Difference between revisions of "Class"
| Line 17: | Line 17: | ||
== Word2Vec == | == Word2Vec == | ||
Skip-Gram | Skip-Gram | ||
| − | |||
| − | |||
| Line 24: | Line 22: | ||
* Word Cloud | * Word Cloud | ||
| − | |||
| − | |||
| − | |||
= Text Classification = | = Text Classification = | ||
| Line 33: | Line 28: | ||
= Financial News = | = Financial News = | ||
| + | |||
Tools: | Tools: | ||
| Line 40: | Line 36: | ||
** Skipgram | ** Skipgram | ||
** Continuous bag of words | ** Continuous bag of words | ||
| + | |||
| + | Multi-channel LSTM Network | ||
| + | Keras wih TensorFlow | ||
| + | Utilize the CloVe and FastText Skipgram pretrained embeddings, allows he underlying network to access larger feature space to build complex features on top of. | ||
| + | |||
| + | Can use utilize combinations of vaious corpus and embedding methods for better performance | ||
| + | |||
| + | Bidirectional LSTM network is used o encode seuential information on the embedding layers. | ||
| + | |||
| + | Dense layer to project fnal output classification | ||
Revision as of 11:53, 23 October 2018
https://courses.nvidia.com/dashboard
Linguistic Concepts
- conference - anaphors
- gang of four design
- null subject
- recursion
Contents
Word Embeddings
- HMMS, CRF, PGMs
- CBoW -Bag of Words / ngrams - feature per word/n items
- 1-hot Sparse input - create a vector the size of the entire vocabulary
- Stop Words
- TF-IDF
Word2Vec
Skip-Gram
- Firth 1957 Distributional Hypothess
- Word Cloud
Text Classification
Text/Machine Translation (MNT)
Financial News
Tools:
- Glove
- dot product
- FastText
- Skipgram
- Continuous bag of words
Multi-channel LSTM Network Keras wih TensorFlow Utilize the CloVe and FastText Skipgram pretrained embeddings, allows he underlying network to access larger feature space to build complex features on top of.
Can use utilize combinations of vaious corpus and embedding methods for better performance
Bidirectional LSTM network is used o encode seuential information on the embedding layers.
Dense layer to project fnal output classification
