Word Embeddings: A Survey

TL;DR

This survey reviews recent methods for creating word embeddings, which are dense vector representations capturing syntactic and semantic information, and highlights their usefulness in various NLP tasks.

Contribution

It provides a comprehensive overview of recent strategies for building word embeddings based on the distributional hypothesis.

Findings

Word embeddings encode syntactic and semantic information.

They improve performance in many NLP tasks.

Abstract

This work lists and describes the main recent strategies for building fixed-length, dense and distributed representations for words, based on the distributional hypothesis. These representations are now commonly called word embeddings and, in addition to encoding surprisingly good syntactic and semantic information, have been proven useful as extra features in many downstream NLP tasks.