Model Online / On-Site
Duration
Modules 5 modules

Course Objective & Scope

A high AUC does not mean a model creates business value. This programme links technical metrics to business measures such as revenue, cost or risk, and covers segment-level error analysis, calibration and group fairness metrics; it then makes model behaviour visible with SHAP, LIME, partial dependence and counterfactual explanations.

Participants prepare an evaluation report and an explainability output for a model adapted to their organisation, structured so it can be shared with business units and audit teams.

Who Should Attend?

  • Data scientists and ML engineers
  • Analytics product managers
  • Model risk and internal audit teams
  • Prerequisite: basic model training experience

Key Takeaways

  • Define a metric set that fits the problem and business goal
  • Translate model performance into business terms with profit/cost curves
  • Find weak spots through segment-level error analysis
  • Produce global and local explanations with SHAP and LIME
  • Compute and report fairness metrics

Curriculum

Metric Selection and Business Alignment
  • Classification, regression and ranking metrics
  • Profit curves, lift and cost matrices
Error Analysis and Calibration
  • Segment and slice-level performance review
  • Reliability diagrams and calibration fixes
Global and Local Explanation Methods
  • SHAP, LIME, permutation importance
  • Partial dependence and ICE plots
Fairness and Responsible Evaluation
  • Demographic parity, equalized odds and Fairlearn
  • Limits of explanations and misinterpretation risks
Evaluation Report
  • Evaluation and explainability output in model card format
  • Structuring the narrative for stakeholder presentation

Other Courses in This Category

  • Model Validation and Model Risk Management

    Independent model validation: data and conceptual soundness checks, backtesting, stability (PSI), challenger models, stress tests and findings reporting, producing a validation report aligned with a model risk management framework.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Time Series Forecasting

    Time series forecasting with ARIMA, ETS, Prophet and LightGBM-based approaches, working hands-on with seasonality, exogenous variables, hierarchical series, backtesting and prediction intervals to build reliable forecasts.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Anomaly Detection with Machine Learning

    Detecting unusual behaviour in unlabelled data with statistical methods, Isolation Forest, LOF, autoencoders and time series anomaly techniques, including threshold setting, alert fatigue and evaluation strategies.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Applied Machine Learning Applied Machine Learning

    End-to-end machine learning with Scikit-learn, XGBoost, LightGBM, and CatBoost, covering feature engineering, Optuna hyperparameter tuning, imbalanced data, and SHAP/LIME explainability.

    Machine Learning & Applied Data Science Corporate, Individual
    Live Course
    3 modules
    View
  • Fraud Detection with Machine Learning

    Behavioural features, gradient boosting, anomaly scores and hybrid rule-model setups for payment, insurance and account fraud, with hands-on work on extreme class imbalance, delayed labels and operational threshold management.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Risk Scoring and Credit Scorecard Modelling

    Variable analysis with WoE/IV, logistic regression scorecards, gradient boosting alternatives, reject inference and score calibration for credit, insurance and supplier risk, building explainable and auditable risk scores.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View

Get in touch