Lightweight Natural Langrage Processing

1
Lightweight Natural Langrage Processing

• Lin Ziheng

2

• Mobile devices is becoming the dominant computing
  platform for apps like web browsing, data/address
  books.
• Use NLP and machine learning algo
• NLP and ML algo are run on high power desktops,
  no concerns for power/memory consumption.
• NLP and ML pipeline for mobile devices needs to
  take into account these constraints.

3

• To develop an entire NLP processing pipeline (POS
  tagging, NE recognition, parsing and semantic
  role labeling..) for use in lightweight
  environment.
• To create a software suite that will be able to
  make accuracy/resource consumption tradeoffs.

4
POS Tagging

• Features
• Capitalization (first letter, all letters)
• Numbers
• Hyphens
• Previous tags (unigrams, bigrams, trigrams), 6804
• Prefix, suffix (up to length of 4), 14023
• Surrounding words (unigrams, bigrams, trigrams),
  153820
• Machine learning
• MaxEnt, SVM
• We use naïve bayes, decision tree

5
(No Transcript)
6
pre/suffix
surrounding words
basic
features
W1 W2 W3 W4