Automated information extraction
from unstructured data sources

Challenge

For supervisory analyses, structured information is needed from thousands of unstructured data sources, such as financial reports and press releases. At present, information is retrieved manually or not at all, resulting in significant loss of time and increased risk of missing important content. There is a need for an intelligent automated solution for detection of tabular data and recognition of contained information.

Solution

We assisted our client in creating a tool which retrieves data from unstructured data sources including internal and external reports, press releases and certificates.Documents which had served as data source were in image or pdf formats.

Result

Automated data retrieval from 30 data sources

Client - European financial authority