Increasing Trust in Answers from Intelligence Applications: the Inference Web Approach

1
Increasing Trust in Answers from Intelligence
Applicationsthe Inference Web Approach

• Deborah McGuinness
• Co-Director and Senior Research Scientist
• Knowledge Systems Laboratory
• Stanford University
• dlm_at_ksl.stanford.edu
• http//www.ksl.stanford.edu/people/dlm
• Inference Web is joint work with Pinheiro da
  Silva, Fikes, Chang, Deshwal, Narayanan, Glass,
  Makarios, Jenkins, Millar, Ding,

2
Semantic Web Layers

• Ontology Level
• Languages (CLASSIC, DAML-ONT, DAMLOIL, OWL, )
• Environments (FindUR, Chimaera,
  OntoBuilder/Server, Sandpiper Tools, )
• Standards (NAPLPS, , W3Cs WebOnt, W3Cs
  Semantic Web Best Practices, EU/US Joint
  Committee, OMG ODM,
• Rules
• SWRL (previously CLASSIC Rules, explanation
  environment, extensibility issues, contracts,
  )
• Logic
• Description Logics
• Proof
• PML, Inference Web Services and Infrastructure
• Trust
• IWTrust, NSF with W3C/MIT

http//www.w3.org/2004/Talks/0412-RDF-functions/sl
ide4-0.html
3
Motivation Trust and Understanding

• If users (humans and agents) are to use, reuse,
  and integrate system answers, they must trust
  them.
• System transparency supports understanding and
  trust.
• Even simple lookup systems benefit from
  providing information about their sources.
• Systems that manipulate information (with sound
  deduction or potentially unsound heuristics)
  benefit from providing information about their
  manipulations.

Goal Provide interoperable infrastructure that
supports explanations of sources, assumptions,
and answers as an enabler for trust.
4
Requirements gathered from

• DARPA Agent Markup Language (DAML)
• Enable the next generation of the web
• DARPA Personal Assistant that Learns (PAL)
• Enable computer systems that can reason, learn,
  be told what to do, explain what they are doing,
  reflect on their experience, respond robustly
  to surprise
• DARPA Rapid Knowledge Formation (RKF)
• Allow distributed teams of subject matter experts
  to quickly and easily build, maintain, and use
  knowledge bases without need for specialized
  training
• DARPA High Performance Knowledge Base (HPKB)
• Advance the technology of how computers acquire,
  represent manipulate knowledge
• ARDA Novel Intelligence for Massive Data (NIMD)
• Avoid strategic surprise by helping analysts be
  more effective (focus attention on critical
  information and help analyze/prune/refine/explain/
  reuse/)
• ARDA Advanced Question Answering for
  Intelligence (AQUAINT)
• Advance QA against structured and unstructured
  info
• Consulting including search, ecommerce,
  configuration,

5
Requirements

• Information Manipulation Traces
• hybrid, distributed, portable, shareable,
  combinable encoding of proof fragments supporting
  multiple justifications
• Presentation
• multiple display formats supporting browsing,
  visualization, summaries,
• Abstraction
• understandable summaries
• Interaction
• multi-modal mixed initiative options including
  natural-language and GUI dialogues, adaptive,
  context-sensitive interaction
• Trust
• source and reasoning provenance, automated trust
  inference
• McGuinness Pinheiro da Silva, ISWC 2003, J.
  Journal of Web Semantics 2004

6
Selected History

• Historical explanation research motivated by
  explaining theorem provers in practice
• Web version originally aimed at explaining hybrid
  (FOL / special purpose) reasoners in a
  distributed environment like the web.
• User demand drove focus on provenance extensions
• Current web environment and programs, such as
  NIMD, drove connections with extraction engines
• Current view Any question answering system can
  be viewed as some kind of information manipulator
  that may benefit from and/or require explanation

7
Inference Web

• Framework for explaining question answering tasks
  by abstracting, storing, exchanging, combining,
  annotating, filtering, segmenting, comparing, and
  rendering proofs and proof fragments provided by
  question answerers
• IWs Proof Markup Language (PML) is an
  interlingua for proof interchange. It is written
  in W3Cs recommended Ontology web language (OWL)
• IWBase is a distributed repository of
  meta-information related to proofs and their
  explanations
• IW Registration services provide support for
  proof generation and checking
• IW Browser provides display capabilities for PML
  documents containing proofs and explanations
  (possibly from multiple inference engines)
• IW Abstractor provides rewriting capabilities
  enabling more understandable presentations
• IW Explainer provides multi-modal dialogue
  options including alternative strategies for
  presenting explanations, visualizations, and
  summaries

