Ga naar de startpagina
Zoek leermateriaalKennis & expertiseCommunities
NL|EN
Privacy policyDeelnemende instellingen
Creative CommonsCreative Commons AttributionSURFFunded by the European Union

Filteren

Zoekresultaten (17)

HBO/WO

MBO

21 juli 2026

Introduction to Stochastic Hydrology

Hydrologic systems are both complex and extremely heterogeneous in time and space. As hydrological models are simplified versions of reality, their predictions are by definition subject to uncertainty. The field of stochastic hydrology focuses on quantifying this uncertainty along with prediction. This book aims to present an overview of the field of stochastic hydrology at an introductory level. A wide range of topics and methods will be treated, while each topic and method is only treated at a basic level: descriptive statistics; probability and random variables; hydrological statistics and extremes; random functions; time series analysis; geostatistics; forward stochastic modeling; and state prediction and Kalman filtering. The book can be used for a 10-15-week course at third-year undergraduate or first-year graduate level.

Document

Marc F.P. Bierkens

8 april 2026

ACA-bot

ACA-bot is an engaging chatbot designed for Automated Conversational Assessment (ACA). Drawing inspiration from traditional oral exams, this text-based conversational agent transforms assessment into a dynamic learning experience. It supports both formative and summative evaluation by offering students personalised feedback and guidance at every step of their learning journey. The first ACA-bot prototype focuses on assessing students’ understanding of basic statistical concepts, but its flexible design allows for easy adaptation to a wide range of subjects and learning contexts. Through interactive dialogue, students can practise explaining ideas clearly under exam-like conditions, apply their knowledge to new and challenging questions, and refine their reasoning as they respond to follow-up prompts. They gain valuable insights into their own understanding, identify areas for improvement, and gradually build confidence for both oral and written assessments. For educators, ACA-bot offers a powerful way to scale oral examinations and enhance formative assessment without increasing workload. It automates parts of the assessment process, provides structured and actionable insights into student performance, and opens the door to more personal, interactive feedback at scale. By combining the strengths of conversational AI and educational assessment, ACA-bot turns evaluation into a collaborative, insightful process. It empowers students to think aloud, reason more effectively, and grow, while giving educators a smarter, more innovative way to understand and support learning. ACA-bot is een slimme en inspirerende chatbot die onderwijs en Artificial Intelligence samenbrengt. Geïnspireerd door het klassieke mondelinge examen tilt deze tekstgebaseerde gespreksagent Automated Conversational Assessment naar een nieuw niveau. Studenten krijgen via natuurlijke gesprekken directe, persoonlijke feedback en begeleiding, waardoor leren en toetsen samenvloeien in één boeiende ervaring. Het prototype van ACA-bot is ontwikkeld om kennis over statistiek te beoordelen, maar dankzij het flexibele ontwerp kan het systeem eenvoudig worden aangepast aan andere vakgebieden en toepassingen. Tijdens interactieve dialogen leren studenten helder uitleg geven, hun kennis toepassen op nieuwe vragen en hun redenering aanscherpen. Ze ontdekken waar hun begrip nog verdiept kan worden en groeien zichtbaar in zelfvertrouwen — waardevolle voorbereiding op zowel mondelinge als schriftelijke examens. Ook voor docenten biedt ACA-bot een slimme oplossing. Het maakt mondelinge toetsing schaalbaar, ondersteunt formatieve beoordeling via gesprek, en vermindert de werkdruk door delen van het beoordelingsproces te automatiseren. Bovendien geeft het gestructureerde inzichten in de voortgang en denkwijze van studenten, waardoor elke dialoog bijdraagt aan beter onderwijs. ACA-bot laat zien hoe Automated Conversational Assessment het leren persoonlijker, interactiever en effectiever maakt. Studenten denken hardop, verkennen hun eigen begrip, en verbeteren met vertrouwen — terwijl docenten op een innovatieve, datagedreven manier inzicht krijgen in hoe kennis werkelijk wordt begrepen.

