Text mining could reveal corruption in public procurement

Researchers at the University of Cambridge have developed a technique for trawling public procurement data to flag up potentially corrupt uses of public funds.

The Corruption Risk Index, announced on 15 June has been developed by Mihály Fazes, a research associate, and Lawrence King, professor of sociology and political economy, in the University of Cambridge’s department of sociology. The index is the basis of the Digital Whistleblower, a consortium of European institutes led by Cambridge that has just secured €3 million from the European research funding programme Horizon 2020.

The index seeks pre-defined ‘red flag’ contractual situations deemed to suggest corrupt behaviour by crawling through public procurement data, now widely available across the European Union, with algorithms and text mining programs. In this manner the team can map levels of corruption risk at a regional and national scale and can even pinpoint suppliers and individual contracts that look suspect.