Problem-Solving Methods in Protégé-2000 Monica Crubézy Stanford Medical Informatics MIS 301 - November 1999.

Slides:



Advertisements
Similar presentations
ARCHITECTURES FOR ARTIFICIAL INTELLIGENCE SYSTEMS
Advertisements

Ch:8 Design Concepts S.W Design should have following quality attribute: Functionality Usability Reliability Performance Supportability (extensibility,
16/11/ IRS-II: A Framework and Infrastructure for Semantic Web Services Motta, Domingue, Cabral, Gaspari Presenter: Emilia Cimpian.
27 January Semantically Coordinated E-Market Semantic Web Term Project Prepared by Melike Şah 27 January 2005.
Analysis Modeling.
Requirements Engineering n Elicit requirements from customer  Information and control needs, product function and behavior, overall product performance,
Internet Reasoning Service: Progress Report Wenjin Lu and Enrico Motta Knowledge Media Institute Monica Crubézy Stanford Medical Informatics.
Software Testing and Quality Assurance
The Unified Software Development Process - Workflows Ivar Jacobson, Grady Booch, James Rumbaugh Addison Wesley, 1999.
The Semantic Web Week 13 Module Website: Lecture: Knowledge Acquisition / Engineering Practical: Getting to know.
Knowledge Acquisitioning. Definition The transfer and transformation of potential problem solving expertise from some knowledge source to a program.
1 SYSTEM and MODULE DESIGN Elements and Definitions.
Knowledge Modelling: Foundations, Techniques and Applications Enrico Motta Knowledge Media Institute The Open University United Kingdom.
An Intelligent Broker Approach to Semantics-based Service Composition Yufeng Zhang National Lab. for Parallel and Distributed Processing Department of.
School of Computing and Mathematics, University of Huddersfield Knowledge Engineering: Issues for the Planning Community Lee McCluskey Department of Computing.
1 System: Mecano Presenters: Baolinh Le, [Bryce Carder] Course: Knowledge-based User Interfaces Date: April 29, 2003 Model-Based Automated Generation of.
Software Requirements
Knowledge Acquisition CIS 479/579 Bruce R. Maxim UM-Dearborn.
PROMPT: Algorithm and Tool for Automated Ontology Merging and Alignment Natalya F. Noy and Mark A. Musen.
Developing an Ontology-based Metadata Management System for Heterogeneous Clinical Databases By Quddus Chong Winter 2002.
Protégé An Environment for Knowledge- Based Systems Development Haishan Liu.
The Software Product Life Cycle. Views of the Software Product Life Cycle  Management  Software engineering  Engineering design  Architectural design.
KMi-SMI collaboration Wenjin Lu and Enrico Motta Knowledge Media Institute Monica Crubézy Stanford Medical Informatics.
1212 Management and Communication of Distributed Conceptual Design Knowledge in the Building and Construction Industry Dr.ir. Jos van Leeuwen Eindhoven.
Course Instructor: Aisha Azeem
The Multi-model, Metadata-driven Approach to Content and Layout Adaptation Knowledge and Data Engineering Group (KDEG) Trinity College,
31 st October, 2012 CSE-435 Tashwin Kaur Khurana.
Knowledge Mediation in the WWW based on Labelled DAGs with Attached Constraints Jutta Eusterbrock WebTechnology GmbH.
Knowledge Management in Geodise Geodise Knowledge Management Team Liming Chen, Barry Tao, Colin Puleston, Paul Smart University of Southampton University.
Made with Protégé: An Intelligent Medical Training System Olga Medvedeva, Eugene Tseytlin, and Rebecca Crowley Center for Pathology Informatics, University.
Katanosh Morovat.   This concept is a formal approach for identifying the rules that encapsulate the structure, constraint, and control of the operation.
Methods for Computer-Aided Design and Execution of Clinical Protocols Mark A. Musen, M.D., Ph.D. Stanford Medical Informatics Stanford University.
Requirements Analysis
Knowledge representation
Design Science Method By Temtim Assefa.
