Welcome to the pages of the working group Applied Computer Sciences headed by Prof. Dr. Jörg Frochte. The working group is assigned to the Department of Electrical Engineering and Computer Science and is primarily active at the Velbert/Heiligenhaus campus. The focus of our activities in teaching and research is on the two topics
- Machine Learning and Data Mining
- Modeling and Simulation
In the field of machine learning and data mining, the working group is active in the Bergische Innovationsplattform für Künstliche Intelligenz (Bergische Innovation Platform for Artificial Intelligence) and networked with scientists from the Bergische University of Wuppertal within the interdisciplinary center "Machine Learning and Data Analytics". In addition, we maintain collaborations with colleagues from Weimar, Leipzig, Lemgo, Wellington (New Zealand) ...
Since 2019/20, the group is also particularly active in the continuing education project WeAI of the Bochum University of Applied Sciences, which offers computer scientists, engineers and natural scientists in the field of Machine Learning a practical form of continuing education at the highest professional level.
A research assistant is usually a person who already has a bachelor's degree and is enrolled in a master's program. A student assistant is a person enrolled in a Bachelor's program without a degree.
There are currently no open vacancies for research/ student asssistants in AKIS or the participating research groups.
The working group is so far active in the areas of:
- Simulation Data Mining: application of data mining to data sets from simulations.
- Educational data mining: application of data mining to learning and examination data
- Machine learning in the area of autonomous robots in cooperation with colleague M. Schmidt
- Machine learning and AI in the area of computer games and virtual environments
Furthermore, we are always open for new exciting applications in the field of machine learning.
Currently, we are working in the area of "machine learning" on the following two topics:
- Learning of Ill-Posed Problems
- Learning on Vertically Partitioned Data
For further information, please visit the english website and the publications section.
The research inthis project focuses on staged learning approaches. Using the approaches explored, we aim to make progress in the following problem areas:
- Learning Autonomous Systems with Limited Resources.
- Privacy & data security in cloud learning
- Improved handling of incomplete data sets
The project is part of the Bergische Innovation Platform for Artificial Intelligence. This project is funded by the European Regional Development Fund (ERDF).
DIBS (Data MIning for Student Advising) is an american slang expression and could probably best be translated in German as "First!" as an exclamation. In this sense, the aim is to implement approaches that make it possible to identify risks of dropping out at an early stage. For this purpose, a first study was carried out with special consideration of data protection and data security in learning systems and procedures. The project was funded by the state of North Rhine-Westphalia.
SimCloud is a project supported by the BMBF (Federal Ministry of Education and Research) for the integration of a finite element simulation into a cloud architecture. The goal is to realize an integrated workflow for a cloud-based FEM application in a prototype. Essential aspects are security and user support. For the distribution to hetrogenic structures we made intensive use of machine learning techniques. The project was funded by the BMBF.
- Maschinelles Lernen: Grundlagen und Algorithmen in Python (3. erweiterte und korrigierte Auflage)
published by the Hanser Fachbuchverlag, Nov. 2020; ISBN-13: 978-3446461444 ; 616 Pages
- Maschinelles Lernen: Grundlagen und Algorithmen in Python (2. korrigierte Auflage)
published by the Hanser Fachbuchverlag, Jan. 2019; ISBN-13: 978-3446459960 ; 406 Pages
- Maschinelles Lernen: Grundlagen und Algorithmen in Python (1. Auflage)
published by the Hanser Fachbuchverlag, Aug. 2018; ISBN-10: 3446446656; ISBN-13: 978-3446446656; 406 Pages
- Finite-Elemente-Methode: Eine praxisbezogene Einführung mit GNU Octave/MATLAB (1. Auflage)
published by the Hanser Fachbuchverlag, Oct. 2016; ISBN-10: 3446452915; ISBN-13: 978-3446452916; 320 Pages
The papers below are all peer-reviewed publications:
- Towards Explainability in Modern Educational Data Mining: A Survey
Basile Tousside, Yashwanth Dama and Jörg Frochte published in: Proceedings of the 14th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management. Pages 212-220. DOI, PDF, BibTex
- Group and Exclusive Sparse Regularization-based Continual Learning of CNNs
Basile Tousside, Janis Mohr and Jörg Frochte published in: Proceedings of the 35th Canadian Conference on Artificial Intelligence. DOI, PDF, BibTex
- Investigation of Capsule Networks Regarding their Potential of Explainability and Image Rankings
Felizia Quetscher, Christof Kaufmann and Jörg Frochte published in: Proceedings of the 14th International Conference on Agents and Artificial Intelligence - Volume 3: ICAART 2022, ISBN 978-989-758-547-0, pages 343-351. (DOI link), (BibTeX), (PDF)
- Regression learning on patches
joint work with Stephen Marsland accepted for publishing at the AusDM 2020
A Learning Approach for Optimizing Robot Behavior Selection Algorithm
