1 / 12

Yannis Kalfoglou, John Domingue, Enrico Motta, Maria Vargas-Vera, Simon Buckingham Shum The Open University

myPlanet : an ontology-driven Web-based personalised news service. Yannis Kalfoglou, John Domingue, Enrico Motta, Maria Vargas-Vera, Simon Buckingham Shum The Open University. Presented by Sang Su Lee. Figures liberally borrowed from paper. Need for the personalised information service.

Sharon_Dale
Download Presentation

Yannis Kalfoglou, John Domingue, Enrico Motta, Maria Vargas-Vera, Simon Buckingham Shum The Open University

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. myPlanet: an ontology-driven Web-based personalised news service Yannis Kalfoglou, John Domingue, Enrico Motta, Maria Vargas-Vera, Simon Buckingham Shum The Open University Presented by Sang Su Lee Figures liberally borrowed from paper

  2. Need for the personalised information service Research goals • Finding interesting stories for himself/herself is non-trivial. • Can be fulfilled by part of an ontology that is tailored to a specific service. • Using lowest-common-denominator medium: an email message. • To improve and ease ontology usability for web users by means of ontology • driven web-based front-ends to personalised services • To provide lightweight means for ontology maintenance triggered by users’ • input by deploying IE techniques along with domain specific templates. Introduction myPlanet: an ontology-driven Web-based personalised news service

  3. User Types Activities What is PlanetOnto? PlanetOnto myPlanet: an ontology-driven Web-based personalised news service - A kind of news publishing system • An integrated suite of tools • Allow ontology-driven document formalization • Augment standard browsing and search facilities with deductive knowledge retrieval • Journalist : send stories to KMi Planet • Knowledge editors : maintain Planet ontology and the Planet knowledge base • Readers : read the Planet stories • Story submission • Story reading • Story annotation • Provision of customized alerts • Ontology editing • Story soliciting • Story retrieval and query answering

  4. What is myPlanet? Ontological interest-profiling Populating the ontology myPlanet? myPlanet: an ontology-driven Web-based personalised news service • web based newsletter by creating a profile based on an ontology. • to improve and ease ontology usability for web users by means of ontology-driven front-ends to personalized services. • to provide lightweight means for ontology maintenance triggered by users’ input by deploying IE techniques along with domain specific templates.

  5. Advantage of using ontology Structure of interests Ontological interest-profiling myPlanet: an ontology-driven Web-based personalised news service • read only the e-Stories of their interest • reason about the categories selected by applying ontology-driven deductive heuristics • find implicitly related e-Stories • Research areas that are investigated in KMi • Research themes that are investigated in KMi • Organizations that KMi collborates with • Projects in KMi • Technologies used in KMi • Application domains that are investigated in KMi • People – members of the KMi lab

  6. String matching with cue phrases and cue words Using OCML language Ontological interest-profiling myPlanet: an ontology-driven Web-based personalised news service • OCML supports the construction of knowledge models by means of several types of constructs. • It allows the specification and operationalization of functions, relations, classes and rules. • abstraction of the category that is associated with • evidence that e-Story which contains it is relevant to that category

  7. Example of myPlanet myPlanet: an ontology-driven Web-based personalised news service 1

  8. Example of myPlanet myPlanet: an ontology-driven Web-based personalised news service

  9. Template-driven information extraction Populating the ontology myPlanet: an ontology-driven Web-based personalised news service • domain specific templates • based on typology of events • ex) visiting-a-place-or-people, academic-conference, event-involving-project • key is the trigger word • ex) [ _, X, _, visited, Y, from, Z, _ ]

  10. Example of ontology population myPlanet: an ontology-driven Web-based personalised news service Visiting-a-place-or-people [ _, X, _, visited, Y, from, Z, _ ] Visitor: “AKT collaborating institutions” Place: “Sheffield” Date: January 29-31 2001” (def-instance visit-of-akt-collaborating-institutions visiting-a-place-or-people ((has-duration ‘3 days’) (start-time january-29-2001) (end-time january-31-2001) (has-location sheffield) (visitor akt-collaborating-institutions)))

  11. Extending type of output Extending the usage of IE techniques with domain specific templates Automation by deploying inductive learning algorithms Avoidance of duplications or omissions of information Ontology-based reasoning Future work myPlanet: an ontology-driven Web-based personalised news service • reason about the kind of output that will be dispatched to the user by analysing his/her profile • suggestions about potential collaborators on a research topic

  12. Thank you myPlanet: an ontology-driven Web-based personalised news service

More Related