Acquiring Synonyms from Monolingual Comparable Texts

Abstract

Abstract. This paper presents a method for acquiring synonyms from monolingual comparable text (MCT). MCT denotes a set of monolingual texts whose contents are similar and can be obtained automatically. Our acquisition method takes advantage of a characteristic of MCT that included words and their relations are confined. Our method uses contextual information of surrounding one word on each side of the target words. To improve acquisition precision, prevention of outside appearance is used. This method has advantages in that it requires only part-ofspeech information and it can acquire infrequent synonyms. We evaluated our method with two kinds of news article data: sentence-aligned parallel texts and document-aligned comparable texts. When applying the former data, our method acquires synonym pairs with 70.0% precision. Re-evaluation of incorrect word pairs with source texts indicates that the method captures the appropriate parts of source texts with 89.5% precision. When applying the latter data, acquisition precision reaches 76.0% in English and 76.3% in Japanese.

Generating Phrasal and Sentential Paraphrases: A Survey of Data-Driven Methods

“…For one, we do not discuss paraphrasing techniques that rely primarily on knowledge-based resources such as dictionaries (Wallis 1993;Fujita et al 2004), handwritten rules (Fujita et al 2007), and formal grammars (McKeown 1979;Dras 1999;Gardent, Amoia, and Jacquey 2004;Gardent and Kow 2005). We also refrain from discussing work on purely lexical paraphrasing which usually comprises various ways to cluster words occurring in similar contexts (Inoue 1991;Crouch and Yang 1992;Pereira, Tishby, and Lee 1993;Grefenstette 1994;Lin 1998;Gasperin et al 2001;Glickman and Dagan 2003;Shimohata and Sumita 2005). 1 Exclusion of general lexical paraphrasing methods obviously implies that other lexical methods developed just for specific applications are also excluded (Bangalore and Rambow 2000;Duclaye, Yvon, and Collin 2003;Murakami and Nasukawa 2004;Kauchak and Barzilay 2006).…”

Combining open-source with research to re-engineer a hands-on introductory NLP course

“…We plan to do so in the context of lexical paraphrase acquisition or synonyms automatically from corpora, a task that relies heavily on this notion of distributional similarity. There has been a lot of work in this area in the past years (Pereira et al, 1993;Gasperin et al, 2001;Glickman and Dagan, 2003;Shimohata and Sumita, 2005), much of which can be easily replicated using the Python-NLTK combination. This would allow for a very hands-on treatment and would allow the students to gain insight into this important, but often omitted, idea from computational linguistics.…”