liveBook · Manning

Compliance due diligence

As explained in Fight Fraud with Machine Learning

Customer due diligence is the process of assessing and managing risks associated with a customer's identity, activities, and relationship with a regulated entity.

Compliance due diligence (CDD) assesses whether a customer can be trusted and determines their risk level within the broader KYC process. section 7.0

Unlike customer identification, which verifies claimed identity, CDD evaluates customer metadata; unlike ongoing monitoring, it focuses on onboarding risk assessment. section 7.0

CDD can use metadata such as:

  • Mobile device type and version used during signup. section 7.0
  • Country or location associated with the customer. section 7.0
  • Credit score and salary level for risk assessment. section 7.4

Automated onboarding risk assessment

A regulated entity can provide customer metadata to a tabular machine-learning model that estimates onboarding risk, while separate systems verify identity and monitor later transactions. section 7.0

Figure 7.1 The 3 different components of KYC along with the various types of KYC tasks listed per component.

Use the KYC component map to distinguish CDD from customer identification and ongoing monitoring. section 7.0

How Fight Fraud with Machine Learning covers it

How deep it goes
Defines CDD inputs and separates it from identification and monitoring, but defers detailed model implementation.
Examples
Uses device type, location, credit score, and salary level as customer-risk metadata examples.
What it assumes
Assumes familiarity with KYC components and customer metadata while defining CDD’s purpose and inputs.
What you can do after
Apply it via tabular machine-learning models to assess customer risk from user metadata.
How it connects
Connects CDD with KYC, customer identification, ongoing monitoring, tabular machine learning, and transaction fraud detection.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Fight Fraud with Machine Learning. The section links point to the parts of the book it draws on.