Storing results of web page segmentation

Segmenting the web page is one of initial steps of information retrieval process performed on that page. While there has been an extensive research in this area, storing the results is usually only a marginal topic. When we consider principles of modern web page design like templates, it is possible to reuse the segmentation result for more pages than just the one it was created with, thus improving the performance of web site processing engines. This paper analyzes what structures have to be stored and what databases are convenient for this purpose in order to enable and support reusing the segmentation algorithms' results.