Associated Information Extraction for Enabling Entity Search on Electronic Map

摘要

Electronic map plays an important role today in our daily life. People use it to find out where a restaurant is before they have a dinner with their friends, or use a navigation device to search a location. These all attributed to the geographic system and the technology of information retrieval. In tradition, the construction of geographic database system costs a lot of human efforts and is not efficient. The goal of this research is to make use of the rich information on the web for automatically constructing the geographic database system.Our system could be separated into two parts: the IE (information extracting) subsystem and IR (information retrieval) subsystem. The first one is to extract the associated information of postal addresses from webpages for constructing the geographic database. Based on the state-of-the-art technology for postal address extraction, we locate via the difference of each address’s path and find out a largest disjunctive subtree as the associated information of each address. The second one applies information retrieval technology to rank the associated information based on users’ queries.