7.4. Publish/Subscribe relevance
In top-k approaches, notified items are computed on the whole set of items, leading to delays of item delivery. In our approach, we consider the set of notified items for a given subscription, the so-called subscription history, and we use this history to filter out in real time the incoming item just after the matching process. This section presents our approach and the definitions adopted, and the instantiation is presented in section 7.5.
7.4.1. Items and histories
In our context, we define an item as a set of terms. Each term is associated with a term weight denoted by wi, which is used to compute distances and similarities. To compute novelty and diversity, a Pub/Sub system must keep already notified items, also called subscription history H. Each one is a time-ordered set of items linked to a subscription. Each time an item is notified for a subscription, it is added to its history.
7.4.2. Novelty
The objective when filtering by novelty is to...