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

Abstract

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.

Detecting light verb constructions across languages

“…Here, we present the most commonly used databases and corpora annotated for LVCs in several languages. Krenn (2008) developed a database of German PP-verb combinations, while Kolesnikova and Gelbukh (2010) collected and classified Spanish verb-noun combinations. Vincze et al (2011b) manually annotated several types of MWEs (including LVCs) in English Wikipedia texts, while Tu and Roth (2011) also published an English corpus, where true LVCs were manually marked.…”