Web-based geospatial information extraction

Abstract

The quantity of information publicly available on the World Wide Web today is enormous and continuously growing. Individual tools available for information retrieval and data mining are capable of extracting data based on search criteria but place a cognitive burden on the user to sift through the search results to find the information of interest. The dynamic nature of web-based information requires frequent searches to ensure that new information is not missed. Since public information is published on the web in many languages, translation into a common language is necessary to ensure that relevant data is not left out of a search. The authors discuss development of a programmatic method of collecting, analyzing and categorizing multi-lingual open source data. The discussed approach is designed to provide greater efficiency in public data searches and analysis by focusing on geospatial capabilities, visualizations, and the correlation of disparate pieces of information.