Maria Vargas-Vera, E.Motta, J. Domingue, S. Buckingham Shum and M. Lanzoni - PowerPoint PPT Presentation

maria vargas vera e motta j domingue s buckingham shum and m lanzoni l.
Skip this Video
Loading SlideShow in 5 Seconds..
Maria Vargas-Vera, E.Motta, J. Domingue, S. Buckingham Shum and M. Lanzoni PowerPoint Presentation
Maria Vargas-Vera, E.Motta, J. Domingue, S. Buckingham Shum and M. Lanzoni

play fullscreen
1 / 15
Download Presentation
Download Presentation

Maria Vargas-Vera, E.Motta, J. Domingue, S. Buckingham Shum and M. Lanzoni

Presentation Transcript

  1. Knowledge Extraction by using an Ontology-based Annotation Tool Maria Vargas-Vera, E.Motta, J. Domingue, S. Buckingham Shum and M. Lanzoni Knowledge Media Institute(KMi) The Open University Milton Keynes, MK7 6AA October 2001

  2. Outline • Motivation • Extraction of knowledge structures from web pages • Final goal -Ontology population • Approaches to semantic annotation of web pages (SAW) • OntoAnnotate [Stab, et al] • SHOE [Hendler et al] • Our solution to SAW problem • Ontology driven annotation • Work so far - we had tried with two different domains (KMi stories and Rental adverts) • Conclusions and Future work

  3. Our system • Our system consists of 4 phases: • Browse • browser selection • Mark-up phase (mark-up text in training set) • Learning phase (learns rules from training set) • Extraction phase (extracts information from a document)

  4. Mark-up phase • Ontology-based Mark-up • The user is presented with a set of tags (taken from ontology) • user selects slots-names for tagging. • Instances are tagged by the user

  5. EVENT 1: visiting-a-place-or-people visitor (list of person(s)) people-or-organisation-being-visited (list of person(s) or organisation) has-duration (duration) start-time (time-point) end-time (time-point) has-location (a place) other agents-involved (list of person (s)) main-agent (list of person (s))

  6. Learning phase • Learning phase was Implemented using Marmot and Crystal. • Mark-up all instances in the training set • Marmot performs segmentation of a sentence: noun phrases,verbs and prepositional phrases. • Example: “David Brown, the Chairman of the University for Industry Design and Implementation Advisory Group and Chairman of Motorola, visited the OU”. • Marmot output: • SUBJ: DAVID BROWN %comma% THE CHAIRMAN OF THE UNIVERSITY • PP: FOR INDUSTRY DESIGN AND IMPLEMENTATION ADVISORY GROUP AND CHAIRMAN OF MOTOROLA • PUNC: %COMMA% • VB: VISITED • OBJ: THE OU

  7. Learning phase (cont) • Crystal derives a set of patterns from a training corpus. • Example of Rule generated using Crystal. • Conceptual Node for visiting-a-place-or-people event: • Verb: visited (active verb) (trigger word) • Visitor: V (person) • Has-location: P (place) • Start-time: ST (time-point) • End-time: ET (time-point) • Example of patterns: • X visited Y on the date Z • X has been awarded Y money from Z

  8. Extraction phase • Badger makes instantiation of templates. • In our example (David’s Brown story), Badger instanciates the following slots of a Event -1 frame: • Type: visiting-a-pace-or-people • Place: The OU • Visitor: David Brown

  9. OCML code(definition of an instance of class visiting-a-place-or-people) (Def-instance visit-of-david-brown-the-chairman-of-the-university visiting-a-place-or-people ((start-time wed-15-oct-1997) (end-time wed-15-oct-1997) (has-location the-ou) (visitor david-brown-the-chairman-of-the-university) ) )

  10. Populating the ontology • David Brown’s story output after the OCML code is sent to Webonto.

  11. Library of IE Methods • Currently our library contains methods for learning: • Crystal (bottom-up learning algorithm) • Whisk (top-down learning algorithm) • We plan to extend the library with other methods besides Crystal and Whisk.

  12. Whisk (second tool for learning) • Whisk: learns information extraction rules • can be applied to semi-structured text (text is un-gramatical, telegraphic). • can be applied to free text (syntactically parsed text). • It uses a top-down induction algorithm seeded by a specific training example. • Whisk has been used: • CNN weather forecast in HTML • BigBook addresses in HTML • Rental ads in HTML (our second domain) • Seminar announcements • job posting • Management succession text from MUC-6

  13. Sample Rule from Rental domain • Domain Rental Adverts: • Ballard - 2 Br/2 Ba, top flr, d/w 1000 sf, $820. (206) 782-2843. • Rule expressed as regular expression: • ID 26 Pattern:: * (Nghbr) * (<digit>) ‘Br’ * ‘$’ (<number>). • Output:: Rental{Neighbourhood $1} {Bedrooms $2} {Price $3}

  14. Whisk example (continuation) • Items in green colour are semantic word classes. • Nghbr :: Ballard | Belltown| … • digit :: 1|2|…|9 • number :: (0-9)* • Complexity : restricted wild card therefore, time is not exponential.

  15. Conclusions and Future Work • We had built a tool which extracts knowledge using and Ontology, IE component and OCML pre-processor. • We had worked with 2 different domains (KMi stories and Rental adverts) • first domain • Precision over 95% • second domain • Precision: 86% - 94% • Recall: 85% - 90% • We will integrate more IE methods in our system. • To extend our system in order to produce XML output, RDFS,… • to integrate visualisation capabilities