Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Get More Instagram Likes with Machine Learning

Get More Instagram Likes with Machine Learning

In this presentation, we will use Python and Cognitive Services to try and understand what makes people like photographs on Instagram more. We will also explore some great ways to run Python code in the cloud, including Visual Studio Codespaces, and some Python tooling, such as Pandas, Scikit Learn, Wordcloud, etc.

D8730970396729af16546cc69ce63b44?s=128

Dmitri Soshnikov

July 15, 2020
Tweet

Transcript

  1. Getting More Instagram Likes with Machine Learning Starting Soon…

  2. Getting More Instagram Likes with Machine Learning Dmitry Soshnikov http://soshnikov.com

    Cloud Developer Advocate, Microsoft Associate Professor, MIPT / HSE / MAI
  3. The Problem

  4. Data Science ♥

  5. Unstructured -> Structured

  6. Microsoft Cognitive Services Learn more about them http://aka.ms/cognitive_docs Home page

    http://aka.ms/cognitive_home MS Learn http://aka.ms/learn_vision
  7. How to Get Cloud http://aka.ms/az4stud http://aka.ms/azfree Photos by Arif Riyanto

    and Siora Photography at Unsplash
  8. Data Science = Python! How to Run it?

  9. Let’s Start Coding!

  10. Let’s Start Coding!

  11. Takeaways Data can Provide Incites In our example, we have

    explored the data and understood statistics of Instagram activities, most covered topics, etc. Data can Drive Decisions By exploring what affects outcomes we can take wiser decisions, e.g. what types of photographs to take, when to post, etc. Models can make Predictions We can build executable models that can predict desired outcomes Visual Studio Codespaces Great environment for any developer in the cloud
  12. We ♥ Feedback! https://aka.ms/Reactor/Survey 7796

  13. Dmitry Soshnikov http://soshnikov.com Twitter/ Instagram: @shwars

  14. None