Detecting corruption in public procurement – DeCoMaP

DeCoMaP lies at the intersection of these three considerations: regulatory developments, open data and automatic tools for fraud detection and economics analysis. Considering that public procurement data are essentially of a relational nature, we will noticeably use machine learning and graph-based approaches to model and automate fraud detection.

The core of our methodological contribution is to propose graph-based fraud detection methods. The relational information can help improve fraud detection and understand its occurrence. Graphs are the natural paradigm to model such information: vertices are used to represent the objects constituting the system (in our case: suppliers and buyers, tenders) and edges to represent the relationships between them (in our case: winning a tender). More precisely, we will use signed graphs as our main modeling paradigm.

- Production and opening of a database of French public contracts over a period of ten years
- Production and opening of a database of court decisions presenting cases of proven fraud in public procurement
- Production of automatic detection tools for fraudulent situations in public procurement data
- Comparison of the relative performance of data mining tools and identification of the added value of approaches based on graph analysis.
- Legal and economic analysis of public procurement vulnerabilities
- Legal and economic analysis of data essential to the control of public procurement

-

-

The societal benefits of opening up public data are expected to be huge. This is particularly true with Public Procurement Data which are supposed to help discover and dismantle corrupt activities by facilitating critical information, tools, and mechanism for judicial enforcement. In a multidisciplinary project, bridging computer science, economics and law, DeCoMap is intended to collect, process and analyze French procurement data in order to create a software tool for automatic identification of corruption and fraud in public procurement (automated red flagging) and provide normative analytical grid by highlighting the main factors that public authorities should identify and pay attention to.

Supported by Transparency International France and Open Contracting Partnership, DeCoMap brings together academic researchers from 7 universities, with strong expertise in procurement and digital law, procurement economics and econometrics, law and economics, graph optimization and complex network analysis. 4 members of Datactivist, a cooperative company that assists organizations from the public, private and non-profit sectors in producing and re-using Open Data, with strong expertise with open data of public procurement, open contracting and open government, complement the consortium.

The project will set up of a comprehensive database of corruption and fraud cases, collecting empirical pieces of evidence from various and heterogeneous legal sources from primary and secondary source documents as well as a survey of procurement experts.

The data requirements for the development and application of the comprehensive methodology are substantial and exceed in terms of volume and complexity any specific data source on corruption and fraud in French public procurement that is known to us. DeCoMaP will provide an innovative law and economics analysis of this ground truth data.

We'll propose different tools and methods to extract signed graphs from (raw or enriched by the analysis conducted) data.

These tools will be complemented by more classical approaches, like econometrics models and machine learning approaches.

In addition to that, the complementary challenge here is the identification of a corruption index in order to enrich the information available in the graph both at the level of its links (contracts between a public buyer and a supplier) and at the level of its vertices (characteristics of the actors).

The results obtained with these methods will guide us in the development of hybrid approaches to fraud detection. Finally, the consortium will calibrate the classification tools based on the ground truth dataset of corrupted procurement practices obtained at the beginning of the project.

By applying the developed tools over the periods on which both market data and court decisions are known, we will be able to identify problematic situations and legal vulnerable environments (types of procedures, nature of clauses, characteristics of actors, etc.), to inform public decision making. Particular attention will be paid to data formats and data mobilized, in order to contribute to the debates on open data and predictive justice.

The production of a functional data-visualization tool for the general public will conclude the project. It will also enable us to analyze and discuss the results obtained with our tool, in a compendium of good practices, for public buyers and legal authorities, in order to identify which criteria could be used to improve real-life public procurement practice.