Software Requirements Engineering CSE 305 Lecture-2.
Second Generation ES1 Second Generation Expert Systems Ahme Rafea CS Dept., AUC.
SOFTWARE DESIGN (SWD) Instructor: Dr. Hany H. Ammar
Odyssey A Reuse Environment based on Domain Models Prepared By: Mahmud Gabareen Eliad Cohen.
Design engineering Vilnius The goal of design engineering is to produce a model that exhibits: firmness – a program should not have bugs that inhibit.
© DATAMAT S.p.A. – Giuseppe Avellino, Stefano Beco, Barbara Cantalupo, Andrea Cavallini A Semantic Workflow Authoring Tool for Programming Grids.
1 Workshop on Business-Driven Enterprise Application Design & Implementation Cristal City, Washington D.C., USA, July 21, 2008 How to Describe Workflow.
Protégé as Professor: Development of an Intelligent Tutoring System With Protégé-2000 Olga Medvedeva Center for Pathology Informatics University of Pittsburgh.
1 Introduction to Software Engineering Lecture 1.
Refining middleware functions for verification purpose Jérôme Hugues Laurent Pautet Fabrice Kordon
Knowledge Representation of Statistic Domain For CBR Application Supervisor : Dr. Aslina Saad Dr. Mashitoh Hashim PM Dr. Nor Hasbiah Ubaidullah.
Elizabeth Furtado, Vasco Furtado, Kênia Sousa, Jean Vanderdonckt, Quentin Limbourg KnowiXML: A Knowledge-Based System Generating Multiple Abstract User.
©Ferenc Vajda 1 Semantic Grid Ferenc Vajda Computer and Automation Research Institute Hungarian Academy of Sciences.
A Context Model based on Ontological Languages: a Proposal for Information Visualization School of Informatics Castilla-La Mancha University Ramón Hervás.
Temporal Mediators: Integration of Temporal Reasoning and Temporal-Data Maintenance Yuval Shahar MD, PhD Temporal Reasoning and Planning in Medicine.
1 USC INFORMATION SCIENCES INSTITUTE CAT: Composition Analysis Tool Interactive Composition of Computational Pathways Yolanda Gil Jihie Kim Varun Ratnakar.
1 USC, INFORMATION SCIENCES INSTITUTE An integrated environment for KA An Integrated Environment for Knowledge Acquisition Jim Blythe
Chapter 6 – Architectural Design Lecture 1 1Chapter 6 Architectural design.
1 Knowledge Acquisition and Learning by Experience – The Role of Case-Specific Knowledge Knowledge modeling and acquisition Learning by experience Framework.
Service Brokering Yu-sik Park. Index Introduction Brokering system Ontology Services retrieval using ontology Example.
1 USC INFORMATION SCIENCES INSTITUTE EXPECT TEMPLE: TEMPLate Extension Through Knowledge Acquisition Yolanda Gil Jim Blythe Information Sciences Institute.
17/1/1 © Pearson Education Limited 2002 Artificial Intelligence & Expert Systems Lecture 1 AI, Decision Support, Architecture of expert systems Topic 17.
Henrik Eriksson Department of Computer and Information Science Linkoping University SE Linkoping, Sweden Raymond W. Fergerson Yuval Shahar Stanford.
Abdul Rahim Ahmad MITM 613 Intelligent System Chapter 10: Tools.
Ontologies Reasoning Components Agents Simulations An Overview of Model-Driven Engineering and Architecture Jacques Robin.
Supporting Collaborative Ontology Development in Protégé International Semantic Web Conference 2008 Tania Tudorache, Natalya F. Noy, Mark A. Musen Stanford.
Protégé/2000 Advanced Tools for Building Intelligent Systems Mark A. Musen Stanford University Stanford, California USA.
Metadata Driven Aspect Specification Ricardo Ferreira, Ricardo Raminhos Uninova, Portugal Ana Moreira Universidade Nova de Lisboa, Portugal 7th International.
1 Software Requirements Descriptions and specifications of a system.
Mechanisms for Requirements Driven Component Selection and Design Automation 최경석.
Human Computer Interaction Lecture 21 User Support
APPLICATION OF DESIGN PATTERNS FOR HARDWARE DESIGN
Introduction To software engineering
Semantic Markup for Semantic Web Tools:
Building Ontologies with Protégé-2000
Presentation transcript:

