Example-based Acquisition of Fine-grained Collocation Resources

15854

Authors

nan

Abstract

AbstractCollocations such as “heavy rain” or “make [a] decision”, are combinations of two elements where one (the base) is freely chosen, while the choice of the other (collocate) is restricted, depending on the base. Collocations present difficulties even to advanced language learners, who usually struggle to find the right collocate to express a particular meaning, e.g., both “heavy” and “strong” express the meaning ‘intense’, but while “rain” selects “heavy”, “wind” selects “strong”. Lexical Functions (LFs) describe the meanings that hold between the elements of collocations, such as ‘intense’, ‘perform’, ‘create’, ‘increase’, etc. Language resources with semantically classified collocations would be of great help for students, however they are expensive to build, since they are manually constructed, and scarce. We present an unsupervised approach to the acquisition and semantic classification of collocations according to LFs, based on word embeddings in which, given an example of a collocation for each of the target LFs and a set of bases, the system retrieves a list of collocates for each base and LF.

Keywords

nan

Full Content

Example-based Acquisition of Fine-grained Collocational Resources Sara Rodr´ıguez-Fern´andez1, Roberto Carlini1, Luis Espinosa-Anke1, Leo Wanner1,2 1NLP Group, Department of Information and Communication Technologies, Pompeu Fabra University C/ Roc Boronat, 138, 08018 Barcelona (Spain) 2Catalan Institute for Research and Advanced Studies (ICREA) sara.rodriguez.fernandez|roberto.carlini|luis.espinosa|leo.wanner@upf.edu Abstract Collocations such as heavy rain or make [a] decision, are combinations of two elements where one (the base) is freely chosen, while the choice of the other (collocate) is restricted, depending on the base. Collocations present difficulties even to advanced language learners, who usually struggle to find the right collocate to express a particular meaning, e.g., both heavy and strong express the meaning ‘intense’, but while rain selects heavy, wind selects strong. Lexical Functions (LFs) describe the meanings that hold between the elements of collocations, such as ‘intense’, ‘perform’, ‘create’, ‘increase’, etc. Language resources with semantically classified collocations would be of great help for students, however they are expensive to build, since they are manually constructed, and scarce. We present an unsu- pervised approach to the acquisition and semantic classification of collocations according to LFs, based on word embeddings in which, given an example of a collocation for each of the target LFs and a set of bases, the system retrieves a list of collocates for each base and LF. Keywords: collocation retrieval, collocation semantic classification, collocation resources, second language learning, word em- beddings 1. Introduction Collocations of the kind make [a] suggestion, attend [a] lecture, heavy rain, deep thought, etc., are restricted lexical co-occurrences between two syntactically related lexical el- ements. One of these elements (the base) is freely chosen, while the choice of the other (the collocate) depends on the base (Hausmann, 1984; Cowie, 1994; Mel’ˇcuk, 1995). For instance, in make [a] suggestion, the choice of sug- gestion is free, while the choice of make is restricted; cf., e.g., *take [a] suggestion. Such collocations are by their nature idiosyncratic and therefore also language-specific. Thus, you make [a] suggestion, but you give [an] advice, you attend [a] lecture, but you assist [an] operation, rain is heavy while wind is strong. In English, you take [a] walk, while in Spanish you ‘give’ it (dar [un] paseo, and in German you ‘make’ it ([einen] Spaziergang machen); in English, rain is heavy, while in Spanish and German it is ‘strong’ (fuerte lluvia/starker Regen). And so on. The idiosyncrasy of such collocations makes them a real chal- lenge even for advanced second language learners (Haus- mann, 1984; Bahns and Eldaw, 1993; Granger, 1998; Lewis and Conzett, 2000; Nesselhauf, 2005; Alonso Ramos et al., 2010). Nesselhauf (2005) and Farghal and Obiedat (1995) report that learners of English paraphrase their discourse in order to avoid using collocations that they do not master. In other words, a learner knows the meaning they want to express, but they fail to do it by means of a collocation, which means that they fail to pick the collocate that ex- presses this meaning. Semantically annotated collocation resources would thus be of great aid. A number of collo- …

Content truncated to 500 words

Labeled Text

AbstractCollocations such as “heavy rain” or “make [a] decision”, are combinations of two elements where one (the base) is freely chosen, while the choice of the other (collocate) is restricted, depending on the base. Collocations present difficulties even to advanced language learners, who usually struggle to find the right collocate to express a particular meaning, e.g., both “heavy” and “strong” express the meaning ‘intense’, but while “rain” selects “heavy”, “wind” selects “strong”. Lexical Functions (LFs) describe the meanings that hold between the elements of collocations, such as ‘intense’, ‘perform’, ‘create’, ‘increase’, etc. Language resources with semantically classified collocations would be of great help for students, however they are expensive to build, since they are manually constructed, and scarce. We present an unsupervised approach to the acquisition and semantic classification of collocations according to LFs, based on word embeddings in which, given an example of a collocation for each of the target LFs and a set of bases, the system retrieves a list of collocates for each base and LF.

Publication Details

Paper ID:
15854
Year:
2016
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan