- Количество слайдов: 11
Open Information Extraction From The Web Rani Qumsiyeh
What is Information Extraction o This article surveys a range of Information Extraction methods. (Particularly Open) o A venerable technology that maps natural language text into structured relational data. o Open Information Extraction is where the identities of the relations to be extracted are unknown and the billions of documents found on the Web necessitate highly scalable processing.
Most common Ways to do IE o Direct knowledge-based encoding. n A human enters regular expressions or rules. o Supervised learning. n A human provides labeled training examples. o Self-supervised learning. n The system automatically finds and labels its own examples.
Direct Knowledge o Not efficient, has to be altered for different domains. o Class Physical. Target space to the term bank in the terrorism domain. o Class Corporation in the joint-ventures domain
Example of Supervised Learning
Self Supervised Knowledge o A system that labels its own training examples. (Example: Know. It. All) o For a given relation n Use generic pattern instantiate relationspecific extraction rules learn domainspecific extraction rules apply rules to web pages and assign them probabilities. o Example: X is a Y (X is a country). n China is a country. n Garth Brooks is a country singer
Open Information Extraction o The challenge of Web extraction is to be able to do Open Information Extraction. n n Unbounded number of relations Web corpus contains billions of documents.
How open IE systems work o learn a general model of how relations are expressed (in a particular language), based on unlexicalized features such as part-of-speech tags. (Identify a verb) o Learn domain-independent regular expressions. (Punctuations, Commas).
Is there a general model of relationships in English
Text. Runner o Works in two phases. 1. 2. Using a conditional random field, the extractor learns to assign labels to each of the words in a sentence. Extracts one or more textual triples that aim to capture (some of) the relationships in each sentence.
Additional Tasks to Accomplish o Opinion mining: in which Open IE can extract opinion information about particular objects (including products, political candidates, and more) that are contained in blog, posts, reviews, and other texts. o Fact checking: in which Open IE can identify assertions that directly or indirectly conflict with the body of knowledge extracted from the Web and various other knowledge bases.