Articles from the New York Times are dsicovered via user generated keywords, and connections between articles are found via a conceptual sematic network.
For example, a search for "food" brings up an article on fajitas. This article has a keyword of "meat", which is semantically connected to the idea of a "butcher shop". Searching for this brings up an article at features an interview with a refugee who used to be a butcher.
