves the underlying model of the
domain in terms of objects, attributes and relations.
– Vocabulary involves assigning symbols or terms to refer
to those objects, attributes and relations.
– Axiomitization involves encoding rules and constraints
which capture significant aspects of the domain model.
•Two ontologies may
– be based on different conceptualizations
– be based on the same conceptualization but use different
vocabularies
– differ in how much they attempt to axiomitize the
ontologies 45

Simple examples

fruit fruit

apple lemon orange pomme citron orange

fruit
fruit
apple citrus pear
tropical temperate
lime lemon orange
46

Copyright 1999 Tim Finin 23

Unicom Seminar on Implementing Agents March 23, 1999

Ontologies vs. KBs

Ontologies are distinguished from KBs not by their
form, but by the role they play in representing
knowledge
– Consensus models for a domain
– Emphasis on properties that hold in all situations
– Emphasis on classes rather than instances
– Intended to support multiple tasks and methods
– Don’t change during problem solving and are suited for
“compiling” into tools
– Need to satisfy a community of use
• Emphasis on collaborative development
• Emphasis on translation to multiple logical formalisms
– Useful for education
47

Ontology Library and Editing Tools

Ontolingua is a language
Lexicons & Common Domain-Specific
for building, publishing, and Skeleton Ontologies Ontologies & Theories Ontologies & Theories

sharing ontologies. Models of

Time
Actions
& Causality
Situations
& Contexts
Operations
Logistics
Sensor Management
–A web-based interface to a Shared
Library
WordNet
Penman Ontology
CYC Upper Ontology
Models of
Space
Physical
Objects
Geography
& Terrain
Battlefield Situations
Command and Control
browser/editor server at
Basic Representation Concepts: Sets, Sequences, Arrays, Quantities, Probabilities
http://ontolingua.stanford.edu/
and mirror sites.
–Ontologies can be translated Browse Compare Compose Extend Check
Editing
into a number of content °
Tools
languages, including KIF,
LOOM, Prolog, CLIPS, etc.

48

Copyright 1999 Tim Finin 24

Unicom Seminar on Implementing Agents March 23, 1999

Ontolingua - Language
•Ontolingua allows full KIF
– 1st order logic with relation
constants in domain of discourse
– Extremely expressive
– Too much for most users
– Too much for most systems!
•Ontolingua provides an object-
oriented projection
•Statements within the o-o
sublanguage easy to make
– But any statement is allowed
•Ontolingua separates
representation from
presentation

49

Ontolingua - Library
•Library of modules
supports reuse
•Authors assemble a new
ontology
– Assembly defines a general
graph
– Cycles are allowed (sports and
medicine)
•Authors may augment
definitions
– But you can never say less!
– Different authors may make
incompatible extensions

50

Copyright 1999 Tim Finin 25

Unicom Seminar on Implementing Agents March 23, 1999

Ontolingua - Usage
• Ontolingua is (one of) the most widely used knowledge
development environments
– Available since 1/94 at http://ontolingua.stanford.edu
– Over 4500 total users, 1200 current users, 300 active users
– Over 4,200,000 user commands executed
– Recently averaging over 7000 commands per day
– Over 800 ontologies stored on the KSL server
– Mirror sites in Spain, Netherlands, UMBC, and corporate sites
• Applications include
– Enterprise modeling, electronic commerce, engineering, ribosomal
structure modeling, workflow modeling, molecular biology, cross-
disciplinary design and simulation, drug interactions, medical
vocabularies, software design reuse, standards development

51

Open Knowledge Base Connectivity

• OKBC is another software tool for building and accessing ontologies
– OKBC is to KBs what ODBC is to Databases
– A standard API to Knowledge Representation Systems (KRS)
• OKBC
– Specifies a protocol for KRS interoperation
– Supports a client-server model for interaction
– Provides an object-oriented view of a KRS
– Supports wide variation in underlying KRS
• Adopted as KRS interoperation protocol within DARPA High
Performance Knowledge Base (HPKB) program
• OKBC drivers available for Loom, Ontolingua, Ocelot, ATP, JavaKB,
TupleKB, ...
• OKBC applications include GKB (SRI), Jot (KSL), Ontolingua (KSL),
Riboweb (SMI), Protégé (SMI), Hike (SAIC), ...
• http://ai.sri.com/~okbc/
52

Copyright 1999 Tim Finin 26

Unicom Seminar on Implementing Agents March 23, 1999

Big Ontologies
•There are several large, general ontologies that are
freely available.
•Some examples are:
– Cyc - Original general purpose ontology
– WordNet - a large, on-line lexical reference system
– World Fact Book -- 5Meg of KIF sentences!
– UMLS - NLM’s Unified Medical Language System
– See http://www.cs.utexas.edu/users/mfkb/related.html for
more
•We anticipate the development of ontologies to
support ecommerce
– see www.ontology.org
– probably in XML
53

