Financial Crime Investigations Empowered by Semantics

Background

The International Centre for Asset Recovery (ICAR) is a specialized division of the Basel Institute on Governance, founded in 2006 as an independent body. Its main purpose is to help countries fight large scale financial crimes involving stolen assets from existing or former government members. The organization provides overall assistance, access to critical intelligence leads from public sources and training to local governments.

The Goal

To create a prototype of a sophisticated system that helps financial crime investigation of internationally traded national assets.

The Challenge

Illegal activities, money transfers or even stealing of national assets by former autocrats and dictators are problems that affect a great number of countries when the political systems are in transition, and the legislation system is still fragile.

Such fraudulent transactions drain from the national economies their limited financial resources. To overcome this problem, ICAR helps the Financial Intelligence Units (FIUs) of these countries in the areas of financial investigation. The fight against corruption and money laundering include asset tracking and recovery, and the provision of legal assistance.

The Solution: Asset Recovery Intelligence System (ARIS)

This project comes to life with the cooperation of Dow Jones, World-Check, and the EDGAR database. The platform is a web-based and secure service, which provides users with the asset-related profile of a named legal entity.

The platform helps financial investigators, analysts, and FIUs with tracking stolen assets. Various data sources are used to reveal information - for example, domain-specific documents from a bank or a financial institution. Other good sources such as Google search and news feeds delivered by the worldwide press agencies also supply the platform.

The system uses financial, and risk-related information from external analytical document banks as Dow Jones Watchlist data feed & Factiva news feed, and WorldCheck Politically Exposed Persons data feed. Apart from these external resources, Sirma AI (trading as Ontotext) uses its World Knowledge Base, which presents information about people, companies, and relations. The service is augmented with text analytics and helps extract named objects, relationships, and factual knowledge from text sources.

All extracted data is ingested in GraphDB. The advantage of using this RDF database is its fast inferencing and efficient outputs to all semantic queries. Drawing from a wide range of available data sources, the platform helps investigators discover hidden relationships between people, suspicious money transfers, companies and bank accounts.

The results

The system creates internal reports about the names of persons of interest, involved in corruption schemes, along with corruption cases reported in regional news agencies and other online information sources. It also provides intelligence officers with on-demand profiles of individuals, including information such as the person’s network, roles, and activities.

Over the years, ARIS has proven its capability to not only support the investigation process, recover stolen national assets and restore justice, but also to enhance international cooperation.

Read the detailed case study here