Towards automatic fine-grained semantic classification of verb-noun collocations

Wanner, Leo

doi:10.1017/s1351324904003328

Cited by 24 publications

(28 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Should this criterion be statistical or semantic? (Wanner, 2004) gives a good concise overview of this debate. The statistical definition of collocation, i.e.…”

Section: Discussion: Testing the Linguistic Statementmentioning

confidence: 99%

Semantic relations between collocations: A Spanish case study

Kolesnikova¹,

Gelbukh²

2012

Rev. signos

View full text Add to dashboard Cite

Linguistics as a scientific study of human language intends to describe and explain it. However, validity of a linguistic theory is difficult to prove due to volatile nature of language as a human convention and impossibility to cover all real-life linguistic data. In spite of these problems, computational techniques and modeling can provide evidence to verify or falsify linguistic theories. As a case study, we conducted a series of computer experiments on a corpus of Spanish verb-noun collocations using machine learning methods, in order to test a linguistic point that collocations in the language do not form an unstructured collection but are language items related via what we call collocational isomorphism, represented by lexical functions of the Meaning-Text Theory. Our experiments allowed us to verify this linguistic statement. Moreover, they suggested that semantic considerations are more important in the definition of the notion of collocation than statistical ones. 45Resumen: La Lingüística, siendo el estudio científico del lenguaje humano, intenta describirlo y explicarlo. Sin embargo, es difícil demostrar la certeza de cualquier teoría lingüística por la naturaleza versátil del lenguaje como una convención humana y también por la imposibilidad de investigar todo lo que se habla y se escribe en la vida real. A pesar de estos problemas, las técnicas y los modelos computacionales pueden proporcionar la evidencia para que las teorías lingüísticas sean comprobadas o refutadas. A través de un estudio de caso, realizamos una serie de experimentos por computador en un corpus de colocaciones de verbo-sustantivo en español usando métodos de aprendizaje de máquina, con el fin de comprobar el hecho lingüístico de que las colocaciones del idioma no conforman un grupo sin estructura sino que son unidades lingüísticas relacionadas por medio de lo que denominamos isomorfismo de colocaciones, representado por las funciones léxicas de la Teoría Significado-Texto. Nuestros experimentos nos permiten verificar esa declaración lingüística. Asimismo, los experimentos sugieren que las consideraciones semánticas son más importantes en la definición de colocaciones que las estadísticas.Palabras Clave: Colocaciones, funciones léxicas, aprendizaje de máquina.

show abstract

“…Should this criterion be statistical or semantic? (Wanner, 2004) gives a good concise overview of this debate. The statistical definition of collocation, i.e.…”

Section: Discussion: Testing the Linguistic Statementmentioning

confidence: 99%

Semantic relations between collocations: A Spanish case study

Kolesnikova¹,

Gelbukh²

2012

Rev. signos

View full text Add to dashboard Cite

show abstract

“…Collocations may have different syntactic structures, and the verb-noun pattern is one of these structures. L.Wanner [16,17] made experiments to classify Spanish verb-noun pairs according to nine lexical functions with the meaning 'perform, experience, carry out something, 'cause the existence of something, 'begin to perform something, 'continue to perform something', etc. Verb-noun pairs were divided in two groups.…”

Section: Related Workmentioning

confidence: 99%

Supervised Machine Learning for Predicting the Meaning of Verb-Noun Combinations in Spanish

Kolesnikova

Gelbukh

2010

Advances in Soft Computing

View full text Add to dashboard Cite

Abstract. The meaning of such verb-noun combinations as take care, undertake work, pay attention can be generalized as DO what is designated by the noun. Likewise, the meaning of make a decision, provide support, write a letter can be generalized as MAKE what is designated by the noun. These generalizations represent the meaning of certain groups of verb-noun combinations. We use supervised machine learning algorithms to predict the meanings DO, MAKE, BEGIN, and CONTINUE of previously unseen verbnoun pairs. We evaluate the performance of the applied algorithms on a training set using 10-fold cross-validation technique. The learnt models have also been evaluated on an independent test set and the predictions have been checked manually to determine the accuracy of the classifiers. The obtained results show that supervised machine learning methods achieve significant accuracy and can be used for semantic annotation of verb-noun combinations.

