| 摘要 |
An affinity generation system according to the present invention analyzes a profiled user's authorship and document usage within an 'intranet' to create a set of affinities between documents and topical classifications used in a hierarchical content catalog. These affinities are weighted depending on the system usage and amount of collected evidence. Once a certain threshold has been reached, the threshold being fixed or dynamically set to achieve a desired policy, the affinities are published into the content catalog. The user looking for specific expertise can then search or browse the content catalog and find both documents and people with strong affinities to the selected content area.
|
