paper-with-me

Papers

Comparing word2vec and GloVe for Automatic Measurement of MWE Compositionality

2020-12-01 · COLING (MWE) 2020 12 · Thomas Pickard

This paper explores the use of word2vec and GloVe embeddings for unsupervised measurement of the semantic compositionality of MWE candidates. Through comparison with several human-annotated reference sets, we find word2vec to be substantively superior to GloVe for this task. We also find Simple English Wikipedia to be a poor-quality resource for compositionality assessment, but demonstrate that a sample of 10% of sentences in the English Wikipedia can provide a conveniently tractable corpus with only moderate reduction in the quality of outputs.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Revisiting Additive Compositionality: AND, OR and NOT Operations with Word Embeddings

2021-05-18 · Masahiro Naito, Sho Yokoi, Geewook Kim, Hidetoshi Shimodaira

It is well-known that typical word embedding methods such as Word2Vec and GloVe have the property that the meaning can be composed by adding up the embeddings (additive compositionality). Several theories have been propo…

Word Embeddings

Detecting Unseen Multiword Expressions in American Sign Language

2023-09-30 · Lee Kezar, Aryan Shukla

Multiword expressions present unique challenges in many translation tasks. In an attempt to ultimately apply a multiword expression detection system to the translation of American Sign Language, we built and tested two s…

TranslationWord Embeddings

mwetoolkit+sem: Integrating Word Embeddings in the mwetoolkit for Semantic MWE Processing

2016-05-01 · LREC 2016 5 · Silvio Cordeiro, Carlos Ramisch, Aline Villavicencio

This paper presents mwetoolkit+sem: an extension of the mwetoolkit that estimates semantic compositionality scores for multiword expressions (MWEs) based on word embeddings. First, we describe our implementation of vecto…

Word Embeddings

An Automated Method to Enrich Consumer Health Vocabularies Using GloVe Word Embeddings and An Auxiliary Lexical Resource

2021-05-18 · Mohammed Ibrahim, Susan Gauch, Omar Salman, Mohammed Alqahatani

Background: Clear language makes communication easier between any two parties. A layman may have difficulty communicating with a professional due to not understanding the specialized terms common to the domain. In health…

Word Embeddings

Are Word Embedding Methods Stable and Should We Care About It?

2021-04-17 · Angana Borah, Manash Pratim Barman, Amit Awekar

A representation learning method is considered stable if it consistently generates similar representation of the given data across multiple runs. Word Embedding Methods (WEMs) are a class of representation learning metho…

ClusteringFairnessPOSPOS Tagging+2