Compositionality: Similarity versus Interpretability
نویسنده
چکیده
The present paper studies the general implications of the principle of compo-sitionality for the organization of grammar. It will be argued that Janssen's (1986) requirement that syntax and semantics be similar algebras is too strong, and that the more liberal requirement that syntax be interpretable into semantics leads to a formalization that can be motivated and applied more easily, while it avoids the technical complications that encumber Janssen's formalization. Moreover, this alternative formalization even allows one to fur-ther`complete' the formal theory of compositionality, in that it is capable of clarifying the role played by model-theoretic interpretation and meaning postulates, two aspects that received little attention in Janssen (1986) and Montague (1970). In its most general form, the principle of compositionality states that`the meaning of an expression is a function of the meanings of its parts and of the way they are syntactically combined' (Partee 1984, p. 281). In other words: the meaning of an expression is determined completely by the meanings of its parts plus the information which syntactic rules have been used to build that expression out of those parts. The principle is also known as`Frege's principle', but this attribu-tion is at best a tribute according to Janssen (1986), who gives a formalization of the principle which is based on Montague's paper`Universal Grammar' (1970). Janssen's formalization and the framework deened in Montague (1970) are, roughly speaking, `diierent views of the same mathematical object' (Janssen 1986, Part 1, p. 91). The main diierence is that Janssen employs many-sorted algebras, whereas Montague uses one-sorted algebras (though with much additional structure). As a consequence of this, Janssen's approach has various advantages (see Hendriks 1993, p. 136). The term`many-sorted algebra' stems from Adj (1977). We deene the notion as follows: 1 h(A s) s2S ; (F) 2? i is a many-sorted algebra of signature (algebra) ii (a) S is a non-empty set (of sorts); (b) (A s) s2S is an indexed family of sets (A s is the carrier of s); (c) ? is a set (of operator indices); (d) (the type-assigning function) assigns to each 2 ? a pair hhs 1 (e) (F) 2? is an indexed family (of operators) such that if To cut a long story short, Janssen argues that the compositionality principle dictates the following: (a) the syntax is a-algebra A = h(A s) s2S ; (F) 2? i with 1 Our terminology deviates from Janssen (1986), where a …
منابع مشابه
Detecting Compositionality of English Verb-Particle Constructions using Semantic Similarity
We present a novel method for detecting the compositionality of English verbparticle constructions (VPCs), based on the assumption that compositionality can be modelled with semantic similarity between VPCs and their base verb in isolation. We also evaluate the contribution of components in compositional VPCs using semantic overlap.
متن کاملUsing Distributional Similarity of Multi-way Translations to Predict Multiword Expression Compositionality
We predict the compositionality of multiword expressions using distributional similarity between each component word and the overall expression, based on translations into multiple languages. We evaluate the method over English noun compounds, English verb particle constructions and German noun compounds. We show that the estimation of compositionality is improved when using translations into m...
متن کاملA Study of Metrics of Distance and Correlation Between Ranked Lists for Compositionality Detection
Compositionality in language refers to how much the meaning of some phrase can be decomposed into the meaning of its constituents and the way these constituents are combined. Based on the premise that substitution by synonyms is meaning-preserving, compositionality can be approximated as the semantic similarity between a phrase and a version of that phrase where words have been replaced by thei...
متن کاملA Word Embedding Approach to Predicting the Compositionality of Multiword Expressions
This paper presents the first attempt to use word embeddings to predict the compositionality of multiword expressions. We consider both singleand multi-prototype word embeddings. Experimental results show that, in combination with a back-off method based on string similarity, word embeddings outperform a method using count-based distributional similarity. Our best results are competitive with, ...
متن کاملMeasuring the Compositionality of Collocations via Word Co-occurrence Vectors: Shared Task System Description
A description of a system for measuring the compositionality of collocations within the framework of the shared task of the Distributional Semantics and Compositionality workshop (DISCo 2011) is presented. The system exploits the intuition that a highly compositional collocation would tend to have a considerable semantic overlap with its constituents (headword and modifier) whereas a collocatio...
متن کامل