A large technology firm was storing massive content libraries of several national news organizations, but it was difficult for journalists to view, search, and derive insights from the material. Journalists were faced with large quantities of unstructured and disparate data and lacked the ability to research trends, information, and patterns. This made collaboration, investigation, and sharing difficult. The organization needed a new research, archival, and collaboration platform that could allow journalists to automatically identify insights, connections, and relationships between content.
Unify designed, custom built, and implemented an artificial intelligence document classification and interpretation engine for use by several high-profile global media companies. The Insights Discovery and Accelerator (IDA) utilized optical character recognition (OCR) to scan text and images to identify elements of metadata. The accelerator was designed to ingest any type of file (structured or unstructured) using OCR, image extraction, and video indexing techniques. The IDA focused on customizable data ingestion pipelines, intuitive user interfaces, easy scalability, and Artificial Intelligence (AI) that is flexible by design. The plug-and-play nature of the IDA solution allowed for many different AI approaches to be bundled into one user-facing solution.
The IDA can successfully analyze text, images, audio, and videos, enabling the organization to build an easy-to-use document archive. Journalists are able to review, search, and derive insights from relevant content, allowing them to quickly and effectively write articles. The increased media exposure has helped to grow the impact of the organization within the national discourse. As an added benefit, the scope of the implementation was able to be scaled beyond journalism, and has been expanded to other document collections within the organization.