Experimental Support for a Categorical Compositional Distributional Model of Meaning

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 1394–1404, Edinburgh, Scotland, UK, July 27–31, 2011. c⃝2011 Association for Computational Linguistics Experimental Support for a Categorical Compositional Distributional Model of Meaning Edward Grefenstette University of Oxford Department of Computer Science Wolfson Building, Parks Road Oxford OX1 3QD, UK edward.grefenstette@cs.ox.ac.uk Mehrnoosh Sadrzadeh University of Oxford Department of Computer Science Wolfson Building, Parks Road Oxford OX1 3QD, UK mehrs@cs.ox.ac.uk Abstract Modelling compositional meaning for sen- tences using empirical distributional methods has been a challenge for computational lin- guists. We implement the abstract categorical model of Coecke et al. (2010) using data from the BNC and evaluate it. The implementation is based on unsupervised learning of matrices for relational words and applying them to the vectors of their arguments. The evaluation is based on the word disambiguation task devel- oped by Mitchell and Lapata (2008) for intran- sitive sentences, and on a similar new experi- ment designed for transitive sentences. Our model matches the results of its competitors in the first experiment, and betters them in the second. The general improvement in results with increase in syntactic complexity show- cases the compositional power of our model. 1 Introduction As competent language speakers, we humans can al- most trivially make sense of sentences we’ve never seen or heard before. We are naturally good at un- derstanding ambiguous words given a context, and forming the meaning of a sentence from the mean- ing of its parts. But while human beings seem comfortable doing this, machines fail to deliver. Search engines such as Google either fall back on bag of words models—ignoring syntax and lexical relations—or exploit superficial models of lexical semantics to retrieve pages with terms related to those in the query (Manning et al., 2008). However, such models fail to shine when it comes to processing the semantics of phrases and sen- tences. Discovering the process of meaning as- signment in natural language is among the most challenging and foundational questions of linguis- tics and computer science. The findings thereof will increase our understanding of cognition and intelli- gence and shall assist in applications to automating language-related tasks such as document search. Compositional type-logical approaches (Mon- tague, 1974; Lambek, 2008) and distributional mod- els of lexical semantics (Schutze, 1998; Firth, 1957) have provided two partial orthogonal solutions to the question. Compositional formal semantic models stem from classical ideas from mathematical logic, mainly Frege’s principle that the meaning of a sen- tence is a function of the meaning of its parts (Frege, 1892). Distributional models are more recent and can be related to Wittgenstein’s later philosophy of ‘meaning is use’, whereby meanings of words can be determined from their context (Wittgenstein, 1953). The logical models relate to well known and robust logical formalisms, hence offering a scalable theory of meaning which can be used to reason inferen- tially. The distributional models have found their way into real world applications such as thesaurus extraction (Grefenstette, 1994; Curran, 2004) or au- tomated essay …

Content truncated to 500 words