Text Mining 4 Curriculum Design

Imagine you are to design a new course on an emerging topic of your field. How would you do this? How would you make sure its meeting both the requirements of the job market as well as international academic standards? How can you design it in a way it will be positively evaluated by accreditation bodies and other stakeholders?

We develop innovative software solutions that support the curriculum design process. Our tool guides you from the first ideas to the first lecture. Our algorithms crawl the web, retrieve relevant data and analyze it applying next generation text mining techniques. You type in the topic area to develop a course or curriculum for, and the solution will suggest topics and skills required to be considered in your curriculum or syllabus.


Today’s Curriculum Design Process shows several weaknesses. First of all, there is little objectivity. The design process often relies on the opinion of one (or a few) professors. Secondly, the needs from the job market are often not understood​. And thirdly, benchmarking – if done at all – takes place on a small scale by checking 2-3 other curricula and then copy them.


The main objectives of this project Text Mining for Curriculum Design​ is to objectify and automate Curriculum Design by using novel data and text mining techniques​. It will be done by deriving a novel semi-automatic, data driven curriculum design process, which is supported by software. In order to assess the proposed process two reference curricula, e.g. one in Data Science and BPM will be designed.


Curriculum Design

Curriculum design concerns all universities around the world. Traditionally, it is performed manually by academics with years of experience in the design process. Decisions about what content to include in a curriculum and what competences to teach are often made based on highly subjective impressions of individuals. These decisions are very rarely backed with solid quantitative measures. Furthermore, since curricula are designed by academics there is a high risk of bias towards content that might be primarily of academic interest.

Graduate needs and roles, however, are so diverse, so multi-disciplinary, the skills needed are multi-faceted and subtle; there is little clear guidance and there are now so many millions of data points in job ads, skill descriptions, tools needed, etc that it is nearly impossible for a human even if unbiased to filter and digest these to develop a curricula.

Also, whereas curricula in many social science disciplines change only relatively slowly over time, curricula from information technology focused disciplines such as information systems require frequent revisions and updates.

Further, students are becoming increasingly analytic when it comes to evaluating and choosing their education. They need evidence the program is strong and well developed and better than alternatives. The tools we develop help them to see that.


Our approach: Data Science

Data science is said to come with fundamental and disruptive changes not just for businesses but also in the way our society works. Europe is well-aware of this. It has reacted by several initiatives (for more information see e.g. "Making Big Data work for Europe"). In the last few years many companies created data science labs and job advertisements show a growing demand for data scientists. More and more universities are trying to fulfill the market demand for graduates in this area by offering specialized graduate programs in data science. At its core data science is supposed to turn data into business value by using sophisticated analysis methods stemming from mathematics and computer science. Often they originate from sub areas such as statistics, machine learning and visualization. We will make use of Data Science to support you in designing your curriculum!


Data Science for Curriculum Design

There is an abundance of information available, which could inform curriculum design, but it can only be handled through semi-automatic means due to the immense volume of information. Typically, the data format is unstructured or semi-structured text found on webpages, ebooks and leaflets. Dealing with unstructured information in large quantities is challenging but also offers opportunities. We believe that relying on a data driven curriculum design allows to better reflect the needs of our economy and society. For example, by considering regional job advertisements, one might identify different demands that lead to different curricula and specializations, comparing your curriculum with the ones of other higher research and education institutions might point you towards your perfect niche!

The data we are able to retrieve is aggregated and processed iteratively by sophisticated algorithms followed by human analysis. We propose a semi-automatic, iterative process. Despite partial automation the required manual work is considerable, in particular for developing the text mining based methodology. However, after an initial investment in automating the design process, future updates to curricula come with reduced effort compared to purely manual design while incorporating a larger amount of information. The methodology that we develop will substantively facilitate curriculum design. It will foster efficiency and improve quality at the same time, making an important contribution to society, which is increasingly fueled by knowledge and subject to increasing speed of change.


The project brings together experts from different academic fields, located at leading European Universities, globally active and recognized in their particular field.


