Geek Answers Handbook

How to create a recommendation system on Amazon Machine Learning

I wonder if I can create an Amazon Machine Learning recommendation system. It doesn't seem to be that simple (according to https://forums.aws.amazon.com/thread.jspa?messageID=617667 ), and I think if Amazon is the best solution for this.

Perhaps someone has experience with recommendation systems and knows how to create it on Amazon ML?

If not, can you offer me other SaaS that I can use?

+5
source share
1 answer

It depends on how many items you need to choose for your recommendations. You cannot use Amazon ML today for recommendations on a huge catalog (for example, Amazon.com or other large retailers), but if you have a relatively small catalog (~ 100 items), you can build a set of binary classifiers for Items.

The quality of the ML model depends on the quality of your data. For example, Amazon.com can make very good recommendations because they have a purchase history for many of your purchases.

A common recommendation mechanism model is collaborative filtering. It can work very well for a relatively large number of users and elements. This and other similar ML algorithms are implemented in frameworks such as Spark ML ( http://spark.apache.org/docs/latest/ml-collaborative-filtering.html ). It takes some technical and scientific and technical efforts to make it work for your use case, but usually it’s worth the effort with pretty good results.

You can also use some of the following services:

+6
source

More articles:


All Articles