Difference between revisions of "讨论:2015-08-17"

From cslt Wiki
Jump to: navigation, search
(Luo Ling 2015-08-17)
 
(清空页面)
 
Line 1: Line 1:
Works in This week:
 
  
1.Finish training word embeddings via 5 models :
 
using EnWiki dataset(953M):
 
CBOW,Skip-Gram(SG)
 
using text8 dataset(95.3M):
 
CBOW,Skip-Gram(SG),C&W,GloVe,LBL and Order(count-based)
 
 
2.Use tasks to measure quality of the word vectors with various dimensions:
 
word similarity(ws)
 
the TOEFL set
 
analogy task
 
text classification
 
named entity recognition(ner)
 
sentence-level sentiment classification (based on convolutional neural networks),just call it 'cnn'
 
part-of-speech tagging(pos)
 

Latest revision as of 04:53, 19 August 2015