More Publications (20)

1Answering Yes/No Questions via Question Inversionis the topranked answer, the answer with the highest confidence value. Our hypothesis is that if an original proposition ...

2Boosting Relation Extraction with Limited ClosedWorld Knowledgepresent a new approach to estimating or ranking the confidence value of learned rules by utilizing limited closedworld knowledge...

3Collaboration On Reference To Objects That Are Not Mutually Known1994  Philip G. Edmondsreferent. hl otrr model, each agerlt associates a numeric confidence value with each of tile attributes in the relSrring expression...

4Learning to Select a Good Translation2000  Dan Tidhar,Uwe Kiissnerncomparab i l i ty Eat;h translation lnodule calculates a confidence value tbr each translation that it; produces. However, since...

5An NPCluster Based Approach to Coreference Resolutionthen presented to the decision tree, which returns a confidence value (CF)2 indicating the likelihood that NPi is coreferential...

6Towards a Unified Approach to Simultaneous SingleDocument and MultiDocument Summarizations2010  Xiaojun Wantrust them so much as the specified document. Thus a confidence value is used to reflect out belief that the document is...

7Active Learning for Chinese Word Segmentationnearest to 0.5. Therefore, we can define the uncertainty confidence value as follows: {0,1} ( ) max (  ) 0.5Un k kyb P y Iφ...

8EndtoEnd Coreference Resolution via Hypergraph Partitioning2010  Jie Cai,Michael Strubea pair of mentions is coreferent or which outputs a confidence value, and (2) a clustering step which groups mentions into...

9CoSTAR: A Cotraining Style Algorithm for Hyponymy Relation Acquisition from Structured and Unstructured Text(not a hyponymy relation)) to instances in x ∈ X with confidence value r ∈ R+, a nonnegative real number. We denote the cl...

10Exploiting Salient Patterns for Question Detection and Question Retrieval in Communitybased Question Answering2010  Kai Wang,TatSeng Chuafor syntactic patterns with a step size of 0.01%. The confidence value for both patterns ranges from 60% to 80% with a step...

11A Global Relaxation Labeling Approach to Coreference Resolutionlinking an anaphor only to its antecedent with maximum confidence value (Ng and Cardie, 2002). Some works propose more elaborated...

12A Transitive Model for Extracting Translation Equivalents of Web Queries through Anchor Text MiningTherefore, a candidate translation has a higher confidence value to be an effective translation if it frequently cooccurs...

13Event Relations At The Phonetics/Phonology Interface(property), start of interval, end of interval and confidence value. This serves as input to event mapping. The output...

14Information Extraction for Question Answering: Improving Recall Through Syntactic Patternsperson (e.g., occupation, position, activities), the confidence value assigned by the extraction modules to this information...

15A MultipleDomain Ontology Builder2010  Sara Salem,Samir AbdelRahmanProbabilistic Ontology Model (POM), which assigns a confidence value for each learnt element. OntoLT (Buitelaar et al...

16Imbalanced Classification Using Dictionarybased Prototypes and Hierarchical Decision Rules for Entity Sense Disambiguationconfident class label to the query instance xˆi, with the confidence value indicated by the value of separating function: C2(xˆi)...

17Summarization of BusinessRelated Tweets: A ConceptBased Approach2012  Annie Louis,Todd Newmanis the absolute difference in positive and negative confidence value from the classifier. This score indicates the degree...

18Improving Alignments for Better Confusion Networks for Combining Machine Translation Systemsbetween the synonyms of the same word according to their confidence value or their POS tag. 35 ONEPASS ALIGNMENT khaddam...

19Coreference Systems Based on Kernels Methodsthe classifier which then returns a class label with a confidence value indicating the likelihood that the candidate is the...

20MultiDimensional Feature Merger for Question Answeringtechniques to weigh and combine features to produce a single confidence value estimating the probability that the candidate answer...