PPDB: The Paraphrase Database

Proceedings of NAACL-HLT 2013, pages 758–764, Atlanta, Georgia, 9–14 June 2013. c⃝2013 Association for Computational Linguistics PPDB: The Paraphrase Database Juri Ganitkevitch1 Benjamin Van Durme1,2 Chris Callison-Burch2,3 1Center for Language and Speech Processing, Johns Hopkins University 2Human Language Technology Center of Excellence, Johns Hopkins University 3Computer and Information Science Department, University of Pennsylvania Abstract We present the 1.0 release of our para- phrase database, PPDB. Its English portion, PPDB:Eng, contains over 220 million para- phrase pairs, consisting of 73 million phrasal and 8 million lexical paraphrases, as well as 140 million paraphrase patterns, which cap- ture many meaning-preserving syntactic trans- formations. The paraphrases are extracted from bilingual parallel corpora totaling over 100 million sentence pairs and over 2 billion English words. We also release PPDB:Spa, a collection of 196 million Spanish paraphrases. Each paraphrase pair in PPDB contains a set of associated scores, including paraphrase probabilities derived from the bitext data and a variety of monolingual distributional similar- ity scores computed from the Google n-grams and the Annotated Gigaword corpus. Our re- lease includes pruning tools that allow users to determine their own precision/recall tradeoff. 1 Introduction Paraphrases, i.e. differing textual realizations of the same meaning, have proven useful for a wide vari- ety of natural language processing applications. Past paraphrase collections include automatically derived resources like DIRT (Lin and Pantel, 2001), the MSR paraphrase corpus and phrase table (Dolan et al., 2004; Quirk et al., 2004), among others. Although several groups have independently ex- tracted paraphrases using Bannard and Callison- Burch (2005)’s bilingual pivoting technique (see Zhou et al. (2006), Riezler et al. (2007), Snover et al. (2010), among others), there has never been an official release of this resource. In this work, we release version 1.0 of the Para- Phrase DataBase PPDB,1 a collection of ranked En- glish and Spanish paraphrases derived by: • Extracting lexical, phrasal, and syntactic para- phrases from large bilingual parallel corpora (with associated paraphrase probabilities). • Computing distributional similarity scores for each of the paraphrases using the Google n- grams and the Annotated Gigaword corpus. In addition to the paraphrase collection itself, we provide tools to filter PPDB to only retain high pre- cision paraphrases, scripts to limit the collection to phrasal or lexical paraphrases (synonyms), and soft- ware that enables users to extract paraphrases for languages other than English. 2 Extracting Paraphrases from Bitexts To extract paraphrases we follow Bannard and Callison-Burch (2005)’s bilingual pivoting method. The intuition is that two English strings e1 and e2 that translate to the same foreign string f can be as- sumed to have the same meaning. We can thus pivot over f and extract ⟨e1, e2⟩ as a pair of paraphrases, as illustrated in Figure 1. The method extracts a di- verse set of paraphrases. For thrown into jail, it ex- tracts arrested, detained, imprisoned, incarcerated, jailed, locked up, taken into custody, and thrown into prison, along with a set of incorrect/noisy para- phrases that have different syntactic types or that are due to misalignments. …

Content truncated to 500 words