MIPA: Mutual Information Based Paraphrase Acquisition via Bilingual Pivoting

Tomoyuki Kajiwara, Mamoru Komachi, Daichi Mochihashi


Abstract
We present a pointwise mutual information (PMI)-based approach to formalize paraphrasability and propose a variant of PMI, called MIPA, for the paraphrase acquisition. Our paraphrase acquisition method first acquires lexical paraphrase pairs by bilingual pivoting and then reranks them by PMI and distributional similarity. The complementary nature of information from bilingual corpora and from monolingual corpora makes the proposed method robust. Experimental results show that the proposed method substantially outperforms bilingual pivoting and distributional similarity themselves in terms of metrics such as MRR, MAP, coverage, and Spearman’s correlation.
Anthology ID:
I17-1009
Volume:
Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
November
Year:
2017
Address:
Taipei, Taiwan
Venue:
IJCNLP
SIG:
Publisher:
Asian Federation of Natural Language Processing
Note:
Pages:
80–89
Language:
URL:
https://aclanthology.org/I17-1009
DOI:
Bibkey:
Cite (ACL):
Tomoyuki Kajiwara, Mamoru Komachi, and Daichi Mochihashi. 2017. MIPA: Mutual Information Based Paraphrase Acquisition via Bilingual Pivoting. In Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 80–89, Taipei, Taiwan. Asian Federation of Natural Language Processing.
Cite (Informal):
MIPA: Mutual Information Based Paraphrase Acquisition via Bilingual Pivoting (Kajiwara et al., IJCNLP 2017)
Copy Citation:
PDF:
https://aclanthology.org/I17-1009.pdf
Code
 tmu-nlp/pmi-ppdb