Data science is an inter-disciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from structured and unstructured data
Wikipedia
... to work effectively with heterogeneous, real-world data and to extract insights from the data using the latest tools and analytical methods.
UC Berkeley MIDS program brochure
However
When you search online this is what you see
Data Science is way more than Machine Learning!
The purpose of visualization is insight, not pictures
The purpose of data analytics is insight, not (just) models
What part of ML do you want to use?
Prediction
![ML Prediction](images/ML Prediction.svg)
How to use it (prediction)?
- Op1. Use models (apis) trained by Google, Amazon, IBM or Microsoft
- Op2. Download models pretrained by others and run them yourself
What can you use it for?
- Photos πΌ
- Videos πΉ
- Document/Text Processing π
- Speech πππΌ
- Structured data πΎ?
Photos πΌ and Videos πΉ
What can I detect on photos πΌ?
- Objects π π π
- Faces π±π½ββοΈπ±ββοΈ
- Celebrities πΎ
- Landmarks πΌ
- Text in images πΌ
Video πΉ is about the same but on streaming
How can I use it?
Develop locally
Document/Text processing π
What can I do with documents π?
- OCR πΌ β π€
- Sentiment analysis ππ‘
- Topic extraction π‘π π£
- Entities detection
- Political Affiliation? ππ
- Psychological Profile?
What can I do with Speech πππΌ?
- Speech recognition ππΌ
- Speech generation π
Let's compare them with a real world example
How is Rappi doing on Twitter?
- 30k tweets in a week of 2019
π‘π πππππ₯°?
- Machine learning π©! ???
- Detects sentiment ! ???
I hired a data π (might be me)
Analyzed 180 tweets
- π‘π πππππ₯°
Would you hire this data π?
Well, actually
- It wasn't a data π
- It was a π»
- Would you use it?
Will you trust it?
I don't
It's up to you!
- Interactivity π Ask questions
- Slice and dice
- Overview first, Zoom/Filter, then details on demand
Rappi Dashboard Link πΒ‘No coma Machine Learning, coma π!
My insights toolset?
![](images/My skills.png)
Training
![ML Training](images/ML Training.svg)
Multivariate Data?
-> Dimensionality Reduction + Clustering
Visual Analytics Applications
- Present information
- Discover insights
- Preprocess + Explore
- Open the black box
Present information
- Prediscovered insight
- General public
Discover insights
- For experts
- Navigate the data
Preprocess
- Understand new data
- Cleanup
MLExplore.js
- Interpret and interact with TSNE+Kmeans