ML and banking: Improving fraud detection

Fraud and Anomaly Detection in Banking

Cover

The scale of fraud detection in banking is staggering, with the U.S. Federal Trade Commission logging millions of fraud-related complaints within the last 5 years.

Unfortunately, traditional rule-based fraud detection systems are often insufficient, showing false positive rates that exceed 90% and are purely reactive.

AI-based systems on the other hand are proactive and can augment regulatory alert systems and improve analysts’ workflow by reducing noise without discarding results.

Access this paper to view the 3 types of anomalies to keep an eye out for, understand examples of how banks use fraud and anomaly detection, and learn how to build a basic, machine learning-based fraud detection system.

Vendor:
Dataiku
Posted:
14 Jan 2020
Published:
14 Jan 2020
Format:
PDF
Length:
20 Page(s)
Type:
eGuide
Language:
English
Already a Bitpipe member? Login here

Download this eGuide!