Improving Lexical Embeddings with Semantic Knowledge
TL
TL;DR
This work proposes a new learning objective that incorporates both a neural language model objective (Mikolov et al., 2013) and prior knowledge from semantic resources to learn improved lexical semantic embeddings.
Abstract
Word embeddings learned on unlabeled data are a popular tool in semantics, but may not capture the desired semantics. We propose a new learning objective that incorporates both a neural language model objective (Mikolov et al., 2013) and prior knowledge from semantic resources to learn improved lexical semantic embeddings. We demonstrate that our embeddings improve over those learned solely on raw text in three settings: language modeling, measuring semantic similarity, and predicting human judgements.
References18 items
1
Distributed Representations of Words and Phrases and their Compositionality
2
Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank
3
Better Word Representations with Recursive Neural Networks for Morphology
4
PPDB: The Paraphrase Database
5