show abstract

“…Incongruent collocations are more difficult than congruent collocations, and they are harder to acquire but, once acquired, they can be processed as efficiently as congruent collocations (Nesselhauf, 2005;Wolter & Gyllstad, 2013;Yamashita & Jiang, 2010) and it is the incongruent collocations that are known to be problematic in learners' production of L2 collocations (Boers, Eyckmans, & Lindstromberg, 2013;Vasiljevic, 2008) While Firth (1957a), Nattinger and Decarrico (1992) refer to mutual expectancies of words, many studies (e.g., Mel'ˇcuk, 1995;Moon, 1998;Nation, 2001;Wanner, 2004) have highlighted that there is not necessarily any special semantic bond between collocating items, so, the semantic of a word as a collocate does not need to be the same as the semantic of that word in isolation. The nonpredictability aspect of collocations has been emphasized by some researchers including : Benson (1989a: Benson ( , 1989b, Smadja (1993), and Nation (2001).…”

Section: )mentioning

confidence: 99%

“…Grammatical collocations combine a lexical word, typically a noun, verb or adjective, with a preposition or grammatical structure such as an infinitive or clause (Michael Lewis, 2001b In terms of degree of idiomaticity lexical collocations have been classified (Howarth, 1998a;Nesselhauf, 2003;Wanner, 2004) The current study will focus on restricted lexical collocations. Lexical collocations are defined as combinations that consist of nouns, adjectives, verbs and adverbs (e.g., Martynska, 2004) and the research study will focus on verb + noun lexical collocations, for example, charge a fee, as these collocations have been found to be a major problem for ESL learners and have been considered significant in the literature (e.g., Bahns, 1993;Erman et al, 2015;Gitsaki, 1999b;Hsu, 2007;Levitzky-Aviad & Laufer, 2013;Li, 2005;Nesselhauf, 2003;Rezaee, Marefat, & Saeedakhtar, 2014;Wible, Kuo, Tsao, Liu, & Lin, 2003).…”

Section: Classification Of Collocations For L2 Acquisitionmentioning

confidence: 99%

See 1 more Smart Citation

Investigating the effectiveness of involvement load in acquisition and retention of ESL learners’ collocational knowledge

Majd¹

View full text Add to dashboard Cite

Learning collocations is one of the difficult tasks in second language acquisition that has been largely overlooked in the research literature. The patterning of words and collocations in a language is not predictable based on grammar and lexicon alone. A variety of factors have been found to affect the acquisition of collocations in the ESL context such as exposure, the degree of L1-L2 differences, language proficiency level, and type of collocations. Previous research has focused predominantly on detecting ESL learners' collocational errors in which lexical miscollocations are among the most common. In particular, verb + noun lexical collocations such as draw conclusions are found to be more difficult than others. However, designing effective ways for second language learners to acquire collocations has not been examined in detail. The current study attempted to address this issue by exploring the effect of tasks with different involvement indexes with regard to verb + noun collocations, and by operationalising the Involvement Load Hypothesis proposed by Hulstijn and Laufer (2001) that provides the theoretical framework for the study.The main aim of this thesis is to increase the understanding of how ESL learners acquire and retain collocations, and to improve the effectiveness of tasks designed for assisting the acquisition and retention of L2 collocations. Learning tasks were designed where Involvement Load was varied from Low to Medium to High. The experimental hypothesis is that tasks with higher involvement load facilitate the learning and retention of L2 collocations better than tasks with lower involvement load. The participants in the study were selected from a single language group, Mandarin speakers from China, who were all university students studying at the University of Queensland.The study was based on a within subjects experimental design that included a detailed system for tracking learning tasks and items across groups so that the effects of different involvement loads could be measured. A pretest, an immediate posttest and a delayed-posttest of learning was performed. Also, at the conclusion of the experimental phase the participants answered a questionnaire that probed their learning experiences and evaluation of the tasks, focusing in particular on the difficulty and perceived effectiveness of the collocation tasks that varied from low to high involvement.The results of the data analysis showed a significant difference between pretest and posttest; and between prestest and delayed posttest, with a very strong effect showing that learning took place and a significant improvement was found for all Levels of Involvement. Hence, the findings from the analysis of the data did not support the prediction that tasks with higher involvement load would be more effective that lower involvement load tasks for learning collocations. In fact the low ii involvement tasks were found to be more effective immediately following the completion of the tasks. Moreover, results from the participants' perc...

show abstract

Towards automatic fine-grained semantic classification of verb-noun collocations

Cited by 24 publications

References 26 publications

Semantic relations between collocations: A Spanish case study

Semantic relations between collocations: A Spanish case study

Supervised Machine Learning for Predicting the Meaning of Verb-Noun Combinations in Spanish

Investigating the effectiveness of involvement load in acquisition and retention of ESL learners’ collocational knowledge

Contact Info

Product

Resources

About