Phosphorus: OntologyBased Matchmaking

1
PhosphorusOntology-Based Matchmaking

• Hans Chalupsky
• Yolanda Gil
• Tom Russ
• Surya Ramachandran
• Information Sciences Institute

2
Ontology-based Matchmaking

• Research goals
• Description-based advertisements and requests
• EXPECTs goal and capability descriptions
• Vocabulary within descriptions derived from
• Performative/Action Ontology
• Domain ontologies
• Broad coverage ontology (e.g., SENSUS)
• Classifier match and partial match
• PowerLoom classifier
• Chameleon partial matcher (combines deduction and
  neural nets)
• Adaptive (trainable) matching
• Multilingual descriptions

3
Classifier Match and Partial Match
Calculate round-trip time (RTT) for aircraft
Find route from location1 to location2
Calculate RTT for combat aircraft
Calculate RTT for transport aircraft
Find egress route from Ryad to Kuwait city
A) Subsumption-based match the request is
subsumed by an agents capability
B) Reformulation-based match the request can
be satisfied by combining the capabilities of two
or more agents
Find route from location1 to location2
Find phone numbers of US citizens in Kuwait
Find addresses of US citizens in Kuwait
Find route from city1 to city2
C) Reverse subsumption-based match an agent can
satisfy some aspect of the request
D) Partial match an agent has a capability that
is similar/related to the original request
4
Agent Matching

• Problem I topic matching (e.g., interests
  matcher e.g., roses)
• Given - thousands/millions of agent
  descriptions
• - a request
• Find a set of agents that can fulfill the
  request
• (and/or something similar to the
  request)
• (and/or can understand some of the
  request)
• (and/or could help reformulate the
  request)
• Problem II task-based matching (e.g.,
  activities matcher)
• Given - a few dozens/hundreds agent
  descriptions
• - a request
• Find the few agents that can fulfill the request
  
• (and refinements of it with
  additional requirements)

5
Matcher Architecture
Term(s) (e.g. CoABS)
Task description (e.g. give demo of TC)
Requests

• Topic-Based
• Matcher

Task-Based Matcher
ml
Ontology-Based Matching Shell
Topic Ontology (e.g., research interests)
Activities Ontology (e.g., research activities)
subsumption
reformulation
abstraction
Agent Descriptions
aa rn qo rs tv
ps
sc
ai vs aa
information gathering agent
scheduling agent
printer agent
researcher
Agents
6
MatchingTask-Based Capabilities and Requests

• Represent task descriptions more declaratively
• (give (obj (spec-of demonstration)) (of
  Teamcore))
• (process (obj (spec-of reimbursement)) (for
  (set-of receipt)))
• (demo (obj Teamcore))
• Reformulations of requests class partition
  sets
• (setup (obj (equipment)))
• (setup (obj
  (lcd)))
• (setup (obj
  (vcr)))
• (demo (obj (Ariadne Teamcore)))
• (demo (obj
  (Ariadne)))
• (demo (obj
  (Teamcore)))

• task qualification parameter
• matches concept in goal
• further specifies task to be done (i.e., how
  action is done)
• allows same method to be used for variety of
  tasks

exploits definitions during matching (demo
(obj sw)) (give (obj (spec-of demonstration))
(of sw))
7
Benefits

• Loose coupling
• Flexible invocation requests do not have to
  mirror the agent descriptions as originally
  stated
• Semantics of the task and its arguments are at
  the core of the matching process through
  subsumption and reformulations
• Declarative representation of task descriptions
• Not only data parameters but also task
  qualification parameters
• Automatic organization of agent capabilities
• Object and task taxonomies are basis for indexing
  agents
• Can support partial matching
• Suggests alternative formulations of requests
  when requests do not match exactly the
  capabilities of available agents

8
Topic Matching with PowerLoom

• Express advertisements and requests as logical
  descriptions
• Domain ontologies provide term definitions
• Representation language is KIF (use XML-rendering
  to embed advertisements on Web pages)
• Use standard PowerLoom inference and
  classification mechanism to support matchmaking
• Use subsumption hierarchy and KILTER partial
  match technology to support relaxed matching

9
Example PowerLoom Advertisement

• Advertisement (advertises
  Yolanda-Gil (kappa (?i) (exists
  (?p) (and (research-interest ?p ?i)
  (subset-of ?i Knowledge-Acquisition)
  ))
• Example Query Who is interested in
  knowledge-based systems? (retrieve ?p
  (and (Person ?p) (exists (?ad)
  (advertises ?p ?ad)
  (subset-of ?ad
  (kappa (?i) (exists (?p)
  (and (research-interest ?p ?i)
  (subset-of ?i
  Knowledge-Based-Systems))))))))

10
(No Transcript)
11
Future Work

• Extend descriptions of agent capabilities
• Tasks agents can perform (including results
  returned)
• Agent invocation guidelines (including inputs to
  be provided)
• Ontological commitments made by the agent
• Additional agents involved
• Agents consulted or invoked to get additional
  information
• Subtasks delegated to other agents
• Qualifications of the agent
• Reliability, efficiency, resources available,
• Model of how tasks are performed by the agent
• Differential properties (comparisons with other
  agents)

12
Description of Task-Based Capabilities Related
Work

• Agent capability languages
• LARKS
• Describing Problem-Solving Methods (e.g., a
  scheduler)
• HPKB PSM Jumpstart
• UPML
• EXPECT
• Process Descriptions
• NIST PSL
• EO
• Workflow
• Process handbook
• Software reuse

13
Issues in Task-Based Matching (I)

• A single agent can perform a wide range of tasks
• Currently, agents can do at most a handful (i.e.,
  one)
• Nominate alternative agents
• Flexible invocation
• A request to Register for local conference is
  treated by a PA as Arrange travel to a meeting
• Invocations of other agents
• Advertise delegation to other agents,
  consultations to get additional information
• Describing peoples capabilities
• Project assistants as everything agents
  (information agents, matching services, proxies
  of travel agents, etc.)

14
Issues in Task-Based Matching (II)

• Requesters will not provide exact description of
  required capability
• e.g. find route to San Diego
• Missing input data from where?
• Imprecise specs surface route? air route?
• Qualification of results expected 3ft segments?
  major points?
• Third parties may need to be invoked to help
  specify all inputs needed
• e.g. find route from LA to San Diego
• Route planner agent needs a map as an input
• Route planner agent takes lat/long as input, not
  city names

15
Research Topics

• Description of agent capabilities
• Using ontologies
• Which ones
• What makes a good ontology
• Partial match
• Learning from experience
• Refining agent descriptions over time
• Negotiation
• Refining a request based on available agents

16
Sample KB Sizes
CC Cycs IKB 99
0.1521 sec with rep req
WG uc
WG Cycs IKB 98
0.1671 sec no rep req
0.5036 sec no rep req
17
Task-Based Matchmaking

• Yolanda Gil
• Surya Ramachandran
• Hans Chalupsky
• Tom Russ
• Information Sciences Institute