Extraction and Geographical Navigation of Important Historical Events in the Web

We propose techniques for achieving the geographical navigation of historical events described in Web pages as “Virtual History Tour”. First, we develop a method for extracting information on the historical events from the Web and organizing it into a chronological table. Our method can effectively handle ambiguous cases – homonyms and multiple location names in a sentence – by using the number of co-occurrences among events, person names, location names, and addresses in the Web. Next, we propose a method for ranking historical entities according to their impacts at specific time and location. We extend the PageRank algorithm to calculate the temporal and spatial impacts of entities. Finally, we introduce our concrete application demonstrating how users can browse historical events through timeline and map interfaces