Course Objective & Scope
Recommending the right item to the right user directly drives conversion and basket size across e-commerce, media, finance and retail. This programme covers user-item interaction matrices, explicit vs implicit feedback, neighbourhood collaborative filtering, matrix factorisation with ALS and BPR, content-based similarity and second-stage re-ranking architectures with LightGBM/CatBoost.
Cold-start, popularity bias and diversity trade-offs are addressed alongside offline metrics such as NDCG@k, Recall@k and MAP and online A/B test design. Participants build a candidate generation and ranking pipeline on an interaction dataset similar to their organisation's and produce an evaluation report.
Who Should Attend?
- Data scientists and ML engineers
- E-commerce, digital product and personalisation teams
- CRM and marketing analytics specialists
- Prerequisite: Python, pandas and basic machine learning
Key Takeaways
- Construct user-item interaction matrices from implicit and explicit feedback
- Train collaborative filtering and matrix factorisation models with implicit and LightFM
- Design cold-start strategies for new users and new catalogue items
- Combine candidate generation with a gradient-boosting re-ranking stage
- Evaluate recommenders with NDCG@k, Recall@k, coverage and diversity metrics
Curriculum
Recommendation Framing and Interaction Data
- Explicit vs implicit feedback and negative sampling
- Temporal train/test splits and leakage control
Collaborative Filtering and Matrix Factorisation
- User-based and item-based similarity
- Latent factor models with SVD, ALS and BPR
Content-Based and Hybrid Approaches
- Building profiles from item and user attributes
- Cold-start and hybrid modelling with LightFM
Two-Stage Architecture and Re-Ranking
- Candidate retrieval and feature enrichment
- Ranking with LightGBM ranker and business rules
Evaluation, Bias and A/B Testing
- Precision@k, Recall@k, MAP and NDCG@k
- Popularity bias, diversity and online experiment design
Other Courses in This Category
-
Failure probability and remaining useful life (RUL) prediction from sensor, telemetry and maintenance logs, with hands-on signal feature engineering, health indices, early-warning thresholds and maintenance cost optimisation.
- Live Course
- 5 modules
-
SKU-location demand forecasting covering promotions, price elasticity, stockout censoring adjustment, intermittent demand (Croston/ADIDA), hierarchical reconciliation and probabilistic forecasts tied to safety stock.
- Live Course
- 5 modules
-
Bridging predictive models and mathematical optimisation: LP/MILP formulation, capacity, scheduling, routing and budget/price allocation with OR-Tools and Pyomo, plus stochastic and contextual decision systems under uncertainty.
- Live Course
- 5 modules
-
Mastering tabular gradient boosting with XGBoost, LightGBM and CatBoost: tree-growth strategies, native categorical and missing-value handling, ranking/quantile/custom objectives, monotonic constraints, Optuna and fast inference.
- Live Course
- 5 modules
-
Automated model selection, hyperparameter search and multi-layer stacking with AutoGluon, FLAML and Optuna, covering compute budget management, leakage checks and moving from rapid baselines to production.
- Live Course
- 5 modules
-
Glass-box models (EBM, GAM), TreeSHAP/KernelSHAP, ALE, counterfactual explanations (DiCE), uncertainty quantification and regulatory compliance for producing auditable, actionable explanations in high-stakes decisions.
- Live Course
- 5 modules
Get in touch
