Conducted in
terms:
2024Z, 2025Z
Erasmus code: 14.0
ISCED code: 0310
ECTS credits:
5
Language:
English
Organized by:
Faculty of Political Science and International Studies
Module- Coding [Python ] 2100-SPP-L-D3MOCO
Module 1: Coding Environment & Prompting Foundations
Coding Preparation: Introduction to setting up a professional coding environment, focusing on advanced code editors (such as VS Code) and the use of extensions and plugins for enhanced productivity.
AI Coding Assistants: Overview of AI-powered coding aids like GitHub Copilot, Cursor, ChatGPT, Claude, and similar models, including how to effectively integrate them into your workflow for code generation, debugging, and learning.
Prompt Engineering Introduction: Fundamental concepts of interacting with AI models, with practical guidance on crafting effective prompts. Introduction to zero-shot and few-shot prompting strategies and their significance in obtaining quality outputs from language models.
Module 2: Regular Expressions & NLP Basics
Regular Expressions: Comprehensive coverage of regex syntax, including character classes, quantifiers, anchors, and grouping for pattern-based text matching.
NLP Preprocessing: Introduction to core text processing techniques such as tokenization, stemming, lemmatization, and stop word removal to prepare textual data for analysis.
Pattern Matching & Extraction: Hands-on practice with searching, extracting, and cleaning text using regular expressions and NLP preprocessing tools, with examples relevant to data science and automation tasks.
Module 3: Data Scraping Essentials
Web Data Extraction: Learn the principles and best practices of scraping structured and unstructured data from websites. This includes navigating page structures and automating data retrieval.
APIs and Data Collection: Explore how to interact with web APIs for accessing data programmatically and efficiently.
Practical Applications: Gain practical skills in data acquisition, understand the legal and ethical boundaries, and discuss responsible data use.
Module 4: Introduction to Large Language Model (LLM) Architectures
LLM Fundamentals: Overview of large language models, their underlying architectures (such as transformer models), and how they are trained and fine-tuned.
Running LLMs Locally and via API: Step-by-step guidance on deploying and utilizing LLMs both on local machines and through cloud/API services, including considerations for model selection, hardware requirements, and practical deployment.
Module 5: Security, Privacy & Risks in AI Models
AI Model Security: In-depth exploration of security challenges specific to AI/ML models, including vulnerabilities, attack vectors, and best practices for mitigating risks.
Data Privacy: Discussion of data privacy concerns, emphasizing what happens when sensitive or personally identifiable information is processed by AI models.
OWASP Top 10 for Machine Learning: Introduction to the OWASP Top 10 Machine Learning Security risks, with real-world examples of threats and guidance on risk management and regulatory compliance.
Module 6: Language Models in Document Analysis
Sentiment Analysis: Techniques for detecting and interpreting emotional tone within documents using pre-trained models and custom classifiers.
Text Classification: Approaches for automatically organizing and categorizing documents (e.g., spam filtering, topic categorization).
Named Entity Recognition (NER): Methods for extracting structured information, such as names, locations, and organizations, from unstructured text.
Module 7: Retrieval-Augmented Generation (RAG) Systems
RAG Overview: Introduction to Retrieval-Augmented Generation, how it combines information retrieval and generative AI for enhanced text generation.
Use Cases: Explore practical examples of RAG in action, such as question answering, document search, and enterprise knowledge management.
Workflow & Implementation: Learn about the underlying workflow, components (retriever, generator), and how to implement and evaluate a basic RAG pipeline.
Module 8: Optical Character Recognition (OCR)
OCR Basics: Introduction to OCR technology, focusing on using Tesseract for extracting text from images and scanned documents.
Image Preprocessing: Techniques for improving OCR accuracy through image enhancement and preprocessing steps.
Hands-on Applications: Real-world projects and exercises involving OCR in document processing, digitization, and automation tasks.
AI usage -- Level 5: Full AI
AI should be used as a ‘co-pilot’ in order to meet the requirements of the assessment, allowing for a collaborative approach with AI and enhancing creativity.
You may use Al throughout your assessment to support your own work and do not have to specify which content is Al generated.
Course coordinators
Term 2025Z: | Term 2024Z: |
Learning outcomes
Knowledge: The student knows and understands:
In an advanced degree, methods of collecting, analyzing, and interpreting quantitative and qualitative data used in the process of creating and analyzing political processes [K_W03]
Skills: The student is able to:
Design and conduct complex social research, particularly in the nature of social diagnosis, select appropriate specialized tools, including analytical tools for research questions and collected data, and justify the choices made [K_U01]
Design, in team collaboration, a complex study assessing the relevance, effectiveness, and efficiency of a social program, and collect and utilize data for this purpose, including the use of modern IT tools [K_U04]
Prepare and deliver a written and oral presentation on a selected social problem, including a complex and atypical problem, and propose and justify solutions to it [K_U06]
Assessment criteria
Course Assessment and Grading:
Homework Assignments: (30 points)
Group Project: (Code evaluation: 15 points, Presentation: 15 points)
The presentation will be evaluated by invited guests from both the fields of political science and machine learning.
Passing Grade: To pass the course, students must earn at least 30 points from the group project and homework assignments combined. Successful completion of the group project is mandatory for passing the course.
Bibliography
Lutz, M. (2013). Learning Python (5th ed.). O'Reilly Media, Inc.
Lutz, M. (2011). Programming Python: Powerful Object-Oriented Programming. O’Reilly Media, Inc.
Pilgrim, M. (2009). Dive Into Python 3. Apress.
Ascher, D. and Martelli, A. (2002). Python Cookbook. O’Reilly.
Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. http://www.deeplearningbook.org
Jurafsky, D. and Martin, J.H. (2009). Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics, and Speech Recognition. Pearson Prentice Hall.
Additional information
Additional information (registration calendar, class conductors, localization and schedules of classes), might be available in the USOSweb system: