Reconstructing Information Retrieved from Multiple Websites

Abstract

This work presents a model for information retrieval from multiple webpages. This model does not need to pre-process, parser or label the webpages; and thus, it can work online and in real time. The model introduces two new techniques for visualization that allow us to automatically reconstruct a new webpage with the information retrieved. This page is structured taking into account the semantically related information. The technique has been implemented and the implementation is discussed focussing on the main problems that appear when the proposed algorithms are integrated into a commercial web browser.