Problem-Solving Methods in Protégé-2000 Monica Crubézy Stanford Medical Informatics MIS November 1999

2 What is a Problem-Solving Method ? A problem-solving method (PSM): realises the task of a KBS (planning, diagnosis) implements a reasoning process in a program (search in a tree) employs the concepts of the KB as knowledge roles (findings) PSMs cover: interpreters, rule engines, algorithms generic methods in AI (propose&revise) domain-specific methods (Protean - ribosome topology identification )

3 Problem-Solving in Protégé Protégé was domain-independent, but method-specific: expert-system shell for Episodic Skeletal Plan Refinement problem- or task-specific KBs (protocol-based advice) Since Protégé-II, the behaviour of a KBS is: an abstract model of problem-solving, i.e. a PSM identifies roles of domain knowledge & ordering of inferences composed from more primitive methods an independent component, reusable across tasks, with libraries of generic PSMs (Eriksson et al. 1995) across domains, with declarative mapping relations (Gennari et al & 1995, Park et al. 1998) Protégé-2000 is entirely application-independent: generic knowledge model and tools to build an ontology of classes, customize KA forms and enter specific instances no embedded problem-solving assumptions

4 What is a PSM in Protégé-2000 ? A separate component in a KBS architecture, that: provides a function, as a “black box” makes its interface requirements on inputs and outputs explicit is recursively decomposed into sub-tasks, solved by sub-methods is stored in a (conceptual) library of reusable PSMs A piece of code: written in any language (CLIPS, C++, Java) applied to a KB stored in any back-end format (CLIPS, DB, XML) run externally or within a Java tab plug-in Examples: Propose&Revise constraint satisfaction (elevator configuration) board-game method (room assignment) EON guideline interpreter (e.g. hypertension guidelines) Eligibility Screening Tab (breast-cancer protocols)

5 Designing a PSM: Important steps Gennari et al Method Selection Task Analysis Method Configuration

6 Task analysis Define real-world activity accomplished by the KBS Specification of a functional goal: inputs and expected outputs of problem-solving contextual requirements, e.g. time, user interaction, flexibility availability and form of domain knowledge Identification of dimensions for method selection: expressed in domain space of discourse formal vs. informal problem characteristics

7 Method selection Find a PSM to realize the task, that: matches task specification takes advantage of domain knowledge characteristics Indexing criteria: task requirements (solution quality, automation level) I/O specification (information at run-time / part of task definition) domain knowledge type and availability (expert reasoning) method properties (computational & space complexity) task and method spaces cognitive distance in representations of problem and knowledge method flexibility w.r.t. task specifics

8 Method configuration Adapt PSM to specific application problem Mapping PSM and domain spaces: identify domain knowledge that fulfils PSM I/O roles identify transformation relations acquire extra domain knowledge if needed Composing complex PSMs from sub-methods: identify sub-tasks of PSM select available sub-components that solve the sub-tasks linkto sub-components inputs and outputs Refining PSMs to fit task-specific needs: narrow scope of PSM with additional ontological commitments specialize behaviour of PSM with alternative sub-components

9 Example Task analysis: elevator configuration (Sisyphus II) set of building specifications and requirements (speed, capacity) find a configuration of elevator components so that no constraints are violated large body of knowledge (elevator components, safety constraints) Method selection: propose-and-revise constraint satisfaction set of parameters and run-time inputs set of constraints that must be satisfied set of fixes to correct violated constraints decomposed in select, propose, verify, revise sub-methods Method configuration: mapping range constraints for task parameters to method fixes using a revise method that fixes constraints locally or globally

10 Building a KBS in Protégé-2000 Domain ontology Protégé 2000 Domain-specific KA tool PSM KBS PSM library KB Adapted from Eriksson et al Mapping relations PSM selection PSM description PSM configuration PSM

11 A PSM description ontology A semi-formal language to describe PSMs uniformly: unifies experience in KA and SE: Unified Problem-Solving Method Description Language (UPML, IBROW 3 project, Fensel et al ) Initial method description language (Gennari et al. 1998) makes knowledge and know-how about PSM components explicit enables PSM indexing, retrieval and integration provides domain experts with access to libraries of reusable PSMs Concepts to describe PSM properties and interface: Nature: Complex or Primitive Pragmatics: author, references, success history Method Ontology: input/output interface and terms Competence: goal, input/output conditions, sub-tasks Operational specification: inference structure, knowledge roles Related concepts: Tasks, Domain Model, Bridges, Refiners

12

13 A PSM Tab for informed PSM selection A Protégé Tab: takes advantage of Protégé API and UI (forms, diagram widget) interfaces a domain KB with external PSM KBs provides improved navigation through PSM instances Browsing PSM libraries from a domain ontology: view PSM descriptions in a guided way (with +/- details) focus attention on function and requirements of PSMs Expressing queries centered on the application problem: allow to use task specification as a guide for selection suggest related PSMs, appropriate to domain requirements Annotating PSM descriptions with practical feedback: add pragmatic information (see also Gennari & Ackerman, 1999) indicate successful vs. ineffectual application cases

14 PSM Tab: Preliminary version PSM Tab in Protégé Available KBs of PSMs PSMs in selected KB Pragmatic properties of selected PSM Reference annotations Usage feedback

15 Competence (I/O roles, conditions, sub-tasks) Operational Description (control, knowledge roles) Capabilities of selected PSM

