We have a few ways to sort through questions:
Five out of these six choices are self explanatory, but I'm curious about the 'frequent' heading. How is a question deemed as such? Is it based on the number of times it is linked to from the comments? Or how many times it is linked from a closed duplicate? How many other factors are there? :)
The frequent questions sort order represent questions that have at least one link pointing towards it. At the top of the list is the post that has the most links point to it.
To see how these links are measured, consider the "most frequent" question How to look up a symbol or identify a math symbol or character? (as of 2013-04-16). In the right-hand column it lists the questions that are linked:
By clicking on the see more linked questions... link, you'll see the search query launched to obtain the count:
url:"https://tex.stackexchange.com/q*/14/*"
Most of these posts have How to look up a symbol or identify a math symbol or character? marked as a duplicate (note the post snippets...) using the automated edit which inserts the duplicate URL. However, some may contain links to the post itself as a form of reference.
It searches for any post (not a comment) that contains the (wildcard) URL https://tex.stackexchange.com/q*/14/*. The q* wildcard captures both default and personalized links to the specific question.
