Supervised Machine Learning

Turn confirmed outcomes into sharper fraud decisions

Build and deploy machine learning models that learn from your labeled outcomes and apply that intelligence to real-time risk decisions. DataVisor connects model development, features, explainability, deployment, and monitoring to the same platform where strategies run.

Request a Demo →
Fraud & Risk Platform/Supervised ML
Solution Pillars

Bring model intelligence into the decision layer

Train on your labeled outcomes, move scores directly into production strategies, and keep performance visible after deployment.

Learn from your outcomes

Use known fraud and good activity to sharpen detection

Train models on labeled behavior and the features that matter to your business so detection reflects the patterns present in your customers, products, and channels.

Trained on your labeled outcomes and features
Put models into production decisions

Move from score to action

Use model scores directly within real-time strategies alongside rules, unsupervised ML, graph intelligence, device signals, and other context.

Scores used directly in live strategies
Build on a shared feature foundation

Keep model and strategy intelligence aligned

Use the platform's feature layer to make relevant risk intelligence available across models and rules, reducing duplicated feature logic and inconsistent definitions.

One feature layer shared across models & rules
Monitor and evolve

Keep performance visible after deployment

Track model behavior and performance over time to identify changes that warrant review, comparison, or retraining. Keep teams informed as data, customer behavior, and fraud patterns evolve.

Ongoing monitoring, review, and retraining
Product Proof

The full model lifecycle

Build and configure models, evaluate performance, understand feature contributions, and deploy model intelligence directly into live decision strategies. Manage the lifecycle in the same environment where fraud decisions are made.

Train
model
→
Evaluate
performance
→
Explainability
→
Deploy to
strategy
Training, evaluation, explainability, and deployment — the model lifecycle from first version to production.
How It Works

From labeled outcomes to live risk scores

1

Prepare labeled outcomes and features.

2

Train and validate the model.

3

Review performance and explainability.

4

Deploy the model into decision strategies.

5

Capture new outcomes and monitor performance over time.

ML connected to the decision layer

Keep model intelligence operational

DataVisor brings supervised models into the same environment as real-time features, unsupervised detection, rules, graph intelligence, and decisioning so a model becomes part of a broader risk strategy instead of another isolated score.

Train for the risks you know

Apply supervised models where outcome history is valuable

Use supervised ML for account takeover, application fraud, card and payment fraud, ACH/wire risk, scams, and other use cases with meaningful labeled outcomes.

Account Takeover Application Fraud Card & Payment Fraud ACH / Wire Risk Scams
Measure model impact

Tie model performance to business performance

Model impact is measured against your organization's detection, false-positive, and deployment-speed goals, with each model validated through feature-importance review and governance documentation before it reaches production.

Connected Platform

Combine known-pattern learning with emerging-threat intelligence

Connect supervised models to Data Orchestration, Rules & Features, Unsupervised ML, dEdge, Knowledge Graph, Real-Time Decisioning, and Analytics.

Real-Time Data Orchestration → Rules, Features & Workflow Automations → Unsupervised Machine Learning → Device & Behavioral Intelligence — dEdge → Knowledge Graph & Link Analysis → Real-Time Decisioning → Analytics & Reporting →
Model governance for production risk

Make performance understandable and controllable

DataVisor pairs explainability and feature-contribution views with rigorous model validation — train, validation, and test-split evaluation, feature-importance review, and governance documentation — so teams can trust a model before it goes live, then keep capturing outcomes to guide future retraining. Externally trained models can also be brought in and served in production decision flows.

  • Explainability and feature-contribution views in alert details
  • Train, validation, and test-split evaluation with feature-importance review
  • Governance documentation supporting model validation before deployment
  • Outcome capture to guide future training and improvement
  • Support for bringing externally trained models into production decision flows
FAQ

Common questions

What data is needed to train a supervised fraud model?

Supervised models learn from labeled outcomes and the associated features. The exact data requirements depend on the use case, available history, and target decision.

Can supervised and unsupervised ML be used together?

Yes. DataVisor can combine supervised models trained on known outcomes with unsupervised detection designed to surface new and coordinated behavior.

How are model decisions explained?

DataVisor provides model-explanation widgets directly within alert details, surfacing the feature contributions behind each model-driven decision so analysts can see why a score or outcome was produced.

See the model inside the strategy

Move from model score to business action

Walk through how DataVisor trains, evaluates, deploys, and uses supervised intelligence in a real-time fraud decision.

Request a Demo →