• Work with Pinheiro da Silva

8
Explainable System Structure
Trust
Explanation
Interaction
Presentation
Abstraction
PML
InferenceML
Proof Interlingua (PML)
Inference Rule Specs
Source Provenance Data
Information Manipulation Data
Source Provenance Data
Information Manipulation Data
9
Registry Information

• IWBase has core and domain-specific repositories
  of meta-data useful for disclosing knowledge
  provenance and reasoning information such as
  descriptions of
• Question answering systems (Inference Engines,
  Extractors, ) along with their supported
  inference rules
• Information sources such as organizations,
  publications and ontologies
• Representation languages along with their axioms

10

11

12
Browsing Proofs

• Enable the visualization of proofs (and
  abstracted proofs)
• Proofs can be extracted and browsed from both
  local and remote PML node sets and can be
  combined
• Links provide access to proof-related
  meta-information

select
select
13
Browsing Proofs
14
Explainer

• Present
• Query
• Answer
• Abstraction of justification (PML information)
• Limited meta information
• Suggests drill down options (also provides
  feedback options)

15
UIMA Explanation
16
Follow-up Metadata
17
Follow-up Assumptions
18
Explaining Extracted Entities (Techies)
Sentences in English
Sentences in annotated English
Sentences in logical format, i.e., KIF
19
Further Observations on Explaining Extracted
Entities
Source fbi_01.txt Source Usage span from 01 to
78
Same conclusion from multiple extractors
conflicting conclusion from one extractor
This extractor decided that Person_fbi-01.txt_46
is a Person and not Occupation
20
Search / Configuration
21
KSL Wine AgentSemantic Web Integration Example

• Uses emerging web standards to enable smart web
  applications
• Given a meal description
• Deborahs Specialty
• Describe matching wines
• White, Dry, Full bodied
• Retrieve some specific options from web
• Forman Chardonnay from DLMs cellar, ThreeSteps
  from wine.com, .
• Info http//www.ksl.stanford.edu/people/dlm/webo
  nt/wineAgent/

22
KSL Wine Agent Semantic Web Integration
Technology
OWL for representing a domain ontology of
foods, wines, their properties, and relationships
between them JTP theorem prover for deriving
appropriate pairings DQL/OWL QL for querying
a knowledge base Inference Web for explaining
and validating answers (descriptions or
instances) Web Services for interfacing with
vendors Connections to online web
agents/information services Utilities for
conducting and caching the above transactions
23

24
Knowledge Provenance Elicitation
XYZ says AB is the answer for my question.
Provenance information may be essential for users
to trust answers.
Data provenance (aka data lineage) is defined and
studied in the database literature. Buneman et
al., ICDT 2001 Cui and Widom, VLDB 2001
Why should I believe this?
Knowledge provenance extends data provenance by
adding data derivation provenance
information Pinheiro da Silva, McGuinness
McCool, Data Eng. Bulletin, 2003
25
IWTrust Trust in Action
Google-2.0 says AB is the answer for my
question.
Trust can be inferred from a Web of Trust.

Why should I trust the answer?
0

?

IWTrust provides infrastructure for building webs
of trust.
The infrastructure includes a trust component
responsible for computing trust values for
answers.
IWTrust is described in Zaihrayeu, Pinheiro da
Silva McGuinness, iTrust 2005
AB
0



0
?
?
26
Explanation Application Areas

• Theorem proving
• First-Order Theorem Provers Stanford (JTP
  (KIF/OWL/)) SRI (SNARK) University of Texas,
  Austin (KM)
• SATisfiability Solvers University of Trento
  (JSAT)
• Information extraction IBM (UIMA), Stanford
  (TAP)
• Information integration/aggregation USC ISI
  (Prometheus,Mediator -gt Fetch) Rutgers ,
  Stanford (TAP)
• Task processing SRI International (SPARK)
• Service composition Stanford, U. of Toronto,
  UCSF (SDS)
• Semantic matching University of Trento
  (S-MATCH)
• Debugging ontologies U of Maryland, College
  Park (SWOOP/Pellet)
• Problem solving University of Fortaleza
• Trust Networks U. of Trento (IWTrust), UMd

27
Inference Web Contributions

• Language for encoding hybrid, distributed proof
  fragments based on web technologies. Support for
  both formal and informal proofs (information
  manipulation traces).

