%autosave 10
Autosaving every 10 seconds
What models and libraries do you use?
scikit-learn
and nltk
Stack, how to scale nltk
to production?
redis
didn't scale, switching to rabbitmq
.nltk
.What customer feedback specifically?
Is Python a bottleneck for high throughput?
Do you do picture analysis?
Have you considered feature reduction using word2vec
https://code.google.com/p/word2vec/:
word2vec
used for feature engineering raw text into numeric arrays.Time series analysis? Seasonality, old documents?
Disambiguation of language, slang?
Do any categorisation or sentiment analysis?