BADM 576 - Data Science and Analytics (ML II)
Program-level details: See program/curriculum.md
Status: Draft Initial outline; pending instructor review.
| Credits: 4 | Term: Fall 2027 (Weeks 1-8) | Instructors: Zilong Liu, Amir Fazli (co-instructor, confirmed 2026-07-10) |
Course Vision
Building on supervised learning foundations from FIN 550 (ML I) and agentic systems experience from Agentic AI, students master advanced ML techniques and the full deployment lifecycle. This course covers advanced ensembles, unsupervised learning, NLP/text analytics, time series, neural networks, and model deployment as cloud-hosted, user-facing applications. By course end, students can build, deploy, and document production-ready ML systems — directly preparing them for the Practicum client project.
Learning Outcomes (L-C-E Framework)
Literacy (Foundational Awareness)
- L1: Understand advanced ML paradigms (unsupervised learning, deep learning, NLP) and when each applies
- L2: Explain the ML deployment lifecycle (training, serving, monitoring, retraining)
- L3: Recognize ethical issues in ML deployment (bias, fairness, transparency, model drift)
Competency (Applied Skills)
- C1: Apply advanced ensemble methods, regularization techniques, and hyperparameter tuning to improve predictive model performance
- C2: Implement clustering and dimensionality reduction methods (K-means, DBSCAN, hierarchical clustering, PCA, t-SNE) to discover patterns and generate business insights from data
- C3: Build text analytics workflows using TF-IDF, embeddings, and machine learning techniques for classification, retrieval, and exploratory analysis
- C4: Package, deploy, and evaluate machine learning models using reproducible pipelines, model serialization, and cloud-hosted applications
Expertise (Advanced Application)
- E1: Develop end-to-end machine learning applications that integrate data preparation, modeling, evaluation, deployment, and user-facing inference
- E2: Compare, select, and justify appropriate machine learning approaches for business analytics problems
- E3: Assess machine learning applications for fairness, bias, transparency, limitations, and responsible business use, and communicate findings through professional model documentation
Learning outcomes updated 2026-06-09 per Zilong Liu (zilongl@illinois.edu).
Week-by-Week Breakdown
| Week | Topic | Lectures | Project Work | Studio Session | Assessment |
|---|---|---|---|---|---|
| 1 | Advanced ensembles + regularization | 2 videos | Team formation + problem scoping | Regularization deep-dive - Ridge, Lasso, ElasticNet, ensemble tuning | Weekly assignment 1 |
| 2 | Unsupervised learning: clustering + dimensionality reduction | 3 videos | EDA + initial segmentation | Clustering workshop - K-means, DBSCAN, hierarchical, PCA, t-SNE | Weekly assignment 2 + Milestone M1 |
| 3 | NLP/text analytics | 2 videos | Text analytics for project | NLP with scikit-learn - TF-IDF, word embeddings, text classification | Weekly assignment 3 |
| 4 | Time series analysis | 3 videos | Forecasting component | Time series workshop - ARIMA, Prophet, evaluation metrics | Weekly assignment 4 + Milestone M2 |
| 5 | Neural networks intro | 2 videos | Neural net model for project | Neural networks - architectures, keras/tensorflow basics | Weekly assignment 5 |
| 6 | Deep learning applications | 2 videos | System architecture + API design | Deep learning - CNNs for tabular data, transfer learning | Weekly assignment 6 + Milestone M3 |
| 7 | Model deployment + MLOps + LLMOps | 2 videos | Deploy model + monitoring | ML in production - Docker, APIs, monitoring, agentic AI deployment | Final deliverable work |
| 8 | Ethics, synthesis, portfolio showcase | 1 video | Final deliverable + reflection | Ethics in ML - bias, fairness, model cards + final presentations | Final deliverable + team oral defense |
Team Project: Production ML System (Team of 3)
One major team project runs across all 8 weeks. Teams build a production-ready ML system that incorporates advanced analytics, forecasting, deep learning, and full deployment with MLOps/LLMOps practices.