Trust
Explanation
Interaction
2. Support (registry, language, services) for
knowledge provenance
Presentation
Abstraction
3. Declarative inference rule representation for
checking hybrid, distributed proofs.
Inference Meta-Language
Proof Markup Language
Inference Rule Specs
Provenance Meta-data
Information Manipulation Data
4. Multiple strategies for proof abstraction,
presentation and interaction.
5. End-to-end trust value computation for
answers.
6. Comprehensive solution for explainable systems
28
Status

• Inference Web infrastructure (PML, browser,
  explainer, registry, toolkit) being used in
  government programs such as PAL and NIMD,
  commercial research labs IBM, Boeing, SRI,
  Universities USC, U MD,
• Integration and registration process underway
  with extraction community
• Useful now for helping decide if information is
  trustworthy, comes from authoritative sources,
  consistent, reliable
• Benefits from more meta data and more information
  population but is useful in an incremental nature

29
Technical Status

• Some focus areas
• Follow-up question support
• Trust
• Contradiction support
• Abstraction techniques
• Extraction extensions
• Task-oriented reasoning support
• Query manager explanation support
• Toolkit for embedding
• Open issues for explanation
• Granularity of explanations
• Meta information filtering
• Abstraction techniques
• Requests / Suggestions?

30
More Info

• Inference Web http//iw.stanford.edu
• OWL http//www.w3.org/TR/owl-features/
  http//www.w3.org/TR/owl-guide/
• DAMLOIL http//www.daml.org/
• WineAgent www.ksl.stanford.edu/people/dlm/webont/
  wineAgent/
• Chimaera http//www.ksl.stanford.edu/software/c
  himaera/
• OWL-QL/DQL http//www.ksl.stanford.edu/projec
  ts/dql/
• UIMA http//www.research.ibm.com/UIMA/
• dlm_at_ksl.stanford.edu

31
Extras

32
Background

• ATT Bell Labs AI Principles Dept
• Description Logics, CLASSIC, explanation,
  ontology environments
• Semantic Search, FindUR, Collaborative Ontology
  Building Env
• Apps Configurators, PROSE/Questar, Data Mining,
  
• Stanford Knowledge Systems, Artificial
  Intelligence Lab
• Ontology Evolution Environments (Diagnostics and
  Merging) Chimaera
• Explanation and Trust, Inference Web
• Semantic Web Representation and Reasoning
  Languages, DAML-ONT, DAMLOIL, OWL,
• Rules and Services SWRL, OWL-S, Explainable SDS,
  KSL Wine Agent
• McGuinness Associates
• Ontology Environments Sandpiper, VerticalNet,
  Cisco
• Knowledge Acquisition and Ontology Building
  VSTO, GeON, ImEp,
• Applications GM Search, etc. CISCO meta
  data org, etc.
• Boards Network Inference, Sandpiper,
  Buildfolio, Tumri, Katalytik

33
KSL Wine Agent Semantic Web Integration (Toy)
Example

• Uses emerging web standards to enable smart web
  application
• Given a meal description
• Deborahs Specialty, a crab dish,
• Describe matching wines
• White, Dry, Full bodied
• Retrieve some specific options from web
• Forman Chardonnay from DLMs cellar, ThreeSteps
  from wine.com, .
• Explain description or specific suggestion
• Info http//www.ksl.stanford.edu/people/dlm/webo
  nt/wineAgent/

34
KSL Wine Agent Semantic Web Integration
Technology

• OWL for representing a domain ontology of
  foods, wines, their properties, and relationships
  between them
• JTP theorem prover for deriving appropriate
  pairings
• Chimaera ontology diagnostics and ontology
  merging
• DQL/OWL QL for querying a knowledge base
• Inference Web for explaining and validating
  answers (descriptions or instances)
• Web Services for interfacing with vendors
• Connections to online web agents/information
  services
• Utilities for conducting and caching the above
  transactions

35
Inference Web in KANI Context
36
Summary

• Tools are emerging that support understanding
  information
• Understanding/Evaluating information can help
  focus a users attention and enable trust, reuse,
  and filtering
• Semantic Web infrastructure (OWL, Structured
  query languages, Semantic Search, Extractors,
  Reasoners, Explanation Infrastructure, .) is
  ready for use and a growing trend

37
Knowledge Provenance Multiple Sources
Answer
Source
Source
38
Extra
39
Inferences drawn by Information Extraction
40
Infrastructure Core IWBase
Statistics for relevant domain independent
meta-data
29
Inference Engines