Ontologies in the FIPA Reference Model

Ontology designer
Non-FIPA
components

Ontology Ontology Ontology

Server Server Server
(Ontolingua) (DB of ODL definitions) ( XML )

OKBC OQL http

Ontology
Ontology servers
Agents Ontology
Agent define and
provide ontology FIPA serve shared
related services ACL Channel components ontologies
(e.g., translation) to agents and
to FIPA agents other systems
OA-2 Agent Agent DF

54

Copyright 1999 Tim Finin 27

Unicom Seminar on Implementing Agents March 23, 1999

Ontology agent services

• Help a FIPA agent in selecting a shared (sub)
ontology for communication,
•Create and update an ontology, or only some
terms of an ontology.
•Translate expressions between different ontologies
(different names with same meanings),
•Respond to query for relationships between terms
or between ontologies,
•Discovery of public ontologies in order to access
them. 55

The FIPA knowledge model

•FIPA specifies fipa-meta-ontology as the ontology
used to talk about ontologies.
•This is largely based on the OKBC model
developed by Stanford and SRI
– See http://ai.sri.com/~okbc
•As its name suggests:
– OKBC is to KBs what ODBC is to Databases
– A standard API to Knowledge Representation Systems (KRS)
•Generic OKBC clients (e.g., browsers, editors)
exist as well as OKBC drivers for a number of
knowledge representation systems.

56

Copyright 1999 Tim Finin 28

Unicom Seminar on Implementing Agents March 23, 1999

Relationships between ontologies

•fipa-meta-ontology also includes:
(ontol-relationship ?O1 ?O2 ?level)
•to describe translation services, where ?level can be:
– Extension - O1 extends the ontology O2
– Identical - O1 and O2 are identical
– Equivalent - O1 and O2 are equivalent
– Strongly-translatable -every term in O1 is translatable to a term in O2
without loss of information
– Weakly-translatable - some terms in O1 are translatable to terms in O2
with some loss of information
– Approx-translatable - translating terms from O1 to O2 may introduce
some inconsistencies
•
57

Ontology Conclusions
• Shared ontologies are essential for agent communication
and knowledge sharing
• Ontology tools and standards are important
– Ontolingua and OKBC are good examples
– XML and RDF may be a next step
• Some large general ontologies are available
– Cyc, WFB, Wordnet, …
• For more information…
– http://www.ontology.org/
– Ontology mailing list: send mail to [email protected] with
“info ontology” in message body for information.
– ANSI Ad Hoc Group on Ontology Standards: http://WWW-
KSL.Stanford.EDU/onto-std/

58

Copyright 1999 Tim Finin 29

Unicom Seminar on Implementing Agents March 23, 1999

Conclusions
and
Prospects
59

Some key ideas

•Software agents offer a new paradigm for very large
scale distributed heterogeneous applications.
•The paradigm focuses on the interactions of
autonomous, cooperating processes which can adapt
to humans and other agents.
•Agent Communication Languages are a key enabling
technology
– Mobility is an orthogonal characteristic which many, but
not all, consider central.
– Intelligence is always a desirable characteristic but is not
strictly required by the paradigm.
•The paradigm is still forming and ACLs will continue
to evolve.
60

Copyright 1999 Tim Finin 30

Unicom Seminar on Implementing Agents March 23, 1999

Prospects
•FIPA’s ACL is likely to be the next iteration of a widely used
standard ACL.
•Its not clear how ACLs will participate in the rapidly evolving
world of Internet languages and protocols
– The ACL “territory” may be overtaken by efforts from a programming
language (e.g., Java, Jini), another interoperability language (e.g.,
CORBA) or Web technology (e.g., XML).
– The Agent community is a small fish compared to, e.g., the Java
community. What will Microsoft do?
•Many are experimenting with XML for agent communication
– XML is a good way to represent structured information (e.g., ACL
messages, KIF-like content) that is easy to use and understand by all
agents, both human and software
– We’ve developed DTDs and style sheets for FIPA ACL and KIF
– XML is not a silver bullet.
61

For More Information

• General information on software agents
– http://www.cs.umbc.edu/agents
• FIPA
– http://www.fipa.org/
• KQML
– http://www.cs.umbc.edu/kqml
• KIF
ask-all – http://www.cs.umbc.edu/kif tell
recommend
subscribe advertise • Ontologies register

– http://www.cs.umbc.edu/ontology/
• Agent Communication Languages
– http://www.cs.umbc.edu/acl/

62

Copyright 1999 Tim Finin 31