Basile Tousside, Janis Mohr, Marco Schmidt and Jörg Frochte published in: Chan C.S. et al. (eds) Intelligent Robotics and Applications. ICIRA 2020. Lecture Notes in Computer Science, vol 12595. Springer, Cham. (Link); BibTex ; PrePrint version (pdf)
- Learning Approach for Ill-Posed Optimisation Problems
joint work with Stephen Marsland pubished in : Le T. et al. (eds) Data Mining. AusDM 2019. Communications in Computer and Information Science, vol 1127, pp 16-27, Springer (Link); BibTex ; PrePrint version (pdf) -- used data for example 2 zip, example 4 zip. Details about example 3 are discussed in Goal-shot, a benchmark problem for ill-posed problems see below
- Case Study On Model-based Application of Machine Learning Using Small CAD Databases for Cost Estimation
joint work with Stefan Börzel published in the Proceedings of the 12th International Conference on Knowledge Discovery and Information Retrieval (KDIR 2019), September 2019 in Vienna Austria; BibTex; PrePrint version (pdf), SciTePress
- Concerning the Integration of Machine Learning Contents in Mechatronics Curricula
joint work with Markus Lemmen and Marco Schmidt published as chapter in the book “
Revolutionizing Education in the Age of AI and Machine Learning” (IGI Global) ; BibTex
- Seamless Integration of Machine Learning Contents in Mechatronics Curricula
joint work with Markus Lemmen and Marco Schmidt to be published in the Proceedings of the IEEE
19th International Conference on Research and Education in Mechatronics (REM 2018), June 2018 in Delft, Netherlands (pdf in der IEEE Xplore); BibTex
- Success Prediction System for Student Counseling Using Data Mining joint work with Irina Bernst accepted for publication in the Proceedings of the International Conference on Knowledge Discovery and Information Retrieval, November 2016
- A Case Study on FMU as Co-Simulation Exchange Format for FEM Models
joint work with Christof Kaufmann accepted for publication in the Proceedings of the International Conference on Applied Computing 2016 (Mannheim, Co-Organized by the University of Mannheim), October 2016
Learning Load Balancing for Simulation in Heterogeneous Systems
joint work with Irina Bernst and Christof Kaufmann published in the Proceedings of the 12th International Conference on Applied Computing 2015 (Greater Dublin, Ireland), ISBN - 978-989-8533-45-6 pages 121 - 128, BibTex, PrePrint version (pdf)
Influence of Plant Model Variants for the Automatic Optimisation of Control Parameters
joint work with Patrick Bouillon and Markus Lemmen published in in the Proceedings of the 16th International Conference on Research and Education in Mechatronics (Bochum, Germany), ISBN - 978-3-945728-01-7; pages 80-87 BibTex, PrePrint version (pdf)
Simulation- and Web-Based E-Learning in Engineering - Open Source Architecture and Didactic Issues -
joint work with Patrick Bouillon published in in the Proceedings of the 16th International Conference on Research and Education in Mechatronics (Bochum, Germany), ISBN - 978-3-945728-01-7; pages 127-134 BibTex, PrePrint version (pdf)
An Approach For Secure Cloud Computing for FEM Simulation
joint work with Christof Kaufmann and Patrick Bouillon;
published in the Proceedings of the IADIS International Conference on Applied Computing 2014; ISBN - 978-989-8533-25-8; pages 234-239, PrePrint version (pdf)
An Approach for Load Balancing for Simulation in Heterogeneous Distributed Systems using Simulation Data Mining
joint work with Irina Bernst, Patrick Bouillon and Christof Kaufmann;
published in the Proceedings of the IADIS International Conference on Applied Computing 2014; ISBN - 978-989-8533-25-8; pages 254-259 PrePrint version ( pdf)
Learning Overlap Optimization for Domain Decomposition Methods
joint work with Steven Burrows, Benno Stein, Michael Völske and Ana Belén Martínez Torres;
accepted for publishing at the 17th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2013)
- Simulation Data Mining for Supporting Bridge Design.
joint work with Steven Burrows, Benno Stein, David Wiesner and Katja Müller.
published in Proc. Australasian Data Mining Conference (AusDM 11), Ballarat, Australia; pages 71-79, December 2011. ACM. ISBN 978-1-921770-02-9
BibTex, PrePrint version (pdf), AusDM Online Version
- Evaluation and Adaptation of Techniques for Higher Index DAE with Respect to Real-Time Simulation
ASIM 2011 Proceedings, September 2011, Winterthur (Switzerland)
ISBN/ISSN: 978-3-899677331 (print-version), 978-3-905745443(CD)
BibTex, PrePrint version (pdf)
- Modelica Simulator Compatibility - Today and in Future
Proceedings 8th International Modelica Conference, Dresden, March 2011
BibTex, PrePrint version ( pdf), Modelica Proceedings
- A numerical Method for a Nonlinear Spatial Population Model with a Continuous Delay
International Conference of Numerical Analysis and Applied Mathematics 2010
AIP Conference Proceedings 1281; ISBN 978-0-7354-0834-0
BibTex, PrePrint version (pdf), AIP Online Version
- A Splitting Technique of Higher Order for the Navier-Stokes Equations
joint work with Wilhelm Heinrichs
published in Journal of Computational and Applied Mathematics (2009)
BibTex, PrePrint version (pdf), Science Direct Online Version
- An Adaptive Higher Order Method in Time for Partial Integro-Differential Equations
International Conference on Numerical Analysis and Applied Mathematics 2008
AIP Conference Proceedings 1048; ISBN 978-0-7354-0576-9
BibTex, PrePrint version (pdf), AIP Online Version
- A third order method for Convection-Diffusion Equations with a Delay term
Numerical mathematics and advanced applications.
