Statistical versus symbolic parsing for captioned-information retrieval

Our MARIE project has been investigating information retrieval of multimedia data using a new idea: putting primary emphasis on caption processing. Although content analysis methods such as substring searching for text media and shape matching for picture media can obviate captions, content analysis usually requires unacceptably-large amounts of time at retrieval time. Captions can be cachings of the results of content analysis, but they can also include auxiliary information like the date or customer for a photograph. Since captions can be considerably smaller than the media-da ta they describe, checking captions before retrieving media data can save time if it can rule out many bad matches quickly, the idea of "information filters" (Belkin and Croft, 1992).