At first glance, of course they are. But if “Jack London” wrote about Helen of Troy’s suitor “Paris” and his desire to name their first daughter “Virginia,” then of course these are not places.
By leveraging NetOwl’s entity extraction to identify and disambiguate different entity types with high accuracy, only actual place names are geotagged in documents.
Even among place names, however, there is ambiguity about which real-world location is being discussed. When “Paris” is identified as a location, which one is it? In France? In Texas? Somewhere else?
Among all the possible candidates from the gazetteer data, NetOwl’s Smart Geotagging selects the most appropriate location given the context provided by the text through a combination of advanced natural language processing and geospatial calculations. “I drove from Paris to Dallas” is most likely to be the “Texas” location, while “I visited Paris on my European vacation” is certainly discussing France. Smart Geotagging outputs a confidence ranking for each possible candidate.
In addition to geotagging named places, NetOwl assigns a latitude/longitude to relative locations like “20 miles southwest of Paris” by calculating it from the base location offset by the direction and distance.
NetOwl also extracts MGRS and UTM coordinates from documents and converts them to latitude/longitude values.
Leveraging its intelligent relationship and event extraction capabilities, NetOwl geotags people, organizations, artifacts, and events mentioned in documents in addition to place names. This advanced geotagging capability opens up new ways to exploit unstructured text data for geospatial analysis. For example, the user can find who travelled to a specific place or where attack events took place geospatially on a map.
Instead of displaying all place names extracted from texts on a map, relationship and event extraction provides semantic context to place names, allowing the users to focus on only those that are important to them. Through an integration with the leading GIS provider ESRI’s ArcGIS suite, a NetOwl Extractor service with Smart Geotagging enabled allows individual users to process their own documents and generate a map view showing information about all linked entities and events that have been extracted for all of the locations in the text.
Provides state-of-the-art accuracy on many different types of text. Fast and scalable for real-time analysis.
Supports multiple languages, including English, Arabic, Chinese (traditional and simplified), French, German, Korean, Persian (Farsi and Dari), Russian, and Spanish.
Provides an interface for easy integration with custom gazetteers. NetOwl includes USGS and NGA gazetteers.
Contact us and discover what NetOwl can do for you!
