Upload
erin-sharp
View
220
Download
0
Tags:
Embed Size (px)
Citation preview
1USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Interactive Knowledge Capturefor Problem-Solving Systems
Jim Blythe
Yolanda Gil
Jihie Kim
www.isi.edu/ikcap
2USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Outline
Main elements of our approach
Experiences and lessons from earlier work: monolithic systems
New work on the Calo project:
open systems
3USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Acquiring procedure knowledge from users
Each user can have unique requirements of a system Their requirements will change over time, perhaps
frequently In travel planning, may be different for each trip
Users need to be able to modify procedure representations in intelligent systems to address their needs
Complements example-based learning
4USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Example: adding procedure knowledge to a travel planning tool
An assessment tool makes judgments about travel itineraries: e.g., the airline should be United or American e.g., the hotel should be within walking distance, unless I am
renting a car
Need to add procedure knowledge to tell the system to make a new kind of judgment:
“the hotel can cost up to 20% more than the government per diem rate for the city.”
or supporting procedures:
“to estimate driving time, divide the distance by 55”
5USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Expect’s Support for procedure KA:Key Technologies
Where does the user start? An acquisition wizard guides the user to start the KA process through
a dialog, based on problem-solving methods. KA takes many steps; users will be lost…
The acquisition wizard manages the process from end to end. Users don’t know the computer language.
An English-based procedure editor– Users modify the English paraphrase of the formal representation.
A search-based expression composer– Suggests valid reformulations of user sentences.
How do users ensure all the needed information is added? An interdependency analyzer understands which pieces of knowledge
are used to solve a problem.
6USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Main elements of our approach
Meta-reasoning: About how knowledge fragments are combined to
solve a goal About missing information, effects of changes to KB About the dialog and context
Putting knowledge in the user’s terms: Generate text descriptions to hide the syntax Browse-and-replace interface for procedures Search-based reformulations of free text Test knowledge on real examples
7USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Outline
Main elements of our approach
Experiences and lessons from earlier work: monolithic systems using Expect
New work on the Calo project:
open systems
8USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Acquisition wizard
Dialog with user to start the process.
Some questions use menus or text input.
Others use the English editor to refine procedural knowledge
9USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
English-based Procedure editor
NL description of method
Alternatives forselected text fragment
(multiply
(obj (look-up
(obj fsa-per-diem-hotel-rate)
(for (r-city ?hotel))))
(by 1.2))
10USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Test new knowledge immediately
Each element is defined and checked
using constraints
Softools, developed under the DARPA AcT program
11USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
U: new port: HavanaS: I need to know if it is an airport or a seaportU: seaportS: I need to know the location and the berths
...
...
r-location
r-berths
r-pols
r-piers
r-storage-area
(evaluate (obj coa) (wrt logistics))
... (r-location port)
... (r-berths seaport)
port
airport seaport
inlandwaterwayseaport
maritimeseaport
Domain Ontology Problem-Solving Methods
Global Procedure Analysis:Test interdependencies, both rule-rule and rule-data
INTERDEPENDENCIES
Interdependencies guide Knowledge Acquisition
12USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
When combined, the tools use information from each other
Application
Acquisitionwizard
Acquisitionanalyzer
Interdependencyanalyzer
Procedure editor
Relation/concept editor
Instance editor
Expression composer
13USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
The acquisition wizard
Guides the user through the initial steps of adding new knowledge.
Structures the knowledge to be added using default procedural knowledge.
Questions are generated from a problem-solving theory.
14USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Problem-solving theory for plan evaluation
A hierarchy of generic types of plan judgments with default procedural knowledge.
DEFINED: check that the value is less than the maximum value
ASK USER: compute a maximum value for each object
judgment
global judgment local judgment bounds check extensional check
completeness judgment
hotel cost judgment
upper bound lower bound positive negative
“Warn if the value is too large?”
15USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Benefits of integration: The acquisition wizard and the method editor
Each component receives information from the other that helps the user:
The wizard provides to the editor: An initial version of the method, with the correct capability An expectation of the result type of the method
The editor provides to the wizard: A more detailed method result type Used to help classify the new task in the ontology
16USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
bounds check
upper bound lower bound
“Warn if the value is too large?”
17USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
The expression composer
Assists users in formalizing informal statements by suggesting composite expressions using terms the system understands.
Uses an ontology of terms and relations, and synonyms derived from WordNet with extensions.
Makes breadth-first forward search, matching keywords
18USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Expression composer example
User types: “max staging post landing”
Tool suggests: “find the maximum of the landing distance available of the runways of the forward staging post”.
find
object of
landing-distance-available
runways
?forward-staging-post
maximum
Function call
reformulation
Typed variablecombining queries
Information element
19USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Integrated with the method editor
Constructs compound terms in KB that include user terms and have desired type
Anytime breadth-first search through space of terms
20USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Outline
Main elements of our approach
Experiences and lessons from earlier work: monolithic systems
New work on the Calo project:
open systems
21USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Tailor: interactive task acquisition forthe Calo office assistant
A personal assistant that manages everyday tasks. Learns from experience, interacts with the user
naturally through several channels. Large DARPA-funded project run by SRI.
Goal2ACT8
sleeping
Fact1ACT2normal
Goal3ACT3
sleeping
Intention Graph
Cue: (TEST (source vision new))
ACT2
Process Library
Process Execution
(source commanders-vision new)
(ACHIEVE (satisfy Incoming-INs))
New Facts & Goals
ExternalWorld
1
2
3
4
5
6
7
8
Cue: (ACHIEVE (satisfy source))
ACT1Facts&
Goals
(ACHIEVE (derive IN-list source))
ACT1current
Spark is Calo’s task manager.
(A PRS-based system developed by Myers et al.)
22USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Approach and challenges in acquiring knowledge within CALO
Carry key elements forward: Explicit reasoning about knowledge dependencies
and acquisition context Expressing the procedures in the user’s terms
Earlier work used Expect, a monolithic system using custom problem-solving system, language, methods, ontology,..
Now use Spark’s performance element and procedures, ontologies provided by several groups.
23USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Example: purchasing a laptop
Spark manages the workflow as the user purchases a laptop. Choose a model, find bids, get authorization, track
purchase…
During the process, Spark finds that the order cannot be completed, because a manager who must authorize the purchase is not available.
User should be able to tell the system “You don’t need authorization when the cost is less than $2000”
24USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Initial solution
Make direct analysis of the Spark procedures Encapsulates a model of Spark’s behaviour Requires additions to language, e.g. types, purpose of
subtasks
Combine the procedure analysis with the expression composer to help interpret user sentences.
Dialog is currently implicit Currently provide follow-up questions based on analysis.
25USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Expression composer suggests valid
condition
User can explore modifications based
on different assumptions
Response to“You don’t need authorization when the cost is less than $2000”
Automatic generation of text from Spark
procedure definitions
26USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Tailor: summary and future work
Currently allow users to modify existing procedures, using global analysis and expression composer.
Next:
Support defining new tasks, integrate with advice for Spark.
Model dialog: build templates for supporting theory, working with Allen/Ferguson on dialog model.
27USC INFORMATION SCIENCES INSTITUTE Interactive Knowledge Capture
Summary
Our work allows users to add procedure knowledge to both custom and pre-existing intelligent systems.
Combination of global analysis and term reformulation can allow users and the system to reach shared understanding.
Interactive knowledge capture complements example-based learning to allow intelligent systems to adapt.
Evaluation strategies: Ablation user studies with fixed and free tasks. Good results with Expect & Constable, still to do with Tailor.