Data Management and Mining in Internet Ad Systems
New systems produce data that often present new challenges for data management and mining problems. For ex-ample, inventory and sales data led to emphasis on data mining problems such as association rule mining; analysis of Internet Packet traffic logs led to data stream management systems; and, growing markup publication systems led to challenges addressed by semi-structured data management. Here the authors' are inspired by systems that have emerged in the past decade that enable advertisements (ads) on the Inter-net. These systems handle billions of transactions every day involving millions of users, websites and advertisers, and are the basis for billions of dollars worth industry. They crucially rely on real-time collection, management and analysis of data for their effectiveness.