Skip to main content

Compliance Management for Insider Trading

Designing an AI-Driven Semantic Search and Classification System for Insider Trading Surveillance

The drawback of conventional search engines like Google and Bing is that they cannot access proprietary data sets. The compliance team within a French banking giant maintained an internal database of PDFs indicating whether certain statements should be flagged. Traders often used keywords or phrases that could suggest insider trading, with the compliance team needing to monitor Bloomberg chat logs to detect this. A high volume of false positives meant the team had to manually review documents to determine whether a flag was valid.

DataSpartan were asked to solve the problem by creating a tool which would allow them to search their internal databases faster for the relevant information.

A custom interface was built using Django which produced PDF previews for key words and phrases to allow the officers to preview the documents manually and by eye for relevancy.

Our experts then helped implement a larger system which includes a document recommendation engine highlighting the usage of the keywords explicitly.

The component is integrated into the client servers and is projected to save each officer 3 hours of work each month. Full integration of the larger system is projected to save 10 hours of work each month.