BDD-Based Combinatorial Keyword Query Processing under a Taxonomy Model

Subscription-Notification Service in Digital Library

Figure 1 shows the basic processing flow of our system. A document is represented in the digital library repository by a description of its content together with an identifier (say, the document's URI) allowing to access the document's content. A number of documents are incoming to the digital library every day, and the identifier list of recent new documents is notified to each user periodically (daily, weekly, or monthly). Each user can specify a subscription to show the composed keywords that are of interest to that user. Every document has a description with a set of keywords, and if one of the keywords matches to some users' subscriptions, the identifier of the document is added to the notification list. Here, we assume that a user's subscription is not so frequently updated as the notification periods.

Figure 1.

Basic processing flow of subscription-notification service

If we consider a nation-wide digital library system, the number of documents and the number of users become quite huge, and it is an important problem to reduce the computation cost for matching (or filtering) of the documents to each user.