Searching Distributed and Heterogeneous Digital Media: The VisualHarness Approach
Amit Sheth, Kshitij Shah, Krishnan Parasuraman, Srilekha Mudumbai · 1999
Current Web-based search engines do a reasonable job when dealing with primarily textual and in some cases semi-structured data. Web-based interfaces to traditional databases also allow us to exploit structured databases. However, there is little help when we wish to find relevant information in visual and so-called new digital media (with significant temporal and spatial components), which are being put on the Internet at an exponential rate. Today, usually a user needs to know the specific repository and use the specific access mechanisms and interfaces that have been provided by the repositories of such information. There is also little help when we wish to deal with a broad variety of heterogeneous media. What if we don’t know which type of digital media we need to target the query to? What if the response to the query is best satisfied by a collection of artifacts of different media? Thus one challenge in the current Internet environment is to have the ability to search and access information in various media, including visual media. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.