Advanced Methods | Data Science and Applications - Advanced Methods | PAIR

Knowledge

This is a follow-on course to “Advanced Methods | Introduction to Python Programming.” Basic Python literacy (variables, control flow, functions, pandas, simple plotting) is expected. No prior machine learning background is required; all modeling concepts are introduced from first principles.

Description

Building on foundational Python skills, this course covers the end-to-end data science workflow from problem framing to evidence-based communication. Students practice data acquisition, cleaning, wrangling, exploratory data analysis, and core modeling for inference and prediction. Emphasis is placed on transparent, reproducible analysis and ethical practice, including bias, fairness, and data protection. Case studies from varied domains illustrate how to translate domain questions into actionable insights and use real-world datasets to strengthen practical skills in analysis, modeling, and storytelling with data.

Learning objectives

Students who successfully complete this course will be able to:

• Frame domain questions as data problems and select suitable analytical methods.
• Collect, clean, and transform tabular and text data using Python libraries.
• Conduct exploratory data analysis and generate clear, reproducible visualizations.
• Apply and interpret core statistical and machine learning models (e.g., regression, classification, evaluation, validation).
• Communicate findings responsibly, addressing uncertainty, limitations, and ethical considerations.

E-learning

Course materials will be available on ILIAS. Practical work will be conducted in Jupyter/Colab; starter notebooks and templates will be provided.

Preparation

No special preparation beyond the Python prerequisite. Optional refreshers on pandas and plotting will be shared in Week 1.

Detailed information about the examinations

The course assessment includes short in-class analytical exercises and a final mini-project demonstrating an end-to-end workflow (data acquisition/cleaning, EDA, a simple model with validation, and visualization) with a brief presentation. Students must complete at least 50% of in-class exercises and submit the final project to receive credit. (Students can work in groups).

Second additional field

Language of the course: English

Building Blocks of the Course:

• Problem Framing and Data Ethics (bias, consent, transparency).
• Data Acquisition and Wrangling (CSV/JSON, APIs; basic scraping where appropriate).
• Exploratory Data Analysis and Visualization (distributions, relationships, dashboards).
• Modeling Fundamentals (train/validation/test, regression and classification, evaluation metrics).
• Communicating Results (uncertainty, limitations, narratives for non-technical audiences).