1 / 10

Fine Grain Entities Recognition

Natural Language Processing. Fine Grain Entities Recognition. Background Information. What about? Machine recognition of English words Identify words of interest Why important? Data mining Deploying internet ads, improving automated searching results etc. Basis for future NLP projects.

yetta
Télécharger la présentation

Fine Grain Entities Recognition

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. Natural Language Processing Fine Grain Entities Recognition

  2. Background Information • What about? • Machine recognition of English words • Identify words of interest • Why important? • Data mining • Deploying internet ads, improving automated searching results etc. • Basis for future NLP projects.

  3. What we want: Existing data: Find relations: President->Politician Politician->Person River->Location • George Washington->President, Politician, Activist • Abraham Lincoln-> President, Legislator, Politician • Mississippi River->River, Location

  4. Principle Goal 1: • Given a context-free word, return all the possible related categories.

  5. Principle Goal 2: • Achieve this goal on large data sets efficiently.

  6. Approach 1: Conditional Probability • If the data suggests that P(Politician|President)>0.9 (or other pre-determined number) • Then we can conclude that any President is a Politician • Calculation:

  7. Approach 2: Random Walk • 6,000 • 3,000 • Activist • Senator

  8. Synonyms • Strongly connect components in the constructed graph can be treated as synonyms

  9. Future work: • Extend the dataset to Wikipedia or the Web • Move from context-independent to context dependent: • Solve ambiguity in natural language • E.g “Michael Jordan” might refer to the basketball player or the Berkeley scholar.

  10. Thank you for attending! • Special Thanks for my mentor: Lev Ratinov • My name: Xiao Cheng • E-mail: cheng88@illinois.edu

More Related