1 / 15

Approach to building ontologies

Approach to building ontologies. A high-level view Chris Wroe. Introduction. Describe key steps in our approach Illustrate with a case study Not a discussion of project management Help inform integration of DL ontology building into wider knowledge base projects. Key steps.

margo
Download Presentation

Approach to building ontologies

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Approach to building ontologies A high-level view Chris Wroe

  2. Introduction • Describe key steps in our approach • Illustrate with a case study • Not a discussion of project management • Help inform integration of DL ontology building into wider knowledge base projects

  3. Key steps • Requirements gathering • Content scoping • Reusing existing components • Construction • Internal testing • Delivery • Evaluation

  4. Requirements gathering • What can a DL based ontology offer and should I use one? • Most people hold misconceptions • Key functions • Organising/ maintaining a large vocabulary within a knowledge base • Integrating vocabularies from several knowledge bases Requirements Scoping Reuse Construction Testing Delivery Evaluation

  5. Case study – a Drug Ontology • Research group builds a knowledge base of prescribing guidelines for specific conditions • KB excludes prescribing ‘common sense’ information. • E.g. ‘Don’t suggest a drug if it will interact with patient’s medication or other conditions’. • Need additional knowledge bases to hold • General drug interactions • General contraindications Requirements Scoping Reuse Construction Testing Delivery Evaluation

  6. Drug Ontology case study - requirements • Require a single vocabulary to integrate the information in each KB in a logically consistent way to support inference • Problems which DL ontologies can address • Vocabulary will be large (1000’s of terms) • Hard to maintain consistently by hand. • Concepts cover a wide range of granularity • Need to be organised in a classification • Concepts are complex • Multiple ways of classifying the same concept Requirements Scoping Reuse Construction Testing Delivery Evaluation

  7. Content scoping • Description Logic Ontology building is descriptive! • Focus taken away from enumeration and manual classification • Determine expected coverage and complexity of concept descriptions required. Requirements Scoping Reuse Construction Testing Delivery Evaluation

  8. Drug Ontology case study - scoping • Sample concepts from each knowledge base. • guideline KB– • if on anti-anginal … • Anti-aginal definition will need to include clinical condition concepts in definition (angina). • Angina definitions will need to include anatomy and physiology concepts in definition (heart, blood flow) Requirements Scoping Reuse Construction Testing Delivery Evaluation

  9. Reusing existing components • Reuse as much as possible especially at the higher levels of the ontology. • Standard upper level ontology • Previously built domain ontologies • Make what you build reusable Requirements Scoping Reuse Construction Testing Delivery Evaluation

  10. Drug Ontology case study - components • Upper level ontology reused • Anatomy and physiology domain ontologies reused and amended. Requirements Scoping Reuse Construction Testing Delivery Evaluation

  11. Construction • Often split into two tasks • Terminology knowledge acquisition • Interacting with domain experts • Terminology knowledge low-level modelling • Expressing knowledge in formal and consistent manner • Use a suite of design patterns and methodologies Requirements Scoping Reuse Construction Testing Delivery Evaluation

  12. Drug Ontology case study – knowledge acquisition • Use an intermediate representation • Simpler, less constrained • Customised to a domain • Authoring tools Requirements Scoping Reuse Construction Testing Delivery Evaluation

  13. Internal testing • What does the logic give you? • Logical consistency checked automatically • Semantic consistency can be assisted by the DL reasoner • By classification – miss-classification • Additional tools • By query and visualisation – missed classification Requirements Scoping Reuse Construction Testing Delivery Evaluation

  14. Case study – internal testing • Pain classed as a nervous system disease • Incorrect definition of pain Requirements Scoping Reuse Construction Testing Delivery Evaluation

  15. Evaluation • Testing of ontology within final application. • Case study - evaluation • Problem integrating existing vocabularies. • Meaning cannot be taken on face value • No human to intervene • Must also explicitly take into account context is which term is used. • Reference material versus Patient record Requirements Scoping Reuse Construction Testing Delivery Evaluation

More Related