Detection of incorrect case assignments in paraphrase generation

Atsushi Fujita, Kentaro Inui, Yuji Matsumoto

Research output: Contribution to journalConference articlepeer-review


This paper addresses the issue of post-transfer process in paraphrasing. Our previous investigation into transfer errors revealed that case assignment tends to be incorrect, irrespective of the types of transfer in lexical and structural paraphrasing of Japanese sentences [3]. Motivated by this observation, we propose an empirical method to detect incorrect case assignments. Our error detection model combines two error detection models that are separately trained on a large collection of positive examples and a small collection of manually labeled negative examples. Experimental results show that our combined model significantly enhances the baseline model which is trained only on positive examples. We also propose a selective sampling scheme to reduce the cost of collecting negative examples, and confirm the effectiveness in the error detection task.

Original languageEnglish
Pages (from-to)555-565
Number of pages11
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publication statusPublished - 2005
EventFirst International Joint Conference on Natural Language Processing - IJCNLP 2004 - Hainan Island, China
Duration: 2004 Mar 222004 Mar 24


Dive into the research topics of 'Detection of incorrect case assignments in paraphrase generation'. Together they form a unique fingerprint.

Cite this