ITTALKS A Case Study in How DAML Helps

1

ITTALKSITTALKSA Case Study in

How DAML Helps

Tim FininTim FininUniversity of Maryland

Baltimore County

Semantic Web for the Military User

June 6, 2001

ask-all

advertisesubscribe

tellrecommend

register

2

Overview

1. ITTALKS web application

2. Two advanced capabilities

3. How does DAML help?

4. What’s needed to build apps?

Joint work with JHU/APL and MIT/Sloan.

See http://umbc.edu/~finin/swmu/ for slides

3

UMBC/JHU/MIT DAML Project

UMBC, JHU, and MIT are working together on a set of issues under funding from DARPA

UMBC (Finin, et. al.) is focused on integrating communicating agents, DAML and the Web

JHU APL (Mayfield, et. al.) is building information indexing and retrieval systems that work with documents and queries that contain a mixture of free text, XML and DAML

MIT Sloan School (Grosof et. al.) is developing techniques for integrating rule based technology and distributed belief into DAML

To be integrated in agent-based applications involving search and using rule-based reasoning.

4

ITTALKS• ITTALKS is a database driven web

site of IT related talks at UMBC andother institutions. The database contains information on– Seminar events– People (speakers, hosts, users, …)– Places (rooms, institutions, …)

• This database is used to dynamically generate web pages and DAML descriptions for the talks and related information and serves as a focal point for agent-based services relating to these talks.

• We are exploring how the semantic web and DAML add value to this web-based application

http://ittalks.org/

1

5

6

Registered users create a profile (encoded in DAML) to describe their preferences and attributes.

7

After logging in, ITTALKS can filter the talks shown based on my interests, schedule and location.

8

9

10

ITTALKS Architecture

Web server + Java servlets

DAMLreasoning

engine

DAMLreasoning

engine

<daml></daml>

<daml></daml>

<daml></daml>

<daml></daml>

DAML files

Agents

Databases

People

RDBMSRDBMSDB

Email, HTML, SMS, WAP

FIPA ACL, KQML, DAML

SQLHTTP, KQML, DAML, Prolog

MapBlast, CiteSeer,Google, …HTTP

HTTP, WebScraping

WebServices

ApacheTomcat

11

ITTALKS Ontologies• We’ve defined and use the following ontologies, all

at http://daml.umbc.edu/ontologies/– calendar-ont.daml – calendar and schedule info– classification.daml – ACM CCS topics– person-ont.daml – people and their attributes– place-ont.daml – talk locations– profile-ont.daml – user modeling info– talk-ont.daml – talks info– topic-ont.daml – topics and interests

12

ITTALKS Features

• Generation from DB to DAML and HTML mediated by MySQL, Java servlets, and JSP.

• Generation of DAML descriptions and user profiles from HTML forms.

• Creation & use of DAML-encoded user models describing interests and ontology extensions.

• Ontologies for events, people, places, schedules, topics, etc.

• Automatic HTML form (pre) filling from DAML.

• Syncing of talks with Palm calendars via Coola.

• Automatic classification of talks into topic ontology

• A XSB-based DAML/RDF reasoning engine.

• ITTALKS agent with KQML API using DAML as content language

• Intelligent matching of people and talks based on interests, locations and schedules.

• Agents using both Jackal and FIPA’s Java Message Service

• Notification via email and mobile devices via SMS and WML.

• Discovery of relevant background papers from NEC CiteSeer

• Automatic generation and maintenance of user models

• Talk recommendations via collaborative filtering

• Integration with STP (Smart Things and Places) ubiquitous computing project

Agent-based featuresGeneral features

13

Two Advanced Capabilities

• I’ll briefly describe two advanced capabilities facilitated by DAML:– Classifying talk topics and user interests

using DAML ontologies– Using DAML as a communication

language among software agents

2

14

Entering talks• Currently, talks manually

entered through a web form interface

• Several things help: (i) recognizing entities, e.g. people) already in the database and (ii) text classification

• Goal: become for research talks what NEC CiteSeer is for research papers– Focused search engine to collect talk announcements in text or

HTML or marked up in a partially understood ontology.– Information extraction using LMCO’s Aerotext to extract

relevant talk parameters and enter into database

15

What are talks about?

