Subject Overview

An advanced, practice-oriented study of data science integrating statistical reasoning, computing, machine learning, data engineering, decision intelligence, governance and organizational value creation.

Learning Goal

Students will be able to frame data-driven problems, acquire and analyze data, build and evaluate statistical and machine-learning solutions, deploy responsible data products, and lead data science initiatives supporting strategy, innovation and digital transformation.

Syllabus

1. Data Science Foundations and Professional Practice

Learning Objective: Explain data science terminology, roles, applications and relationships among statistics, analytics, AI, machine learning, business intelligence and data engineering; use Python or R, notebooks, Git and reproducible workflows.

Category: foundation

2. Problem Framing and the Data Science Lifecycle

Learning Objective: Translate organizational challenges into analytical questions, hypotheses and measurable outcomes using CRISP-DM, OSEMN, design thinking and scientific research methods; distinguish descriptive, diagnostic, predictive and prescriptive analytics.

Category: foundation

3. Mathematical, Probability and Statistical Foundations

Learning Objective: Apply linear algebra, calculus, optimization, probability distributions, sampling, estimation, Bayesian reasoning and statistical assumptions underlying data science models.

Category: core

4. Data Acquisition, Storage and Management

Learning Objective: Acquire structured, semi-structured and unstructured data through databases, APIs, files, sensors and web sources; apply SQL, data modeling, metadata, warehousing, lakes and lakehouse concepts.

Category: core

5. Data Quality, Preparation and Feature Engineering

Learning Objective: Profile, clean, integrate and transform data; address missingness, outliers, imbalance, leakage and bias; construct reproducible pipelines and informative features for numerical, categorical, text and temporal data.

Category: core

6. Exploratory Data Analysis and Visualization

Learning Objective: Use summary statistics, distributional analysis, multivariate exploration and visual design principles to discover patterns, test assumptions and communicate findings through dashboards, narratives and interactive visualizations.

Category: core

7. Statistical Inference, Experimentation and Causal Analysis

Learning Objective: Conduct confidence estimation, hypothesis testing, power analysis, A/B and multivariate experiments; evaluate observational evidence using causal diagrams, matching, instrumental variables and difference-in-differences.

Category: advanced

8. Supervised Machine Learning

Learning Objective: Build regression and classification models using linear and generalized linear models, trees, ensembles, support vector machines and nearest-neighbor methods; understand regularization, optimization and bias–variance trade-offs.

Category: core

9. Unsupervised Learning and Pattern Discovery

Learning Objective: Apply clustering, dimensionality reduction, association analysis, anomaly detection and representation learning; validate discovered structures and connect patterns to actionable domain insights.

Category: core

10. Time-Series, Forecasting and Prescriptive Analytics

Learning Objective: Model trend, seasonality and dependence using classical and machine-learning forecasting methods; quantify uncertainty and integrate forecasts with simulation, optimization and resource-allocation decisions.

Category: application

11. Deep Learning and Contemporary AI

Learning Objective: Explain neural networks and apply deep learning to text, images and sequences; evaluate embeddings, transformers, foundation models and generative AI for data science while recognizing reliability, cost and risk constraints.

Category: contemporary

12. Model Evaluation, Explainability and Robustness

Learning Objective: Select suitable metrics and validation designs; perform tuning, calibration, error analysis, fairness assessment, stress testing and uncertainty quantification; interpret models using global and local explanation techniques.

Category: advanced

13. Scalable Data Engineering, Cloud and MLOps

Learning Objective: Design batch and streaming pipelines using distributed and cloud platforms; operationalize models through APIs, containers, orchestration, feature stores, CI/CD, experiment tracking, monitoring and drift management.

Category: application

14. Data Products, Decision Intelligence and Transformation

Learning Objective: Convert analytical outputs into decisions and scalable products; define KPIs, estimate business value, manage data science portfolios and teams, communicate with stakeholders, and lead analytics-enabled digital transformation.

Category: strategic

15. Responsible Data Science, Governance and Future Directions

Learning Objective: Apply privacy, security, consent, fairness, transparency, accountability and regulatory principles; establish data and AI governance, auditability and human oversight while evaluating AutoML, synthetic data, federated learning, edge AI and emerging research.

Category: contemporary