Word2Vec: Optimal Hyper-Parameters and Their Impact on NLP Downstream Tasks.
Tosin P. AdewumiFoteini LiwickiMarcus LiwickiPublished in: CoRR (2020)
Keyphrases
- hyperparameters
- closed form
- model selection
- cross validation
- natural language processing
- bayesian framework
- support vector
- tasks in natural language processing
- dynamic programming
- co occurrence
- bayesian inference
- variational bayes
- random sampling
- maximum a posteriori
- bayesian methods
- parameter space
- prior information
- machine learning
- incremental learning
- parameter settings
- data mining
- sample size
- posterior distribution
- text mining
- worst case