Victoria Ubaldo
@Vikyale
Introduction to NLP :
How to improve accessibility
with Machine Learning
CONNECT REIMAGINE 2021
Slide 3
Slide 3 text
Hello!
Computer and System Engineer. Peruvian
Data Analyst @Interbank
Msc Candidate Computer Science, PUCP.
Enjoy mentoring, illustration and dancing!
Slide 4
Slide 4 text
Agenda
● Machine Learning and Accessibility
● Natural Language Processing
● Tensorflow for NLP
● Next steps
Slide 5
Slide 5 text
No content
Slide 6
Slide 6 text
No content
Slide 7
Slide 7 text
No content
Slide 8
Slide 8 text
Machine Learning is
programming with data
Slide 9
Slide 9 text
No content
Slide 10
Slide 10 text
No content
Slide 11
Slide 11 text
Natural Language Process (NLP)
Slide 12
Slide 12 text
No content
Slide 13
Slide 13 text
Where find NLP ?
Chatbots
Predictive text
(autocomplete)
Transcription
Language translation
Others: Text Analysis
Email filters
Search Result
Sentiment Analysis
Slide 14
Slide 14 text
Challenges in NLP
Slide 15
Slide 15 text
No content
Slide 16
Slide 16 text
Challenges in NLP : Textual Accessibility
Visual Auditory Speech
Slide 17
Slide 17 text
Accessible = Comprehensible
Slide 18
Slide 18 text
Challenges in NLP : Textual Accessibility
Transcription Speech recognition
Language Translation
Slide 19
Slide 19 text
Case :
Recognize sentiment in Text
Slide 20
Slide 20 text
No content
Slide 21
Slide 21 text
Amazon and Yelp reviews
1 : Good
0: Bad
Slide 22
Slide 22 text
Pre-processing data
Slide 23
Slide 23 text
Text Processing
Slide 24
Slide 24 text
Tokenization
Slide 25
Slide 25 text
Out Of Vocabulary (OOV)
EXTRA WORDS
unknown tokens (UNK)
Slide 26
Slide 26 text
Stop Words
Slide 27
Slide 27 text
Stemming
Slide 28
Slide 28 text
Lemmatization
Slide 29
Slide 29 text
Padding
Slide 30
Slide 30 text
Tools!
● Python 3.+
● Jupyter Notebook or Google Colab
● and using a framework for ML
Slide 31
Slide 31 text
¿Why Tensorflow?
Whether you’re an expert or a beginner, TensorFlow is an end-to-end platform that makes it
easy for you to build and deploy ML models.
Slide 32
Slide 32 text
Trying in Google Colab
bit.ly/2TU9MJz
Slide 33
Slide 33 text
Challenges in NLP : Textual Accessibility
Transcription
Speech recognition
Slide 34
Slide 34 text
No content
Slide 35
Slide 35 text
Challenges in NLP : Textual Accessibility
Language Translation
Slide 36
Slide 36 text
No content
Slide 37
Slide 37 text
Next steps
Try new tools :
- Recurrent Neural Network (RNN)
- Long short-term memory (LSTM)
Slide 38
Slide 38 text
Next steps
Increase your corpus with diverse data.
- PDFMiner
- Tweepy API
Slide 39
Slide 39 text
Next steps
Datasets with native or indigenous
language or code mixing
Slide 40
Slide 40 text
“We can build a much brighter future
where humans are relieved of menial work
using AI capabilities”
- Andrew Ng
Slide 41
Slide 41 text
Thanks!
@womenwhocode
#WWCode
Keep in touch!
linkedin.com/in/victoriaubaldo
@vikyale