Domain name registrars and URL shortener service providers placeadvertisements on the parked domains (Internet domain names which are not inservice) in order to generate profits. As the web contents have been removed,it is critical to make sure the displayed ads are directly related to theintents of the visitors who have been directed to the parked domains. Becauseof the missing contents in these domains, it is non-trivial to generate thekeywords to describe the previous contents and therefore the users intents. Inthis paper we discuss the adaptive keywords extraction problem and introduce analgorithm based on the BM25F term weighting and linear multi-armed bandits. Webuilt a prototype over a production domain registration system and evaluated itusing crowdsourcing in multiple iterations. The prototype is compared withother popular methods and is shown to be more effective.
展开▼