56
Axioms
38
Declarative Rules
select

10
Method Rules
6
Derived Rules
12
Languages
select
41
Explaining Answers GUI Explainer
Users can exit the explainer providing feedback
about their satisfiability with explanation(s)
Select action
Users can ask for alternative explanations or
summaries
42
Follow-up PML Abstraction(Techies only)
43
Browsing

• Present
• Query
• Answer
• Alternate formats (KIF, English, Raw, )
• Graph Structure (with lens view)
• Annotations

44
Knowledge Provenance Multiple Sources
Answer
Source
Source
45
IWTrustImproving User
Trust in Answers from the Web

• Ilya Zaihrayeu
• ITC-IRST
• Paulo Pinheiro da Silva
• Deborah L. McGuinness
• Stanford University

46
Trusting Answers

• It may be challenging for users to establish
  their degrees of trust, untrust, mistrust and
  distrust in a web application answer if the
  answer is provided without any kind of
  justification
• Knowledge Provenance (KP) is a description of
  both the origins of knowledge and the reasoning
  process to produce an answer
• Users may need KP to establish a degree of trust
  in the answer
• Which sources were used?
• Who are the authors of such sources?
• Which engines were used?
• What are the assumptions of the engines? Are the
  engines rules sound?
• KP itself may not be enough for trusting the
  answer
• I may not know anything about one or more sources
  in the KP
• I may have no information about the reliability
  of one or more of then engines in the KP

47
Trusting Answers from the Web

• The overall process of establishing a degree of
  trust in answers from web applications is
  particularly complex since applications may rely
  on
• Hybrid and distributed processing, e.g., web
  services, the Grid
• Large number of heterogeneous, distributed
  information sources, e.g., the Web
• information sources with more variation in their
  reliability, e.g., information extraction
• Sophisticated information integration methods,
  e.g., SIMS, TSIMMIS
• The definition of trust is a significant part of
  the process

• The task of keeping, encoding, sharing and
  gathering KP for partial answers towards the
  generation of the KP for answers is another part
  of the process
• The use of KP to derive trust values for answers
  is yet another part of the process

48
The Inference Web

• The Inference Web is an infrastructure supporting
  explanations for answers from the web
• The Proof Markup Language (PML) is used to encode
  answer justification, i.e., information
  manipulation traces, proofs
• IWBase is used to annotate PML documents with
  proof-related data, i.e., trust values for
  sources and engines

• User U1 asks question Q
• A question answering system returns the set of
  answers A1,A2,,An

PML Documents
IWBase
S1
(A1, t11, t12,...)
IE1
Q(U1)
S2
(A2, t21, t22,...)
...
S3
...
(An, tn1, tn2,...)
IE2
49
Inference Web and KP

• Inference Web is an infrastructure supporting
  KP for answers derived by multiple methods
• Information extraction IBM (UIMA), Stanford
  (TAP)
• Information integration USC ISI
  (Prometheus/Mediator) Rutgers University
  (Prolog/Datalog)
• Task processing SRI International (SPARK)
• Theorem proving
• First-Order Theorem Provers SRI International
  (SNARK) Stanford (JTP) University of Texas,
  Austin (KM)
• SATisfiability Solvers University of Trento
  (J-SAT)
• Expert Systems University of Fortaleza (JEOPS)
• Service composition Stanford, University of
  Toronto, UCSF (SDS)
• Semantic matching University of Trento
  (S-Match)
• Debugging ontologies University of Maryland,
  College Park (SWOOP/Pellet)
• Problem solving University of Fortaleza
  (ExpertCop)

50
The Inference Web Trust (IWTrust)

• IWTrust extends the Inference Web to support
  trust computation
• IW TrustNet is a social network of source
  recommenders
• A trust component implementing an algorithm to
  compute trust values for answers
• Trust values are used to rank answers and answer
  justifications
• User U1 trusts U3 to a degree t1-3

PML Documents
IWBase
S1
(A1, t11, t12,...)
IE1
Q(U1)
S2
(A2, t21, t22,...)
...
S3
...
(An, tn1, tn2,...)
IE2
51
Conclusions

• IWTrust provides infrastructure for building a
  trust graph from a user asking a question to the
  answers for the question
• Knowledge Provenance is a key element of the
  trust graph and a requirement for trusting
  answers in general
• Inference Web is a Semantic Web solution for
  Knowledge Provenance
• iw.stanford.edu
• Adaptive explanations based on user modeling
• IWBase registration of a large set of software
  systems
• Registration of a comprehensive set of primitive
  rules
• Established library of explanation tactics

• IWTrust intends to be a solution for the Semantic
  Web trust layer

• Inference Web is a solution for the Semantic Web
  proof layer