Digital images and video clips are becoming popular due to the increase in the availability of consumer devices that capture them. Digital content is also growing over the Internet. Applications that benefit from video are education and training, marketing support, medical, etc. The increase of this digital content creates a need for user-friendly tools to browse through large volumes of digital material. However, there are two basic impediments to wider use of digital video. The first is cataloging, which includes video digitization, compression and annotation, and the second is the lack of fast and effective search and browse techniques for this massive video content. The authors are interested in this second problem. One method that they believe is promising is the augmentation of a metadatabase with information on video content so that users can be guided to appropriate data sets. An automated technique is presented that combines manual annotations and knowledge produced by an automatic content characterization technique (i.e. clustering algorithms) to build higher level abstraction of video content.