A Quantum-Theoretic Approach to Distributional Semantics
نویسندگان
چکیده
In this paper we explore the potential of quantum theory as a formal framework for capturing lexical meaning. We present a novel semantic space model that is syntactically aware, takes word order into account, and features key quantum aspects such as superposition and entanglement. We define a dependency-based Hilbert space and show how to represent the meaning of words by density matrices that encode dependency neighborhoods. Experiments on word similarity and association reveal that our model achieves results competitive with a variety of classical models.
منابع مشابه
On Quantum Generalizations of Information-Theoretic Measures and their Contribution to Distributional Semantics
Information-theoretic measures such as relative entropy and correlation are extremely useful when modeling or analyzing the interaction of probabilistic systems. We survey the quantum generalization of 5 such measures and point out some of their commonalities and interpretations. In particular we find the application of information theory to distributional semantics useful. By modeling the dist...
متن کاملLexicalised compositionality
In this paper, we propose an approach to distributional semantics which can be formally related to a simple model-theoretic approach. We describe treatments of some of the traditional lexical semantic relationships within this framework, and also outline accounts of some phenomena which have been considered within Generative Lexicon theory. We further argue that distributions should be based on...
متن کاملCategory-theoretic quantitative compositional distributional models of natural language semantics
This thesis is about the problem of compositionality in distributional semantics. Distributional semantics presupposes that the meanings of words are a function of their occurrences in textual contexts. It models words as distributions over these contexts and represents them as vectors in high dimensional spaces. The problem of compositionality for such models concerns itself with how to produc...
متن کاملGraded Entailment for Compositional Distributional Semantics
The categorical compositional distributional model of natural language provides a conceptually motivated procedure to compute the meaning of sentences, given grammatical structure and the meanings of its words. This approach has outperformed other models in mainstream empirical language processing tasks. However, until now it has lacked the crucial feature of lexical entailment – as do other di...
متن کاملDistributional Sentence Entailment Using Density Matrices
Categorical compositional distributional model of [9] suggests a way to combine grammatical composition of the formal, type logical models with the corpus based, empirical word representations of distributional semantics. This paper contributes to the project by expanding the model to also capture entailment relations. This is achieved by extending the representations of words from points in me...
متن کامل