OSLC-KM

A Knowledge Management specification for OSLC-based resources

Jose María Alvarez-Rodríguez, Juan Llorens, Manuela Alejandres and Jose Fuentes

INCOSE IS 2015, Seattle, US

Knowledge Management

in a nutshell

Adapted from: S. McIntyre, M. Gauvin, and B. Waruszynski, “Knowledge management in the military context,” Can. Mil. J., vol. 4, no. 1, pp. 35–40, 2003.

Representation

 


Services

COMMON

Reuse

 


Human-centric

Natural language

Major objectives

Representation

  • Formats
  • Protocols
  • Query Languages
  • Objectives
  • Budgets
  • ...

...a text?

E.g. A requirement

"When the voltage level is below 11.5v the battery shall send 'a low battery load level' message to the power control systems"

...an UML diagram?

...a Modelica physical system model?

...an E/R model?

...a PBS?

...a graph?

...even a person?

How many different types of artifacts are generated during the development life-cycle?

tools, formats, protocols, query languages,  etc.

Needs and Challenges

  • A common representation model
  • Interoperability
  • Knowledge Management processes
  • Natural language to express queries
  • ...

A remainder...

Natural language

as main

communication mechanism

#me also speak SQL, SPARQL, XPATH, etc...

Services

  • Semantic Indexing and Retrieval
  • Quality checking
  • Visualization
  • Traceability
  • ...

Semantic indexing and retrieval?

"Sine voltage source connected to a potentiometer (variable resistor) before a RC filter"

Results but types of artifacts?

...quality?

E.g. Requirements

...and for any artifact?

...visualization?

Specific

Generic

...traceability?

E.g. Requirements and Models

...and for any artifact?

Needs and Challenges

  • A common representation model
  • Interoperability
  • Knowledge Management processes
  • Natural language to express queries
  • ...

KCSE

Knowledge-centric Systems Engineering

...and now...

Interoperability

Open Services for Life-cycle Collaboration

OSLC

OSLC 

Prelimnary Evaluation

  • Great effort on interoperability
  • Community (industry) effort
  • Different types of artifacts (resource shapes)
  • OASIS standards (Core, Change Management, etc.)
  • ...

...still work to do...

  • Define new domains, resource shapes, etc.
  • Common services
  • Perspective of knowledge management (beyond data exchange)
  • ...

Wrap-up

  • Knowledge Management is a key activity in any organization
  • It requires:
    • A common representation model
    • A set of common of services
    • Interoperability
    • Process different types of artifacts
    • ...
  • In the SE domain, this perspective is becoming crucial
  • Upcoming domains and resource shapes in OSLC

OSLC-KM

New domain

New resource shape

Knowledge Representation

  • Different schemas
  • ...paradigms
  • ...purposes
  • ...execution engines

OSLC Context

Semantic Web Stack 2005

Semantic Web Stack 2015

RDF

Resource Description Framework

  • Common & shared Data model
  • Triples
    • (subject, object, predicate)
  • Binary relationships
  • Underlying Directed Graph
  • W3C Recommendation (2004)
  • Query languages

RDFS

RDF Schema

  • Data modeling for RDF data
  • Classes
  • Properties (domain & ranges)
  • RDF serialization
  • W3C Recommendation (2004)

OWL

Ontology Web Language

  • Vocabulary for defining formal ontologies
  • Logic-oriented
  • RDF serialization
  • Flavours (2.0): EL, QL, RL
  • W3C Recommendation (2012, v2.0)

RIF

Rule Interchange Format

  • Exchange of business rules
  • Rule-oriented
  • Flavours: Core, PRD, BLD, etc.
  • XML serialization
  • W3C Recommendation (2013, v2.0)

RSHP

Relationship "arship"

  • Property Graph
  • Any kind of relationship
    • Arity and Cardinality
  • Industry-oriented
  • Native tool support
  • Queries based on natural language
  • First publication 2004

Others

  • SBVR
    • Semantics of Business Vocabulary and Rules
  • ODM
    • Ontology Definition Metamodel
  • RAS
    • Reusable Asset Specification 

Preliminary Evaluation

  • RDF good option for exchange data
    • Restrictions: arity and cardinality of relationships, lack of native tools, logics, etc.
  • RDFS and OWL
    • Oriented to define formal ontologies
  • RSHP
    • High level of Expressivity
    • Native tool support
    • ...

Approach

OSLC 

(Data Exchange)

 

 

 

RSHP*

(Internal representation: metadata+contents 

and services)

The Specification

 

Shape for KM resources

Metadata+Contents

Mapping to RDF resources

OSLC Overview

Situating contributions...

OSLC Overview

Situated contributions...

Resource Shape

Based on the W3C SKOS Recommendation

Simple Knowledge Organization System

OSLC Core

Provenance*
Access 
(W3C HTTP Access)

Metadata (Dublin Core, traces, etc.)
Contents (resource shape)

Visualization (SVG)

Summary of Properties

Properties

All in one...

What about services?

Functional architecture

Implementation

Knowledge Manager by

The REUSE Company

Vocabulary

Conceptual Model

  • Normalization
  • Standardization
  • Suggestions
  • X Breakdown Structures
  • ...

Patterns

  • Restrictions
  • Examples
  • Suggestions

1-Listing the concepts of a domain vocabulary

2-Standardize a term or text-based description

3-Get synonyms for a term or a text-based description

4-Get the list of patterns 

to guide the writing of requirements

Real Example

Vocabulary through OSLC in CKEditor

On-going work...

Process any kind of OSLC Resource or RDF data...

E.g. Modelica

Summary

  • Use of the W3C Recommendations
  • Concepts and relationships are the entities to be exchanged 
  • Services for: search, trace, naming, visualize, etc.

Metadata+Contents

Artifact

Advantages

  • Standard exchange of data: RDF+OSLC
  • Enhance expressiveness (RSHP)
  • Reuse of existing standards and vocabularies
  • Native Tool support (Knowledge Manager)
  • Cross-cutting services (semantic-based)
  • Elastic approach
  • ...

Drawbacks

  • Scope of knowledge management (needs)
  • Potential overlapping with other tools and specs 
  • Need of spread the approach

Conclusions

  • Knowledge Management: a key process

    • ​REUSE!

  • Need of:
    • Reuse of standards and service-oriented functionalities (OSLC)
    • Take the most for data exchange (RDF) and representation (RSHP)
    • Bring technology to a human-oriented environment (NLP)
  • Technical issues
    • Completeness of the specification?
    • Implementation

Future work

  • Merge and extend the spec with other knowledge standards

    • E.g. STEP, Industrial Internet activities

  • Refine of the resource shape 

  • Full implementation and support to all services

  • Integration patterns governed by knowledge

  • ...

Industrial Linked Data

OSLC applications

  • Semantic Impact Analysis
  • Risk Analysis
  • Continuous Engineering
  • ...

Speakers 

  •  Dr. Jose María Alvarez-Rodríguez
Made with Slides.com