• Topic hierarchies provide indexing terms– ACM CCS topic hierarchy

– Open Directory

• Encoded as DAML ontologies

• These allow users to specify interests as well as browse the database of talks by topic

• Automatic classification of talks (based on title and abstract) and users (based on his web pages, CV, papers, etc.)

• Discovery of mapping rules between CCS to OD ontologies using IR techniques

16

Classifying Talks

ACM CCS Ontology

Training corpus

CMU Bow

statisticaltext analysis

tools

CMU Bow

statisticaltext analysis

tools

ACM CCSclassifier

Now is the time for all good men to come to the aid of the country. Now is the time for

topics

e.g.: ACMCCS

e.g.:5K ACMabstracts

Topics Ontology

uses

uses

17

Mapping between topic ontologies

Topic ontology T1

Training corpus T1

CMU Bow

statisticaltext analysis

tools

CMU Bow

statisticaltext analysis

tools

T1T2mapper

{(t2:bar, 0.8), (t2:qux, 0.7), …}

Topic ontology T2

Training corpus T2

T1

T2

t1:foo

18

DAML and Agents• Much multi-agent systems work is grounded in Agent

Communication Languages (e.g., KQML, FIPA) and associated software infrastructure such as the DARPA Grid– The paradigm has been peer-to-peer message oriented

communication mediated by brokers and facilitators.• The DAML program invites different paradigms which will

require some changes in ACLs and their associates software systems.– Agents “publish” beliefs, requests, and other “speech acts”

on web pages.– Agents “discover” what peers have published on the web.

• The software agent research community is very interested in the semantic web and DAML

ask-all

advertisesubscribe

19

ITTALKS Agent

• ITTALKS offers a web interface for its human users and can send notifications to humans via email, WAP and SMS.

• We are also developing an agent API so that software agents can interact with ITTALKS.

• Currently, the ITTALKS agent can send notifications to agents via KQML using DAML as the “content language”.– We will support richer, mixed initiative dialogs between

ITTALKS and agents in the future

20

ITTALKSagent

Travelagent

Calendaragent

Useragent

DAMLreasoning engine

BrokerAgent

AgentNameServer

user’s daml profile

mapquest

user’s calendar app

ITTALKS app

DAML reasonerCommon agent infrastructure

KQML

API

Communicationprotocol

1

8

7 6

5

3

2

4

9

10

11

12

13

17

16

15

14

18

21

How Does DAML Help?

• Does it Help? Yes– We’ve identified five general areas in

which DAML adds value to the application or facilitates building or maintaining the application

• Is DAML needed? No– Not strictly (yet), although the alternative

technologies are not designed for the web and thus suffer from deficiencies.

3

22

How does DAML Help?

ontologylanguage

usermodels

interoplanguage

agentcommunication

webservices

23

ontologylanguage

usermodels

interoplanguage

agentcommunication

webservices

Information in ITTALKS is exposed or published in DAML on the web.

DAML’s descendant will become the “semantic interlingua” for applications and systems

24

ontologylanguage

usermodels

interoplanguage

agentcommunication

webservices

DAML is used • As a DB conceptualschema language

• To help specify APIs• To aid human understanding

25

ontologylanguage

usermodels

interoplanguage

agentcommunication

webservices

DAML is used to encode a common user model that

• Are stored in the user’s file space

• Contain information which can be shared by many applications

• Can contain information specific to certain applications

26

ontologylanguage

usermodels

interoplanguage

agentcommunication

webservices

DAML is used to support agent communication as a “content language” used to encode the content of a KQML message

Future: as an encoding for an entire FIPA ACL message and as a way of publishing speech acts on web pages

27

What’s needed to build apps?

• Where does SW markup come from?– From Databases, just like much of the HTML on today’s web

– But, where does the DB content come from?• From legacy systems

• From web forms or custom HCIs

• From focused search engines feeding into web scrapers or information extraction apps

• Are the DAML tools there?– Some in beta form

– Many XML and RDF tools are very handy

– We used protégé and XMLSpy to create and edit ontologies

4

28

Conclusion

• ITTALKS is a useful, fairly sophisticated web application

• The semantic web concepts and DAML in particular – Make it easier to develop and maintain ITTALKS– Support some features of ITTALKS

• Visit http://ittalks.org/– To use ITTALKS– For more information, including a paper, a demo

“movie”, and these slides

• mailto:[email protected] to request a domain for your organization.

29