D12-1066 engineering new features to improve paraphrase recognition . Lastly , we note that precision
J10-3003 Sekine ( 2006 ) shows how to use paraphrase recognition to cluster together extraction
J10-3003 from this task . In general , paraphrase recognition can be very helpful for several
D15-1185 of methods can help solve the paraphrase recognition problem , which is a subset of
J10-3003 available test set to evaluate paraphrase recognition methods . On a related note ,
J09-2001 information could then be used for paraphrase recognition . Prepositions have deservedly
I05-5002 its use as a tool for evaluating paraphrase recognition algorithms . It has already been
J10-3003 generation applications rely heavily on paraphrase recognition . For a multi-document summarization
D09-1083 TWEAK in other problems such as paraphrase recognition and near - duplicate detection
A83-1002 grammar , permits easy and natural paraphrase recognition , although there are linguistic
C04-1051 techniques to the problems of paraphrase recognition and generation . We feel that
J10-3003 briefly describes the tasks of paraphrase recognition and textual entailment and their
I05-5001 4.1 Methodology Evaluation of paraphrase recognition within an SMT framework is highly
E12-1036 tasks of textual entailment and paraphrase recognition , see Androutsopoulos and Malakasiotis
J13-3001 various alternatives , an automated paraphrase recognition mechanism would be useful . One
J10-3003 , in this sense , the task of paraphrase recognition can simply be formulated as bidirectional
D12-1066 exploiting a large set of features for paraphrase recognition . A detailed quantified typology
D10-1090 evaluation , summary evaluation , and paraphrase recognition . To facilitate future research
J10-3003 Iftene 2009 ) and might even use paraphrase recognition to improve their performance
J10-3003 techniques for the purpose of paraphrase recognition ( Rus , McCarthy , and Lintean
hide detail