Reveal relationships in unstructured news data
Monitoring the OCCRP
The Organized Crime and Corruption Project focuses on publishing investigations and news exposing issues to the public. Much of the source material for these investigations is made available through a search platform. However, even though the information is searchable it is still lacking true structure and cannot be analyzed to its full extent.
“Who are the associates of Paul Manafort?” is a difficult question to answer with unstructured data. Traditional NLP techniques like Named Entity Recognition (NER) will only help identify what names are in a text, not how these names relate to each other. A fundamentally more advanced approach is required to uncover the relationships between the entities in the text.
Built with Constellation
Constellation is the world’s most advanced Text Analysis Platform. Constellation read each investigation published by the OCCRP and automatically identified relationships between the people, organizations, and locations referenced in each article. This structured data powers the search tool provided above.
The memo was attached to one of thousands of emails obtained by the OCCRP that document a decade of correspondence between Manafort, Lovochkin, and others, starting in 2008.
This Relationship Investigator is designed to be an ongoing resource for investigators. Accordingly, the data is freely available to anyone under a Creative Commons license and can be downloaded or accessed via API here.
The Constellation Platform is the most robust platform for transforming unstructured information into enterprise intelligence. Constellation is utilized by leading companies to create structured data used for advanced analytics, information security, workflow automation, and more.