Problem Options:
- Customer intelligence platform (segmentation + churn prediction + demand forecasting)
- Financial analytics system (market analysis + sentiment + price forecasting)
- Operations intelligence platform (supply chain optimization + anomaly detection + demand forecasting)
- Student choice (approved)
Weekly Assignments (Weeks 1-6, Individual)
Hands-on exercises that build technical skills feeding into the team project:
| Week | Assignment | Focus |
|---|---|---|
| 1 | Ensemble methods lab | Ridge, Lasso, ElasticNet comparison; advanced ensemble tuning |
| 2 | Clustering + dimensionality reduction | K-means, DBSCAN, hierarchical clustering; PCA, t-SNE visualization |
| 3 | Text analytics pipeline | TF-IDF, embeddings, text classification with scikit-learn |
| 4 | Time series forecasting | ARIMA, Prophet models; MAE, MAPE, RMSE evaluation |
| 5 | Neural network fundamentals | Architecture design, keras/tensorflow basics, training documentation |
| 6 | Deep learning application | CNNs for tabular data, transfer learning, comparison with traditional methods |
Rubric per assignment (3 dimensions):
| Dimension | Excellent (A) | Proficient (B) | Developing (C) |
|---|---|---|---|
| Technical Execution | Correct implementation, well-tuned parameters | Functional code, reasonable choices | Incomplete or poorly tuned |
| Interpretation | Clear business insights from results | Adequate explanation | Minimal interpretation |
| Code Quality | Clean, commented, reproducible | Readable code | Disorganized or undocumented |
Project Milestones (Progressive, Team)
Milestones build progressively toward the final deployed system:
| Milestone | Due | Deliverable |
|---|---|---|
| M1: Problem Scoping + Data | End of Week 2 | Problem definition, dataset selection, EDA, initial clustering/segmentation analysis, team charter |
| M2: Model Development | End of Week 4 | Trained models (ensemble + time series + baseline neural net), evaluation metrics, model comparison |
| M3: System Architecture | End of Week 6 | System design document, API specification, deployment plan, LLMOps integration plan |
Rubric per milestone (3 dimensions):
| Dimension | Excellent (A) | Proficient (B) | Developing (C) |
|---|---|---|---|
| Progress | On track, all deliverables complete | Most deliverables complete | Behind schedule or incomplete |
| Technical Depth | Rigorous analysis, justified decisions | Sound approach | Superficial or unjustified |
| Team Collaboration | Clear task division, all members contributing | Adequate collaboration | Uneven contributions |
Final Project Deliverable (Week 7-8, Team)
Deliverables:
- Model Documentation:
- Model card (what it does, performance, limitations, ethical considerations)
- Data sheet (dataset provenance, bias analysis)
- System design document (inputs, outputs, dependencies)
- Deployment:
- Containerized model (Docker)
- REST API (Flask/FastAPI)
- Cloud deployment (AWS Lambda, Heroku, or similar)
- Monitoring + Operations:
- Unit tests + integration tests
- Performance monitoring (track model accuracy over time)
- Data drift detection (alert if input distribution changes)
- Retraining strategy (how often to retrain?)
- LLMOps Component:
- Integration of agentic AI concepts (how LLM-based tools fit in the ML system)
- Prompt management and versioning strategy (if applicable)
- Fairness + Ethics Analysis:
- Evaluate model for bias (across demographic groups if applicable)
- Document limitations + intended use cases
- Identify risks + mitigation strategies
- Peer evaluation of team contributions
- GitHub repo with all code + tests + Docker file + documentation
Rubric (5 dimensions):
| Dimension | Excellent (A) | Proficient (B) | Developing (C) |
|---|---|---|---|
| Deployment | Production-ready, containerized, accessible | Works on cloud | Local only |
| MLOps/LLMOps | Comprehensive monitoring, drift detection, LLM integration | Basic tracking | No monitoring |
| Model Quality | Multiple well-tuned models with rigorous comparison | Functional models | Single or poorly tuned model |
| Fairness Analysis | Thoughtful bias evaluation + mitigation | Addresses fairness | Ignores fairness |
| Documentation | Model card + system design complete | Adequate docs | Minimal documentation |
Oral Defense (Week 8, Team)
Teams present their deployed system, demonstrate the live API, walk through the model card, and answer questions on design decisions, fairness analysis, and deployment trade-offs.
