Word2vec Skip-Gram Dimensionality Selection via Sequential Normalized Maximum Likelihood

Word2vec Skip-Gram Dimensionality Selection via Sequential Normalized Maximum Likelihood

In this paper, we propose a novel information criteria-based approach to select the dimensionality of the word2vec Skip-gram (SG). From the perspective of the probability theory, SG is considered as an implicit probability distribution estimation under the assumption that there exists a true context...

Full description

Bibliographic Details
Main Authors: Pham Thuc Hung, Kenji Yamanishi
Format: Article
Language:English
Published: MDPI AG 2021-07-01
Series:Entropy
Subjects:
model selection
information criteria
minimum description length
sequentially normalized maximum likelihood
word embedding
word2vec
Online Access:https://www.mdpi.com/1099-4300/23/8/997

Internet

https://www.mdpi.com/1099-4300/23/8/997

Similar Items