The architecture of corporate information and news engine
Panayot Dobrikov, Preslav Nakov · 2003
The paper describes the architecture of a corporate information and news engine, providing an effective bass for analysis, selection, navigation, and presentation of large sets of dynamic corporate contents. We show that common text processing techniques, when combined in an appropriate way, can handle a variety of sources, including: e-mails, structured and unstructured documents from internal Web sites, files shared over the LAN, etc. The major factors the system takes into account in order to offer a fully automated data gathering and personalized presentation include: document recency, sender's authority, closeness to a pre-selected topic and the user's interests.