App

Aki Härmä, Julie Bauer

6 juni 2025

How To Write A Quantitative Research Paper

So, you need to write a ‘quantitative’ research paper, and you want to know: what is the correct way of writing such a paper? The answer is, unfortunately: it depends … For one, there is no universally accepted, ‘correct’ way of writing any paper in the social sciences in general. Moreover, even just within the so-called ‘quantitative’ research tradition specifically, there exist different viewpoints on what makes for a good research paper. To muddy things up even further, how to write a paper will also depend on the specific research question(s) that it addresses. Hence, even the same scientist might opt for varying approaches in different papers. Having said that, it is possible to identify some common rules and practices that are often followed, as well as a number of practices that should generally be avoided. Based on these common rules and practices, this document is meant to help you on your way with writing a ‘quantitative’ research paper by offering a number of tips, tricks and rules-of-thumb. None of these rules and practices are set in stone, and you (or your teacher/supervisor) may have excellent arguments to deviate from them. However, in general, following them will greatly increase the quality of your paper.

Document

Alex Lehr, Roderick Sluiter

1 november 2024

Computational Modelling of Terrains

An open textbook about the algorithms and methodologies to reconstruct, manipulate, and extract information from terrains (or DTMs or DEMs). It covers different representations of terrains (eg TINs, rasters, point clouds, contour lines) and presents techniques to handle large datasets. The book is currently used for the course Digital Terrain Modelling (GEO1015) in the MSc Geomatics at the Delft University of Technology in the Netherlands. The course is tailored for MSc students who have already followed an introductory course in GIS and in programming. Each chapter is a lesson in the course, and each lesson is accompanied by a video introducing the key ideas and/or explaining some parts of the lessons.

Open (Access) Tekst Boek

Hugo Ledoux

23 augustus 2023

Lecture: Historical Introduction to Scientific Explanation Models

Understanding of the basic inference and explanation modes in the modern world

Website

HCAIM Consortium

23 augustus 2023

Lecture: Risk & Risk mitigation

Learner understands the competing concepts of risk Learner understands how risk is perceived by a the modern society Learner understands approaches to risk evaluation Learner understand approaches to risk mitigation

Website

HCAIM Consortium

23 augustus 2023

Lecture: Model Compression - Edge Computing

Understand the concept of model compression Provide the rationale behind the techniques of pruning, quantization and knowledge distillation Prepare for understanding of basic implementations using a high-level framework like TensorFlow

Website

HCAIM Consortium

23 augustus 2023

Lecture: Unsupervised Learning

Learn basics of unsupervised learning

Website

HCAIM Consortium

23 augustus 2023

Lecture: Decision Networks

Naive Bayesian networks Bayesian networks Decision networks Students can define a multivariate joint distribution: multinomial and Gaussian Students can explain the difference between association versus causation Students can define observational, causal, and counterfactual inference Students can define fairness using observational and counterfactual reasoning

Website

HCAIM Consortium

2021

Tracking student skills real-time through a continuous-variable dynamic Bayesian network

The field of Knowledge Tracing is focused on predicting the success rate of a student for a given skill. Modern methods like Deep Knowledge Tracing provide accurate estimates given enough data, but beingbased on neural networks they struggle to explain how these estimates are formed. More classical methodslike dynamic Bayesian networks can do this, but they cannot give data on the accuracy of their estimatesand often struggle to incorporate new observations in real-time due to their high computational load.This paper presents a novel method, Performance Distribution Tracing (PDT), in which the distribution of the success rate is traced. It uses a dynamic Bayesian network with continuous random variablesas nodes. By tracing the success rate distribution, there is always data available on the accuracy of any success rate estimation. In addition, it makes it possible to combine data from similar/related skills tocome up with a more informed estimate of success rates. This makes it possible to accurately predictexercise success rates, even when an exercise requires a combination of different skills to solve. Andthrough the use of the right conjugate priors, all distributions are available in analytical form, allowing efficient online updates upon new observations. An application to an engineering physics course has shown that the PDT algorithm works well inreal-life settings. Students mainly appreciate the overview given of their learning process as well as the corresponding practice recommendations.

Document

Hildo Bijl