Project Directors

Prof. Dr. Jan vom Brocke
University of Liechtenstein, Liechtenstein
Prof. Dr. Dr. h.c. Dr. h.c. Jörg Becker
University of Münster, Germany
Prof. Dr. Kieran Conboy
National University of Ireland, Galway, Ireland

Project Lead

Dr. Johannes Schneider
Project Lead, University of Liechtenstein
Dr. Denis Dennehy
Lead, NUI Galway
Dr. Armin Stein
Lead, University of Münster

Project Team

Benedikt Hoffmeister
University of Münster
Marcus Basalla
University of Liechtenstein
Joshua Handali
University of Liechtenstein
Michael Gau
University of Liechtenstein

International Partners

Prof. Dr. Heikki Topi
Bentley, USA
Prof. Dr. Bernard Tan
NUS, Singapore
Prof. Dr. Eija Helena Karsten
Abo Akademi Handelshögskolan, Finland


Project partners are researchers active in both curriculum design and data analytics, and they have published in leading international journals:

Big Data Analytics

  • Conboy, K., Dennehy, D. and O'Connor, M., 2018. ‘Big time’: An examination of temporal complexity and business value in analytics. Information & Management.
  • Schneider, J., & Vlachos, M. (2018, May). Topic Modeling based on Keywords and Context. In Proceedings of the 2018 SIAM International Conference on Data Mining (pp. 369-377). Society for Industrial and Applied Mathematics.
  • Damevski, K., Shepherd, D. C., Schneider, J., & Pollock, L. (2017). Mining sequences of developer interactions in visual studio for usage smells. IEEE Transactions on Software Engineering, 43(4), 359-371.
  • Schneider, J., & Vlachos, M. (2017). Scalable density-based clustering with quality guarantees using random projections. Data Mining and Knowledge Discovery, 31(4), 972-1005.
  • Schmiedel, T., Müller, O., vom Brocke, J.(2018), Topic Modeling as a Strategy of Inquiry in Organizational Research: A Tutorial with an Application Example on Organizational Culture, in: Organizational Research Methods (ORM), (ABCD: A*, ABS: 4*; ISI: 4.78, VHB: A).
  • Lehrer, C., Wieneke, A., vom Brocke, J., Jung, R., Seidel, S. (2018), How Big Data Analytics Enables Service Innovation: Materiality, Affordance, and the Individualization of Service, in: Journal of Management Information Systems (JMIS), forthcoming (ABDC: A*; ABS: 4; FT 50; ISI: 3.775; VHB: A).
  • Müller, O., Junglas, I., Debortoli, S., vom Brocke, J. (2016),Using Text Analytics to Derive Customer Service Management Benefits from Unstructured Data, in: Management Information Systems Quarterly Executive (MISQe), 15(4), 243-258 (ABDC: A; ABS: 2; ISI: 4.961; VHB: B). ICIS Senior Scholars` Best Publication of 2016 Award Winner,
  • Müller, O., Junglas, I., vom Brocke, J., Debortoli, S. (2016), Utilizing Big Data Analytics for Information Systems Research: Challenges, Promises and Guidelines. European Journal of Information Systems (EJIS), Volume 25, Issue 4, pp 289–302 (ABDC: A*; ABS: 3; ISI: 3.505; VHB: A).

Curriculum Design

  • vom Brocke, J., Tan, B., Toppi, H., Weinmann, M. (Eds.). (2017). AIS Global IS Education Report, The Global Report of the Association for Information Systems on Information Systems Education 2017: Association for Information Systems.
  • vom Brocke, J., Tan, B., Toppi, H., Weinmann, M. (Eds.). (2016). AIS Global IS Education Report, The Global Report of the Association for Information Systems on Information Systems Education 2016: Association for Information Systems.
  • vom Brocke, J., Rosemann, M. (2015), „Business Process Management“, In: Association for Information Systems. Reference Syllabi, Ed.: J. vom Brocke, Eduglopedia.org, 2015. Available at: http://eduglopedia.org/reference-syllabus/AIS_Reference_Syllabus_Business_Process_Management.pdf


The projects connects to a number of innovative tools, which are developed in related research and entrepreneurial projects.



Contact us if you want to know more. We are happy to provide more updated information on the project as well as to get you involved to contribute to this stream of research.

Prof. Dr. Jan vom Brocke, Project Director, jan.vom.brocke@uni.li
Dr. Johannes Schneider, Project Manager, johannes.schneider@uni.li

University of Liechtenstein
Institute of Information Systems
9490 Vaduz