The code can be found here
In this exercise, we have an app store dataset. Each user's application download history is shown by the app ids.
The goal is to recommend/predict the next 'n' apps that users are likely to download (recommend the apps to them).
I had two ideas to solve this problem:
1. Using RNN to predict next 'n' apps.
2. Using recommender systems (Cosine similarity)
The details can be found in the code