Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Introduction to NLP : How to improve accessibility with Machine Learning
Search
Victoria Ubaldo
June 11, 2021
Technology
1
180
Introduction to NLP : How to improve accessibility with Machine Learning
WWCode CONNECT REIMAGINE 2021
Victoria Ubaldo
June 11, 2021
Tweet
Share
More Decks by Victoria Ubaldo
See All by Victoria Ubaldo
Analizar Datos con Python y Google Colab
vickyale
1
54
¿Cómo estar al día en Tecnología, como profesional tech?
vickyale
1
310
Primeros pasos en AI/ML
vickyale
0
85
ML Zero to Hero in GCP
vickyale
0
46
Agile & Big Data
vickyale
1
24
Clean Data with Python
vickyale
0
160
AoG Workshop for IWD19
vickyale
0
98
Introducción a Actions for Google Assistant
vickyale
0
39
Kit para crear un sitio web móvil asombroso!
vickyale
0
20
Other Decks in Technology
See All in Technology
皆がすなるカオスエンジアリングといふものを、ネットワークオペレーションでもしてみむとてするなり
tjmtrhs
0
120
Webエンジニアのためのデータエンジニアリング概説
mtoriyama000
5
400
Tohoku.Tech #1 「EC-CUBE/AWSの構築をChatGPTに相談してみました」by テンダ
jun2882
0
130
テストだけで品質は上がらない?! エセ自己組織化した品質組織からの脱却 / JaSST'24 Tokyo
visional_engineering_and_design
9
2.9k
本気でプロダクトに向き合うCTOになるために必要な事 (技育祭2024春)
mosa_siru
33
11k
How to Build a Strong Engineering Culture
alperhankendi
0
120
新卒1年目がプロジェクトを進めるときにコケたポイント
ryunakayama
1
110
書を捨てよ、現場へ出よう
nwiizo
11
8.7k
OpenStack再入門「アーキテクチャ編」
kajinamit
0
240
JaSST_Tokyo経験発表_20240313.pdf
ktanizaki
0
190
履歴データテーブルとの向き合い方_PHPerKaigi2024
gennei
25
6.7k
Automate your changelogs! Release Drafter
onenashev
PRO
2
410
Featured
See All Featured
Making the Leap to Tech Lead
cromwellryan
122
8.4k
Fontdeck: Realign not Redesign
paulrobertlloyd
75
4.8k
The Invisible Customer
myddelton
114
12k
The Pragmatic Product Professional
lauravandoore
24
5.7k
Mobile First: as difficult as doing things right
swwweet
215
8.5k
Responsive Adventures: Dirty Tricks From The Dark Corners of Front-End
smashingmag
242
20k
Embracing the Ebb and Flow
colly
78
4.1k
Why You Should Never Use an ORM
jnunemaker
PRO
50
8.5k
Why Our Code Smells
bkeepers
PRO
330
56k
Raft: Consensus for Rubyists
vanstee
130
6.2k
Web development in the modern age
philhawksworth
201
10k
jQuery: Nuts, Bolts and Bling
dougneiner
57
7.1k
Transcript
None
Victoria Ubaldo @Vikyale Introduction to NLP : How to improve
accessibility with Machine Learning CONNECT REIMAGINE 2021
Hello! Computer and System Engineer. Peruvian Data Analyst @Interbank Msc
Candidate Computer Science, PUCP. Enjoy mentoring, illustration and dancing!
Agenda • Machine Learning and Accessibility • Natural Language Processing
• Tensorflow for NLP • Next steps
None
None
None
Machine Learning is programming with data
None
None
Natural Language Process (NLP)
None
Where find NLP ? Chatbots Predictive text (autocomplete) Transcription Language
translation Others: Text Analysis Email filters Search Result Sentiment Analysis
Challenges in NLP
None
Challenges in NLP : Textual Accessibility Visual Auditory Speech
Accessible = Comprehensible
Challenges in NLP : Textual Accessibility Transcription Speech recognition Language
Translation
Case : Recognize sentiment in Text
None
Amazon and Yelp reviews 1 : Good 0: Bad
Pre-processing data
Text Processing
Tokenization
Out Of Vocabulary (OOV) EXTRA WORDS unknown tokens (UNK)
Stop Words
Stemming
Lemmatization
Padding
Tools! • Python 3.+ • Jupyter Notebook or Google Colab
• and using a framework for ML
¿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.
Trying in Google Colab bit.ly/2TU9MJz
Challenges in NLP : Textual Accessibility Transcription Speech recognition
None
Challenges in NLP : Textual Accessibility Language Translation
None
Next steps Try new tools : - Recurrent Neural Network
(RNN) - Long short-term memory (LSTM)
Next steps Increase your corpus with diverse data. - PDFMiner
- Tweepy API
Next steps Datasets with native or indigenous language or code
mixing
“We can build a much brighter future where humans are
relieved of menial work using AI capabilities” - Andrew Ng
Thanks! @womenwhocode #WWCode Keep in touch! linkedin.com/in/victoriaubaldo @vikyale
None