How to build an integrity monitoring platform to minimize fraud, error and waste within a complex market?

Anomaly detection is just one of the methodologies the RoZetta data science team employs to fulfill the value creation goals.

This paper has been prepared by:

• Phil Anderson, Head of Product Management & Strategy
• Scott Matthews, Chief Data Scientist
• Dr Inigo Jauregi Unanue, Data Scientist

Introduction

Some industries rely on third parties to deliver complex services to end-users, including health services, education, wealth management, capital markets, taxation services and some franchise operations.

This environment is susceptible to error, waste, abuse, and fraud as the service deliverer, or intermediary, behave in a way inconsistent with the organization’s intention and expectations.

Anomaly detection also referred to as outlier or novelty detection, identifies events or observations that deviate significantly from the expected result, or a majority of the data, or do not conform to a well-defined notion of normal behavior. These are often related to behavioural events of error, waste, abuse, or fraud.

This article covers focuses on anomaly detection, an important tool in monitoring human and systems interactions that are too complex for a human alone to comprehend.

For more on anomaly detection see our interview with David Wright, Executive Director and Chairman, RoZetta Technology.