CAMEO Ontology Based Real-time Political Event Data

We are generating Structured Political event data in the "who-did-what-to-whom" format using the CAMEO (Conflict and Mediation Event Observation) ontology. The system runs 24/7 (from October 2017) on Apache SPARK based distributed system and relies on dynamic ontology extension for more accurate modeling of the raw text data. We consider 250+ news agency websites to collect news articles that are processed with Stanford CoreNLP/Universal Dependency (python udpipe package) parser to generate required metadata for the PETRARCH event coder. Finally, the data is served through an API and publicly accessible over the Internet around the world. A detailed description and API access policy are described here http://eventdata.utdallas.edu/