1

I have what I think is a very basic question. What are the best practices for integrating a machine learning algorithm with Firebase data structure? To be more specific, I have a data structure that continually grows on Firebase that collects, essentially, survey data. Think Survey Monkey.

I also have a script written in python that pulls the relevant data I need out of the data structure and turns the survey responses into clean, usable numbers for the users.

The way I have been doing it is simply exporting the JSON from Firebase. Of course, as the data structure grows it is going to become unreasonable to do this on my local computer... I think.

What is the best practices for this workflow? I see that Google Cloud has some database management programs available. How do I do my pre-processing on the cloud, or at least move from this incredibly janky manual solution I'm currently doing, that is not scalable at all.

Thank you!

Ted
  • 22,696
  • 11
  • 95
  • 109
Quin_917
  • 11
  • 2
  • I don't see why the Datastructure matters. Firebase doesn't run machine learning algorithms, so you need to extract your model and reformat it anyway into scikit-learn or whatever – OneCricketeer Sep 21 '17 at 23:15

1 Answers1

0

now we have firebase predictions!

Firebase Predictions applies machine learning to your analytics data to create dynamic user groups based on your users' predicted behavior. These predictions are automatically available for use with Firebase Remote Config, the Notifications composer, and A/B testing.

please check the link for more info

https://firebase.google.com/docs/predictions/

Ted
  • 22,696
  • 11
  • 95
  • 109