How to improve my recommendation result? I am using spark ALS implicit

Question

First, I have some use history of user\'s app.

For example:
user1, app1, 3(launch times)
user2, app2, 2(launch times)
user3, app1, 1(launch times)

Answer 1

There are two things you should try:

1. Standardise your data so that it has zero mean and unit variance per user vector. This is a common step in lots of machine learning. It helps to reduce the effect of outliers, which cause the close-to-zero values you are seeing.
2. Remove all users that have only a single app. The only thing you will learn from these users is a slightly better "mean" value for the app scores. They will not help you learn any meaningful relationships though, which is what you really want.

Having removed a user from the model, you will lose the ability to get a recommendation for that user directly from the model, by providing the user ID. However, they only have a single app rating anyway. So, you can instead run a KNN search over the product matrix to find apps most similar to that users apps = recommendations.