Natural Language Processing

Capire il Machine Learning

Sara Billen

Curriculum Manager at DataCamp

Natural Language Processing (NLP)

The ability for computers to understand the meaning of human language

Capire il Machine Learning

Bag of words

Bag of Words

Capire il Machine Learning

Bag of words

"U2 is a great band"

Word Count
U2 1
Queen 0
is 1
a 1
great 1
band 1

"Queen is a great band"

Word Count
U2 0
Queen 1
is 1
a 1
great 1
band 1
Capire il Machine Learning

Bag of words: n-grams

"That book is not great"

Word Count
That 1
book 1
is 1
not 1
great 1

2-gram (bi-gram)

Word Count
That book 1
book is 1
is not 1
not great 1
Capire il Machine Learning

Bag of words: limitations

  • Word counts don't help us consider synonyms
  • Example: "blue"
    • "sky-blue"
    • "aqua"
    • "cerulean"
  • Want to group as a single feature

blue

Capire il Machine Learning

Word embeddings

Word embeddings

  • Create features that group similar words
  • Features have a mathematical meaning:

$$

king - man + woman = queen

Capire il Machine Learning

Language translation

Translation with neural network

Capire il Machine Learning

Applications

  • Language translation
  • Chatbots
  • Personal assistants
  • Sentiment analysis
  • ...

Man chatting with chat bot on smartphone

Capire il Machine Learning

Deep learning

  • Two types of problems
    • Computer vision
    • Natural language processing
  • Why deep learning?
    • Complex problems
    • Automatic feature extraction
    • Lots of data

Capire il Machine Learning

Let's practice!

Capire il Machine Learning

Preparing Video For Download...