Paper: Paraphrase Recognition Via Dissimilarity Significance Classification

ACL ID W06-1603
Title Paraphrase Recognition Via Dissimilarity Significance Classification
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2006
Authors

We propose a supervised, two-phase framework to address the problem of para- phrase recognition (PR). Unlike most PR systems that focus on sentence similarity, our framework detects dissimilarities be- tween sentences and makes its paraphrase judgment based on the signi cance of such dissimilarities. The ability to differenti- ate signi cant dissimilarities not only re- veals what makes two sentences a non- paraphrase, but also helps to recall addi- tional paraphrases that contain extra but insigni cant information. Experimental results show that while being accurate at discerning non-paraphrasing dissimilar- ities, our implemented system is able to achieve higher paraphrase recall (93%), at an overall performance comparable to the alternatives.