Source PaperYearLineSentence
P08-1004 2008 105
The CRF is then used to labelinstances relations for each possible entity pair, sub ject to the constraints mentioned previously.Following extraction, O-CRF applies the RE SOLVER algorithm (Yates and Etzioni, 2007) to findrelation synonyms, the various ways in which a relation is expressed in text (self citation)
W10-0911 2010 88
ShopBot learns comparison shopping agents via self-supervision using heuristic knowledge (Doorenbos et al, 1997); WIEN induces wrappers for information extraction via self-supervision using joint inference to combine simple atomic extractors (Kushmerick et al, 1997); Mulder answers factoid questions by leveraging redundancy to rank candidate answers extracted frommultiple search query results (Kwok et al, 2001); KnowItAll conducts open-domain information extraction via self supervision bootstrapping from Hearst patterns (Etzioni et al, 2005); Opine builds on KnowItAll and mines productreviews via self-supervision using joint inference over neighborhood features (Popescu and Etzioni, 2005); Kylin pop ulates Wikipedia infoboxes via self-supervision bootstrapping from existing infoboxes (Wu and Weld, 2007); LEX conducts Web-scale name entity recognition by leveraging collocation statistics (Downey et al, 2007a); REALM improves sparse open-domain information extraction via relational clustering and language modeling (Downey et al,2007b); RESOLVER performs entity and relation resolution via relational clustering (Yates and Etzioni, 2007); Tex tRunner conducts open-domain information extraction via self-supervision bootstrapping from heuristic rules (Bankoet al, 2007); AuContraire automatically identifies contradictory statements in a large web corpus using functional re lations (Ritter et al, 2008); HOLMES infers new facts from TextRunner output using Markov logic (Schoenmackers et al, 2008); KOG learns a rich ontology by combining Wikipedia infoboxes with WordNet via joint inference using Markov Logic Networks (Wu and Weld, 2008), shrinkage over this ontology vastly improves the recall of Kylin?s extractors; UCR performs state-of-the-art unsupervised coreference resolution by incorporating a small amount of domain knowledge and conducting joint inference among entity mentions with Markov logic (Poon and Domingos, 2008b); SNE constructs a semantic network over TextRunner output via relational clustering with Markov logic (Kok and Domingos, 2008); WebTables conducts Web-scale information extraction by leveraging HTML table structures (Cafarella et al, 2008); IIA learns from infoboxes to filter open-domain information extraction toward assertions that are interesting to people (Lin et al, 2009); USP jointly learns a semantic parser and extracts knowledge via recursiverelational clustering with Markov logic (Poon and Domingos, 2009); LDA-SP automatically infers a compact repre sentation describing the plausible arguments for a relation using an LDA-Style model and Bayesian Inference (Ritter et al, 2010); LOFT builds on USP and jointly performs ontology induction, population, and knowledge extraction via joint recursive relational clustering and shrinkage with Markov logic (Poon and Domingos, 2010); OLPI improves the efficiency of lifted probabilistic inference and learning via coarse-to-fine inference based on type hierarchies (Kiddon and Domingos, 2010) (self citation)
D10-1106 2010 46
Two other notable systems that learn inference rules from text are DIRT (Lin and Pantel, 2001)and RESOLVER (Yates and Etzioni, 2007) (self citation)
P11-1058 2011 65
Synonym resolution on relations extracted from web text has been previously studied by Resolver (Yates and Etzioni, 2007), which finds synonyms in relation triples extracted by TextRunner (Banko etal., 2007)
P11-1058 2011 31
Like many other systems(Miller, 1995; Yates and Etzioni, 2007; Lin and Pan tel, 2002), ConceptResolver represents each output concept ci as a set of synonymous noun phrases, i.e., ci = {xi1, xi2, ..., xim}
P11-1058 2011 230
We also adopt the precision/recall measure from Resolver(Yates and Etzioni, 2007), which we dub the Resolver metric
D12-1094 2012 51
Resolver (Yates and Etzioni, 2007) resolves objects and relation synonyms
P12-1037 2012 49
DIRT (Lin and Pantel, 2001) and RESOLVER (Yates and Etzioni, 2007) learn inference rules, alsocalled entailment rules that capture synonymous re lations and entities from text