KnowItAll System

AKA: KnowItAll, KnowItAll Web IE System.
Context:
- It implements the KnowItAll Algorithm, a Semi-Supervised Named Entity Recognition Algorithm.
- It is a Knowledge Base that populates its facts, concepts, and relationships by extracting the information from the web.
- It is intended to be scalability and have high throughput in order to access the information on the Web.
  - It avoids the use of deep parsing techniques.
- It is divided into four components:
  - Extractor,
  - Search Engine Interface,
  - Assessor,
  - Database.
- It requires a set of generic Extraction Pattern seeds (not Instance seeds).
- It includes a predefined set of topics (ontology) such as:
  - cities, states, countries, actors and films.
- It is developed at the University of Washington.
- It is based on the KnowItAll Hypothesis that: "Extractions drawn more frequently from distinct sentences in a corpus are more likely to be correct.
Counter-Example(s):
- TextRunner System.
- Opine System.
See: Open Information Extraction. ReVerb System.

References