How best to test recommendation engine results in a user study?

1. How do you check the accuracy of a recommendation engine?
2. How do you test a recommender system?
3. How do you evaluate a content based recommender system code?

How do you check the accuracy of a recommendation engine?

The click-through rate (CTR) is a metric that measures how many people click on the recommendations. The basic notion is that if more people click on the recommended things, the recommendations are more relevant to them. In news recommendations, the CTR is a widely used metric.

How do you test a recommender system?

For recommender systems, the solution is offline evaluation, where historical data is used to estimate how a user might have reacted to a different set of recommendations placed in front of them at a certain point in time, by using the knowledge of what they really did react to later.

How do you evaluate a content based recommender system code?

The top-N documents with the highest similarity scores is the recommended list for a user based on a specific query. After that, you can identify a value for each document in the recommendation list either relevant or non-relevant based on the similarity with the content of the query and the recommended document.