The Audience Interests: Cross-Browsing Behavior and Relevance Score are two metrics that help identify valuable websites that target audiences visit.
Cross-Visitation: Average percentage of users who visited both this website and the analyzed website on the same day.
"Average" is the aggregation of unique users that visited both websites that month, averaged per day.
Relevance Score: The relevance score is based on a joint audience; i.e., the score between pairs of sites is calculated based on users who visited both sites relative to the size of these sites.
A simple mathematical explanation (we are doing something close to) can be the ratio between the users who visited both sites to the users who visited at least once.
In addition, there are a few more considerations taken into the score, like the size of the two sites (so that amazon wouldn't get some small sites as the most relevant, for example), and the intent of the user in certain situations (i.e., if users who visited sites A and B, always starts with A the score between them will not be symmetric).
Cross-Browsing Behavior and Audience Interests data is based on unique visits.
These metrics will identify sites that share a percentage of users based on the selected time period.
To learn more, visit Understand Audience Interests.