I'm create a application to log everything user purchase to drink.The drink have 3 value to provide the taste of the drink.So how can i use Firebase Prediction to generate 3 predicted value about what user like to drink to create a list about "Suggest Drink" for user base on 3 predicted value.
2 Answers
From the documentation:
By default, Predictions provides two types of predictions: churn, which helps you identify users likely to stop using your app (that is, they will not open the app or app-related notification messages), and spend, which helps you find users who are likely to spend money in your app. You can also create your own predictions based on custom conversion Analytics events that you collect in your app.
Other types of prediction are currently not possible with just Firebase Predictions.
What you could do is train your own custom Machine Learning model on the analytics data, and use that custom model that in your app using ML Kit.

- 565,676
- 79
- 828
- 807
-
sorry cause im late response,this is the accepted answer.trying to use Tensor Flow as well – shadow Aug 28 '19 at 06:22
Since Firebase Predictions can not predict based param value of an Analytics event, one approach for you would be creating three types of Analytics events for each possible value and create three custom-predictions for each of them. The new version of the Firebase Predictions actually calibrates the probabilities for you which means the score for these three different predictions are comparable.

- 1