Distributed Structures and Distributional Meaning
نویسندگان
چکیده
Stemming from distributed representation theories, we investigate the interaction between distributed structure and distributional meaning. We propose a pure distributed tree (DT) and distributional distributed tree (DDT). DTs and DDTs are exploited for defining distributed tree kernels (DTKs) and distributional distributed tree kernels (DDTKs). We compare DTKs and DDTKs in two tasks: approximating tree kernels TK (Collins and Duffy, 2002); performing textual entailment recognition (RTE). Results show that DTKs correlate with TKs and perform in RTE better than DDTKs. Then, including distributional vectors in distributed structures is a very difficult task.
منابع مشابه
Towards Syntax-aware Compositional Distributional Semantic Models
Compositional Distributional Semantics Models (CDSMs) are traditionally seen as an entire different world with respect to Tree Kernels (TKs). In this paper, we show that under a suitable regime these two approaches can be regarded as the same and, thus, structural information and distributional semantics can successfully cooperate in CSDMs for NLP tasks. Leveraging on distributed trees, we pres...
متن کاملA Frobenius Model of Information Structure in Categorical Compositional Distributional Semantics
The categorical compositional distributional model of Coecke et al. (2010) provides a linguistically motivated procedure for computing the meaning of a sentence as a function of the distributional meaning of the words therein. The theoretical framework allows for reasoning about compositional aspects of language and offers structural ways of studying the underlying relationships. While the mode...
متن کاملDistributed representations for compositional semantics
The mathematical representation of semantics is a key issue for Natural Language Processing (NLP). A lot of research has been devoted to finding ways of representing the semantics of individual words in vector spaces. Distributional approaches—meaning distributed representations that exploit co-occurrence statistics of large corpora—have proved popular and successful across a number of tasks. H...
متن کاملCompositional Distributional Cognition
Building on existing compositional distributional models of meaning [12, 47], we present an adaptation of the Integrated Connectionist/Symbolic Architecture (ICS) introduced in [52]. In particular, we build a novel compositional distributional model of cognition (DISCOG) that resolves representational issues in ICS via mapping of same-type symbolic structures to a finite shared meaning space. I...
متن کاملCompositional Distributional Semantics Models in Chunk-based Smoothed Tree Kernels
The field of compositional distributional semantics has proposed very interesting and reliable models for accounting the distributional meaning of simple phrases. These models however tend to disregard the syntactic structures when they are applied to larger sentences. In this paper we propose the chunk-based smoothed tree kernels (CSTKs) as a way to exploit the syntactic structures as well as ...
متن کامل