NetOwl Extractor’s Smart Geotagging capabilities support intelligent exploitation of geo-codable information found in text. Smart Geotagging combined with NetOwl’s rich semantic ontology provides unprecedented capabilities to expand the range of geospatial analysis for documents.
Intelligent Extraction and Disambiguation
Smart Geotagging automatically assigns latitude/longitude values to geo-codable entities in text extracted by NetOwl’s entity extraction. NetOwl disambiguates place names from other entity types (e.g., “Austin” as place vs. person) with high accuracy. For each extracted place entity Smart Geotagging intelligently performs location resolution, as many place names are ambiguous (e.g., “Springfield” and “Alexandria”). It selects the most appropriate location given the context provided by the text and using a combination of advanced natural language processing and geospatial calculations. It also provides additional intelligent features:
- Assigns latitude/longitude values to relative location phrases (e.g., “a town 10 km northwest of Paris”)
- Converts MGRS coordinates to latitude/longitude values
- Outputs a confidence ranking for each possible geotagging candidate
Smart Geotagging provides unique capabilities – thanks to NetOwl’s advanced link and event extraction – to geotag people, organizations, artifacts, and events in documents in addition to place names. For example, the user can find:
- Locations where a certain person visited
- Locations where certain events occurred
This advanced geotagging capability opens up new ways to exploit unstructured text data for geospatial analysis.
MultilingualSmart Geotagging supports multiple languages, including:
- Chinese (traditional and simplified)
- Persian (Farsi and Dari)
Language IDOffers a seamlessly integrated language ID capability where the language of the input text is automatically detected, and the text is processed accordingly. Both microblog and standard document lengths are supported. A mixed language document, where sections of the document are written in multiple languages, can also be handled automatically.
Name NormalizationEnables disambiguation and normalization of place names. Springfield could be located in Massachusetts, Illinois, or many other states. Historical names as well as spelling variants are also normalized, e.g., Burma ➞ Myanmar, Kyiv ➞ Kiev. Name normalization is ideal for cross-document name resolution for applications such as faceted search and link analysis.
Semantic DisambiguationRecognizes and classifies concepts using linguistic context. This sophisticated feature distinguishes semantic ambiguities like:
- "Ford" (place vs. person vs. company)
- "Obama" (place vs. person)
GazetteersSmart Geotagging includes NGA and USGS gazetteers and also provides an interface for easy integration with custom gazetteers.
Coreference ResolutionResolves co-referring extracted entities, whether they are name aliases, pronouns, or definite noun phrases, identifying them as referring to the same object. For example:
- "the city" ➞ "San Franciso"
- "his country" ➞ "United States"