Single-Course Engelsk 5 ECTS

Advanced Business Analytics

Overall Course Objectives

Business Analytics (BA) concerns the use of data and models to understand organisational systems and support decisions. This course covers advanced data-centric methods across the three main directions in BA: descriptive (“what happened?”), predictive (“what will happen?”), and prescriptive (“what should happen?”). Emphasis is on connecting modelling choices to decision contexts: problem formulation, data management, model evaluation and validation, and communicating results under uncertainty. Methods are applied to business cases to illustrate how data and machine learning can be used for decision support, including modern approaches where relevant. Throughout, the course highlights how responsible analytics can contribute to societal good—by improving decisions in areas such as sustainability, public services, and equitable outcomes—while recognising limitations, risks, and trade-offs. The course relies on advanced machine learning techniques; students are therefore expected to have prior knowledge of data science and machine learning fundamentals, as well as solid programming skills.

See course description in Danish

Identify business and societal impact opportunities related to effective data utilization
Summarize the identifying characteristics of advanced machine learning approaches for descriptive, predictive and prescriptive analytics
Select and apply appropriate machine learning (regression, classification, reinforcement learning, clustering, recommender systems, computer vision, LLMs)
Conduct machine learning experiments in complex settings (e.g. noisy data, high uncertainty, combination with optimization components, parametric and non-parametric models)
Apply one or more explainable AI techniques (e.g SHAP, Lime) in data-driven decision advisory situations
Understand the technical principles and potential of foundation models (e.g. LLMs)
Build on the latest LLM techniques (RAG, LoRA, Chain-of-Thought reasoning) to develop application-ready tools for business and society
Quantify uncertainties in predictive modelling, through quantile regression and heteroskedasticity modelling
Understand and be able to explain causal vs correlational relationships between variables, endogeneity, intervention and counterfactual reasoning
Be able to provide a clear and informative summary (executive summary) for data-driven analyses and tools, including insights for business and critical questions

The course is delivered in two main blocks. The first half of the semester focuses on advanced business analytics techniques and is taught in an intensive format. This block combines multiple teaching approaches, including interactive lectures integrated with hands‑on Python exercises as well as flipped‑classroom sessions supported by pre‑recorded video lectures and practical in‑class activities. The second half of the semester is dedicated to project work and includes a series of invited talks from industry, the public sector and academia.

The main topics covered in the course includes web data mining; Augmented/Virtual reality; natural language processing; Large Language Models; recommender systems; explainable AI; deep learning; reinforcement learning; spatio-temporal prediction models; ensemble models; survival analysis; prediction uncertainty and computer vision.

Lectures, practical laboratories and online learning, i.e. self-learning with online resources (e.g. with iPython Notebook).