16 PSM configuration tools Mapping PSM requirements to domain concepts: use the method ontology, that explicits I/O interface identify mapping relations between PSM and domain ontologies (Gennari et al. 1994, Park&Musen 1998) ontology of mapping relations (from simple renaming to complex combination or addition of knowledge) mapping interpreter generates a “virtual knowledge base” use typical cases of applying PSMs to suggest mapping Adapting PSM for the application at hand: replace sub-components, via PSM-task bridges that map sub- methods together in complex PSMs define specific refinements of PSMs via PSM refiners (Fensel et al. 1998) to solve domain task by specialization (Eriksson et al. 1995) to match a class of tasks

17 Work in progress… Next steps ( Test the PSM ontology: create PSM instances search propose&revise, cover&differentiate, board-game method, chronological backtracking, etc. Protean and other methods used in molecular biology identify missing or redundant distinctions Improve the PSM Tab: enhance browsing relationships between different PSMs, or PSMs and tasks method input/output ontology, conditions on knowledge roles develop query possibilities focus editing for annotation Integrating mapping tool support

18 Conclusions KBS development is a continuous modeling activity: revises task analysis as more insight is gained on domain/methods leads to the refinement of generic PSMs defines the scope and use of domain knowledge PSMs are components which are complex to describe: integrate modeling of declarative and procedural knowledge must explicit an interface to map to domain knowledge clearly defined in a PSM ontology PSMs can be reused more easily if: cognitive distance from method to task space is minimal underlying conceptual model is understandable mapping relations are explicit and principled Protégé-2000 enables to handle PSMs in those respects

19 References: PSM libraries, reuse, mapping Eriksson, H., Shahar, Y., Tu, S.W., Puerta, A.R., and Musen, M.A. (1995). Task modeling with reusable problem-solving methods. Artificial Intelligence 79. SMI Gennari, J.H., Tu, S.W., Rothenfluh, T.E. and Musen, M.A. (1994). Mapping domains to methods in support of reuse. International Journal of Human-Computer Studies 41. SMI Gennari, J.H., Altman, R.B. and Musen, M.A. (1995). Reuse with Protege-II: From Elevators to Ribosomes. ACM-SigSoft 1995 Symposium on Software Reusability, Seattle. SMI Gennari, J.H., Stein, A.R., and Musen, M.A. (1996). Reuse for Knowledge-Based Systems and CORBA Components. In: 10th Knowledge Acquisition for Knowledge-Based Systems Workshop, Banff, Canada. SMI Park, J.Y., Gennari, J.H. and Musen, M.A. (1998). Mappings for Reuse in Knowledge- Based Systems. In: 11th Workshop on Knowledge Acquisition, Modeling, and Management, Banff, Canada. SMI Park, J.Y., and Musen, M.A (1998). VM-in-Protege: A Study of Software Reuse. In: Medinfo '98: Ninth World Congress on Medical Informatics, Seoul, South Korea, IOS Press. SMI

20 References: Method ontology, UPML Gennari, J.H., Grosso, W.E., and Musen, M.A. (1998). A Method-Description Language: An initial ontology with examples. In: 11th Workshop on Knowledge Acquisition, Modeling, and Management, Banff, Canada. SMI Fensel, D., and Motta, E. (1998). Structured Development of Problem Solving Methods. In: 11th Workshop on Knowledge Acquisition, Modeling, and Management, Banff, Canada. ksi.cpsc.ucalgary.ca/KAW/KAW.html Fensel, D., Benjamins, V.R., Motta, E., and Wielinga, B. (1999). UPML: A Framework for knowledge system reuse. In: Proceedings of the International Joint Conference on AI (IJCAI-99), Stockholm, Sweden. Fensel, D., Benjamins, V.R., Decker, S., Gaspari, M., Groenboom, R., Grosso, W., Motta, E., Musen, M.A., Plaza, E., Schreiber, G., Studer, R., and Wielinga, R. (1999). The Unified Problem-solving Method Development Language, UPML. Benjamins, V.R., Plaza, E., Motta, E., Fensel, D., Studer, R., Wielinga, B., Schreiber, G., Zdrahal, Z., and Decker, S. (1998). IBROW3: An intelligent brokering service for knowledge-component reuse on the World-Wide Web. In: 11th Workshop on Knowledge Acquisition, Modeling, and Management, Banff, Canada. Gennari, J.H., and Ackerman, M. (1999), Extra-Technical Information for Method Libraries. In: Proceedings of the 12th Workshop on Knowledge Acquisition, Modeling and Management, Banff, Canada. ksi.cpsc.ucalgary.ca/KAW/KAW.html