Proceedings of ENUMATH 2007
Springer (2008). ISBN-10: 3540697764
BibTex, PrePrint version (pdf), Springer Online Version
Members of the working group supervise the following courses at the Velbert/Heiligenhaus campus. The names correspond to the new exmination regulations, which are valid since the WS 2015/16. An international website on the lectures can be found here.
Password and login for the documents will be announced in the course.
- Analysis 2 (documents, aids in the exam etc.)
- Machine Learning & Data Mining (documents, aids in the exam etc.)
- In-depth Simulation (documents etc.) Examination form is usually a paper, not a written examination
"Machine Learning & Data Mining" can also be taken in the Computer Science (Bochum Campus) elective module at CVH.
The following courses are offered as part of the Master's program in Mechatronics and Information Technology at the Velbert/Heiligenhaus campus
- Numerical Mathematics and Simulation (documents, aids in the exam etc.),
- Applied Artificial Intelligence (documents etc.)
- 2022 – M. Born: Bachelor thesis "Analyse der Eignung von piezoelektrischen Keramikkondensatoren in frequenzvariablen Resonanzinvertern zur Ansteuerung von BLDC-Motoren im elektrifizierten Antrieb"
- 2021 – M. Heimbach: Master thesis "Performance-Vergleich verschiedener Reinforcement Learning Verfahren auf den beschränkten Rechenressourcen eines Kriechroboters"
- 2021 - F. Quetscher: Master thesis "Investigation of Capsule Networks regarding the Explainability of Search Engine Rankings"
- 2021 - S. Meier: Master thesis "Analyse und Vergleich von Verfahren in Predictive Maintenance"
- 2020 - L. Friedrichsen: Master thesis "Impact of modularization on learning behavior and detection rate of Convolutional Neural Networks"
- 2020 - F. Breidenbach: Bachelor thesis "Objekterkennung durch maschinelles Lernen anhand von Oberflächentexturen -Ansatz auf der Basis von Convolutional Neural Network"
- 2019 - H. Richter: Master thesis "Einsatzmöglichkeiten von Convolutional Neural Networks zur Klassifizierung von Vogelstimmen"
- 2019 - L. Jakob: Master thesis "Transfer Learning für die Handgestenerkennung mit Bilddaten geringer Auflösung"
- 2018 - H. Richter: Bachelor thesis "Informationsextraktion aus Ausweisdokumenten mittels Deep Neural Networks"
- 2017 - P. Bouillon: Master thesis "Simulation-Based Pretraining for a Multilevel Reinforcement Learning on Mobile Robots"
- 2017 - Tobias S. Fischer: Bachelor thesis "Nonintrusive Load Monitoring – Erkennung von Finite State Machines mittels Sequential Pattern Mining"
- 2016 - J. Beran: Master thesis "Vergleich verschiedener Ansätze für lernende Agenten in strategischen Planspielen"
- 2015 - T. E. Preuß: Bachelor thesis "Training eines Saugroboters in einer virtuellen Umgebung"
- 2015 - O. P. Müller: Master thesis "Klassifikation von Handgesten in drei Dimensionen mittels maschinellen Lernens"
- 2015 - J. Dambacher: external bachelor thesis
- 2014 - P. Bouillon: Bachelor thesis "Training eines humanoiden Roboters durch maschinelles Lernen"
- 2014 - D. Cziesla: Bachelor thesis "3D Visualisierung technischer Daten in Webapplikationen"
If you are a student of the Bochum University of Applied Sciences and are thinking about writing a bachelor's or master's thesis with me, you can inform yourself about the focus of the working group on the pages Projects and Theses, Research and Development, and Publications. In general, in my experience, the following applies: If you come up with your own idea or if we find a topic together that you are really interested in, the results will generally be good. So if your topic does not appear in the topics on these pages, don't let this scare you off at first, just present your idea.
Concerning the form, the same hints apply as below for the term papers. If you are unsure, it is best to address the open questions directly at the beginning of the paper.
The working group regularly offers the option of carrying out Communication and Information Systems, Software and Development projects in our focus areas. Suggestions for topics are regularly posted in the showcases. Of course you are also welcome to come up with your own ideas.
Students of computer science from Bochum with an interest in machine learning or applied AI are of course also welcome for their software projects, etc.
Please consider this example only as an orientation guide. It should give you an impression of how a paper is typically structured and which classic mistakes (see e.g. note on handling sources and citations at the bottom of the page) should be avoided. Also the basics do not always have to be called basics, but can be modified as a heading to suit the paper.
Of course you do not have to use LaTeX. If you want to use Word or LibreOffice you can simply follow the pdf above.If you want to use LaTeX under Windows I recommend MiKTeX.
Please note that form and citation technique are important things in such papers and will also be reflected in the grade. Take the hints to heart.