2018-01-11

Adjusting Word Embeddings with Semantic Intensity Orders

clustering words for intensity ordering首先需要从Google N-gram中抽取形容词的程度顺序信息。使用的方法是基于模版匹配的方法。比如从good but not great我们可以总结出一个规则xx but not xx，然后可以发现后者比前者的程度深。然后使用mixed integer linear programming (MILP)来进行最优化排序

Papers

2018-01-10

Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinction

他们想增强一些能够明显区分同义词和反义词的特征。

Papers

2018-01-09

Ontologically Grounded Multi-sense Representation Learning for Semantic Vector Space Models

IntroductionIn this paper thye propose two novel and general approaches for generating sense-specific word embeddings that are grounded in an ontology. 但是通用的词向量只为一个词分配一个词向量，无法解决解决一词多义的问题。虽然Yu and Dre

Papers

2018-01-07

AutoExtend Extending Word Embeddings to Embeddings for Synsets and Lexemes

Introduction这篇文章提出了一个叫做AutoExtand的方法，来运用其他的信息来增强word embedding的性能。用于学习用来表示synsets和lexemes。用于一些公开的知识库上，比如WordNet, Wikipedia和Freebase。 synset是指一个词语集合，这里面的词可以在一定的条件下相互替换。lexeme会将一个特定的拼写和发音和一个特定的意思匹配在一起。也

Blog

归档: 2018/1

Adjusting Word Embeddings with Semantic Intensity Orders

Integrating Distributional Lexical Contrast into Word Embeddings for Antonym-Synonym Distinction

Ontologically Grounded Multi-sense Representation Learning for Semantic Vector Space Models

AutoExtend Extending Word Embeddings to Embeddings for Synsets and Lexemes