Back to results
US 10,592,310 B2Granted
System and method for detecting, collecting, analyzing, and communicating event-related information
AssigneeGeorgetown University
InventorsPeter Chao-yuan Li, Jeffrey R. Collmann, Jane W. Blake, Mark G. Polyak, James M. Wilson, Jae In Yoon, Manabu Torii, Carla S. Thomas
Life of this patent
Estimating the term…
Abstract
A system and method involves detecting operational social disruptive events on a global scale, modeling data in conjunction with linguistics analysis to establish responsive actions, and generating visualization and executing models for communicating information.
Claims24
A system for detecting food security events or precursors to food security events, the system comprising:
- an information collection and processing subsystem comprising at least one repository database stored in non-transitory computer readable storage media containing documents downloaded from at least one publicly available information source;
- an information analysis and reporting subsystem that:
- stores indirect indications and warnings associated with socially disruptive events, the indirect indications and warnings being associated with food security events;
- and
- searches the documents for the indirect indications and warnings associated with food security events;
- a text classification system that uses a machine learning model or a combination of machine learning models to identify one or more documents that suggest that an unreported food security event or a precursor to an unreported food security event is unfolding based on the indications and warnings;
- and
- an information communications subsystem that outputs the documents that suggest that a food security event or a precursor to a food security event is unfolding.
5 more elements
A method of detecting food security events or precursors to food security events, the method comprising:
- storing, by an information analysis and reporting subsystem, indirect indications and warnings associated with socially disruptive events on non-transitive computer readable media, the indirect indications and warnings being associated with food security events;
- downloading, by an information collection and processing subsystem, documents from at least one publicly available information source;
- storing the documents in a repository database stored in non-transitory computer readable storage media;
- searching the documents, by the information analysis and reporting subsystem, for indirect indications and warnings associated with food security events;
- and
- identifying, by a text classification system that uses a machine learning model or a combination of machine learning models, one or more documents that suggest that an unreported food security event or a precursor to an unreported food security event is unfolding based on the indications and warnings.
3 more elements
Dependencies are inferred from the claim text.