Rubric (3 dimensions):
| Dimension | Excellent (A) | Proficient (B) | Developing (C) |
|---|---|---|---|
| Technical Depth | Clear explanation of architecture, model choices, and trade-offs | Adequate explanation | Superficial or confused |
| Live Demo | Confident demo of deployed system, handles edge cases | System works but limited demo | Demo fails or only screenshots |
| Q&A | Handles questions confidently, demonstrates deep understanding | Answers most questions | Unable to answer or deflects |
AI Tools Integration
Weeks 1-3 (Weekly Assignments + Project M1):
- Use Claude/ChatGPT to:
- Explain regularization trade-offs
- Debug clustering and NLP pipeline issues
- Suggest dimensionality reduction approaches
- Generate feature engineering code
Weeks 4-6 (Weekly Assignments + Project M2-M3):
- Use AI to:
- Explain ARIMA parameter selection
- Debug neural network training issues
- Suggest architecture choices
- Generate evaluation code
Weeks 7-8 (Final Deliverable + Deployment):
- Use AI to:
- Write Docker/API code
- Generate monitoring and drift detection code
- Create model cards and documentation
- Review design for production readiness
Studio Session Topics:
- Week 1: Regularization + advanced ensembles
- Week 2: Clustering + dimensionality reduction visualization
- Week 3: NLP pipelines + text feature engineering
- Week 4: Time series decomposition + ARIMA
- Week 5: Neural network training + debugging
- Week 6: Deep learning applications + transfer learning
- Week 7: Model deployment + containerization + LLMOps
- Week 8: ML ethics + fairness + team presentations
Assessment Summary
| Component | Weight | v3 Bucket | Notes |
|---|---|---|---|
| Weekly assignments | 30% | ICA | Weeks 1-6, individual. Canvas Live Session engagement folded in. |
| Project milestones | 25% | Project — Team Deliverables | M1 (Wk 2), M2 (Wk 4), M3 (Wk 6), team |
| Final project deliverable | 15% | Project — Team Deliverables | Weeks 7-8, team |
| Oral defense | 20% | Project — Individual Deliverables | Week 8, team |
| Peer evaluation (Peerceptiv) | 10% | Engagement | Wk 4 formative 2% + Wk 8 summative 8% |
No traditional exam. One major team project with weekly individual skill-building assignments.
AI Usage Levels (AIAS)
| Assessment | AIAS Level | AI Permitted |
|---|---|---|
| Weekly Assignments | 2 | AI for debugging, parameter guidance, code explanation — with attribution |
| Project Milestones | 2 | AI for EDA, model selection guidance, architecture suggestions — with attribution |
| Final Project Deliverable | 3 | AI as collaborator for Docker/API code, model cards, monitoring scripts — with full disclosure |
| Oral Defense | 0 | No AI |
| Studio sessions (ungraded — output folds into milestones) | 1 | AI for exploration during exercises |
Technology Stack
- ML Libraries: scikit-learn, XGBoost, LightGBM, keras/tensorflow
- Data: pandas, numpy, feature-engine
- Text: scikit-learn (TF-IDF), gensim, spacy (optional)
- Time Series: statsmodels, prophet
- Deep Learning: keras, tensorflow
- Deployment: Docker, Flask/FastAPI, AWS Lambda
- Monitoring: evidently (for ML monitoring), custom scripts
- Testing: pytest, unittest
- IDE: VS Code with GitHub Copilot; Google Colab (browser alternative)
- Notebooks: Jupyter Notebooks (via Colab or VS Code)
- Version Control: GitHub
Prerequisites
- Completion of FIN 550 (supervised ML foundations) + BADM 558 (cloud infrastructure)
- Comfortable with Python programming + SQL
Bridge Module: ML Refresher (Pre-Course, ~3 hours)
Complete before Week 1. Available in Canvas at the start of Fall 2027. There is an approximately 8-month gap between ML I (FIN 550, Fall 2026) and ML II (this course). This module helps students rebuild fluency before diving into advanced topics.
| Unit | Topics | Format | Self-Check |
|---|---|---|---|
| 1. Supervised Learning Review (1 hr) | Train/test splits, cross-validation, overfitting/underfitting, bias-variance tradeoff | Narrated Jupyter notebook walkthrough | Quiz: identify overfitting in a learning curve, explain train/test split |
| 2. Model Evaluation Refresher (1 hr) | Accuracy, precision, recall, F1, ROC/AUC, confusion matrix, regression metrics (MAE, RMSE, R²) | Interactive Jupyter exercises with pre-built models | Quiz: interpret a confusion matrix, choose the right metric for a scenario |
| 3. Core Algorithms Quick Review (1 hr) | Linear/logistic regression, decision trees, random forest, gradient boosting — when to use each | Cheat sheet + short exercises comparing model outputs | Quiz: given a problem description, recommend an algorithm and justify |
Readiness check: Students who pass all 3 self-check quizzes (70% threshold) are ready for Week 1. This module is strongly recommended for all students, not just those who feel rusty.
| Course Sequence: ← BADM 557 — Business Intelligence | Next: Practicum → |