Distributional Semantics in R with the wordspace Package
نویسنده
چکیده
This paper introduces the wordspace package, which turns Gnu R into an interactive laboratory for research in distributional semantics. The package includes highly efficient implementations of a carefully chosen set of key functions, allowing it to scale up to real-life data sets.
منابع مشابه
Encoding Syntactic Dependencies using Random Indexing and Wikipedia as a Corpus
Distributional approaches are based on a simple hypothesis: the meaning of a word can be inferred from its usage. The application of that idea to the vector space model makes possible the construction of a WordSpace in which words are represented by mathematical points in a geometric space. Similar words are represented close in this space and the definition of “word usage” depends on the defin...
متن کاملEncoding syntactic dependencies by vector permutation
Distributional approaches are based on a simple hypothesis: the meaning of a word can be inferred from its usage. The application of that idea to the vector space model makes possible the construction of a WordSpace in which words are represented by mathematical points in a geometric space. Similar words are represented close in this space and the definition of “word usage” depends on the defin...
متن کاملThe distributional Henstock-Kurzweil integral and measure differential equations
In the present paper, measure differential equations involving the distributional Henstock-Kurzweil integral are investigated. Theorems on the existence and structure of the set of solutions are established by using Schauder$^prime s$ fixed point theorem and Vidossich theorem. Two examples of the main results paper are presented. The new results are generalizations of some previous results in t...
متن کاملUNIBA: Super-sense Tagging at EVALITA 2011
This paper describes our participation in EVALITA 2011 Super Sense Tagging (SST) task. The goal of the task is to annotate each word in a text within a general semantic taxonomy defined by the WordNet lexicographer classes called super-senses. In this task, we exploit structured learning based on Support Vector Machine. Moreover, we propose to solve the data sparseness problem by incorporating ...
متن کاملSemantic Vectors: a Scalable Open Source Package and Online Technology Management Application
This paper describes the open source SemanticVectors package that efficiently creates semantic vectors for words and documents from a corpus of free text articles. We believe that this package can play an important role in furthering research in distributional semantics, and (perhaps more importantly) can help to significantly reduce the current gap that exists between good research results and...
متن کامل