Uncovering Adverse Information with Entity Extraction: a Use Case

Entity Extraction, Geotagging, Homeland Security, Intelligence Analysis, Risk Management

Financial crime is a big problem that continues to grow. In response, governments are imposing increasingly stringent requirements on organizations that might be exposed to or exploited for financial crime. These institutions need to make sure that their prospective customers and business partners are legitimate and carry no questionable or criminal baggage.  As part of their Know Your Customer (KYC) obligations, they must carry out conventional due diligence and screening against watch and sanctions lists as well as Politically Exposed Person (PEP) lists. They must perform this screening at onboarding time and also on an on-going basis.

This new regulatory environment has meant that many organizations need outside help, and one of our clients has set up a comprehensive program to provide it.  Their program provides support on all aspects of customer onboarding from screening against sanctions lists to helping their clients keep up to date with evolving regulatory requirements. Our client also realized that today it is possible to find adverse information in news reports and social media on a real-time basis using AI-based technology. Capturing this information would represent a big step beyond the coverage provided by the usual sanctions and PEP lists, since they are always somewhat behind the times. That’s where NetOwl Extractor comes in.

How NetOwl’s Entity Extraction Supports Improved Customer Onboarding

After an extensive evaluation, our client chose NetOwl Extractor as its adverse media monitoring tool not only because they found NetOwl to be highly accurate and scalable but also because it offers unique, AI-based relationship and event extraction that is relevant to adverse media monitoring. The goal of our client is to get a total picture – not just of all prospective customers, but also of their executives, partners, hires, consultants, etc. and to locate all adverse information about them that would increase risk exposure.  News stories from around the world are, of course, a major source of valuable adverse information. Social media is also a great information source. Think about a company’s officials being indicted, arrested, or prosecuted for involvement in a crime such as bribery, corruption, or conspiracy. Certainly any company contemplating doing business with a compromised entity should be aware of the risk.

The Power of AI-based NetOwl Extractor to Mine Big Data

NetOwl Extractor provides automatic identification of semantic concepts in text, such as people, organizations, and places. Being able to recognize named entities is important and necessary but not sufficient to address the risk management challenge. To be able to find adverse information, an Entity Extraction product like NetOwl must also be able to identify advanced semantic concepts such as relationships and events. For instance, given a company of interest, NetOwl’s Relationship Extraction will allow you to discover news involving people associated with it – perhaps high-ranking officials or even their relatives. Most important, NetOwl’s advanced Event Extraction will reveal events containing adverse information (e.g., arrests, indictments, crimes) about the company’s employees and associates.  NetOwl’s event ontology includes over 100 types of events out of the box.

How NetOwl Extractor Works

NetOwl works by applying Natural Language Processing (NLP) technology to analyze unstructured data. Its major strength is that it can recognize previously unknown instances of names, relationships, and events by analyzing the linguistic contexts they occur in. You may be well aware of who’s in the org chart of a prospective customer, but what you may not know are very recent personnel changes at the executive level or the new relationships the existing executives may have acquired recently – their new relatives, associates, etc. That’s where NetOwl shows its strength: it is able to detect the presence of new names, new relationships, and new events in unstructured text and report them to you.

Summary

Our client has a broad solution that addresses all the needs of financial institutions in combatting financial crime. NetOwl contributes a critical capability to it. Identifying adverse information in today’s fast-moving global economy requires real-time mining of text-based Big Data from from a world-wide range of sources. NetOwl Extractor’s advanced, AI-based entity, relationship, and event extraction provides the key capabilities required to meet their requirements in support of adverse information collection and ultimately risk management