1. Information for participants

Thank you very much for your interest in our study on the topic “Retrieval-Augmented Generation (RAG) chatbot for the Introduction to Data Mining lecture”. Please consider the following information carefully.

Purpose of the study

This study evaluates which questions user aks a RAG chatbot and how good the corresponding answers are to support student learning.

Responsible people for this study

The study “Retrieval-Augmented Generation (RAG) chatbot for the Introduction to Data Mining lecture” is performed by the research group Knowledge Representation and Machine Learning at the Faculty of Technology, Bielefeld University. The head of the research group is Prof. Benjamin Paaßen. The participating researcher is Lukas Gehring. During the study, only Lukas Gehring (who does not teach the course) has access to the server.

Study procedure

During the study, you are asked to

• 1. Use the RAG chatbot to ask questions about the lecture to support your learning
• 2. Rate the recieved answers from the chatbot on a scale of (1-5).

During the study, we record your chat requests, the chatbot answers, and how you rated the answer. Irrespective of the study, we record IP addresses to protect the functionality of the website. These are not coupled with any study-related data.

We do not record data that could directly identify you and ask you to avoid giving any hints about your identity (e.g. via your user name or by including your name in the chat window).

Benefits and risks

By participating in the study, you contribute to research on RAG chatbots in education, and more specific in the RAG chatbot for the Introduction to Data Mining coruse. Results of this study will, as such, benefit future students who use the system and scientific progress overall. Furthermore, the chatbot can help you in your learning for the course.

Be aware, that the chatbot can make mistakes. We highly suggest to check the answers with the proviede references.

Voluntary nature of the study

The participation in this study is entirely voluntary. Teachers of the “Introduction to Data Mining” course have no access to the server data and have no knowledge who uses the chatbot or participates in the study. You cannot use the chatbot without participating in the study. If you do not want to participate, just close the window.

Privacy and usage of data

The data recorded during this study are treated confidentially. Details about data usage are explained in the attached privacy statement.

Study results will be published as scientific articles and presentations.

After the study has concluded, the data will be published in a suitable research data repository. The purpose, type, and extent of usage of this data are not foreseeable at this time. Data will only be published in a form that does not permit any inference regarding individual participants. With this procedure the study is following recommendations of the German Research Foundation for quality control in research.

Contact data

If you have any further questions regarding participation, voluntary nature of the study, and data usage, please contact

2. Consent to participate and to data processing

By clicking „I consent to participate in the study“, I confirm that:

I have read and understood the participant information as well as the privacy statement for the study “Retrieval-Augmented Generation (RAG) chatbot for the Introduction to Data Mining lecture”. I have been sufficiently informed and had the opportunity to ask further questions via e-mail or phone.

I am aware that my participation in this study is voluntary and that I do not face any repercussions by declining to participate. I am aware that I can use the tutoring system without participating in the study and that the course teachers have no knowledge who uses the system or who participates in the study. I can withdraw participation at any time. No justifications are needed for withdrawing and no repercussions result from doing so. In case I withdraw my participation, my data will be deleted from the study. By withdrawing the lawfulness of data processing up to this point is not impacted. That means: The data from already existing study results do not have to be retroactively deleted but the data cannot be used from this point forward and has to be deleted.

To withdraw from the study, I can simply close the browser window. Data entered until this point will still be part of the study.

I had an opportunity to receive a copy of the participant information, the privacy statement, and this consent form via this link.

3. Information regarding the processing of person-related data

Information regarding the processing of person-related data as part of the study “Retrieval-Augmented Generation (RAG) chatbot for the Introduction to Data Mining lecture” (this document was designed on July 13, 2025).

With this information Bielefeld University complied with the duty to provide information regarding articles 13 and 14 of the European General Data Protection Regulation (GDPR) for the aforementioned processing of data.

Regarding the terms “person-related data”, “processing”, “responsible person”, etc. we refer to the definitions in article 4 of the GDPR.

1. Contact Data

2. Legal basis for processing

The legal basis for processing person-related data is based on the participants‘ consent as described in Art. 6, 1a GDPR.

The legal basis for processing IP addresses is Art. 6, 1e GDPR to protect the technical functioning of the website.

3. Processing of person-related data and purposes

As part of the study, the following data is recorded and processed.

• To protect the technical functioning of the website: IP addresses

The person-related data are exclusively used for the purposes described here and will only be used inside Bielefeld University.

Recording is performed via the RAG chatbot and are temporarily stored on the system’s server at Bielefeld University. The only person with access to the server is Lukas Gehring. Not person related data ist stored. The data on the server is wiped on September 30, 2026. Before publication, the data will further be checked for any accidental leaks of person-related data (as part of the user requests). Only after this additional step will the data be considered anonymized and published.

4. Data transmission

During the study, only Lukas Gehring has access to the server.

As a matter of principle, we do not pass on personal data which is processed during a visit to the study’s internet pages to third parties.

In some circumstances, this may be necessary if we are required by law to do so, for example, transmission to criminal prosecution authorities to clarify criminal offenses within the framework of the provisions of the German Code of Criminal Procedure (StPO) or for the purpose of asserting claims for damages in the event of copyright infringements.

Where technical service providers have access to personal data, this happens on the basis of a contract pursuant to Art. 28 of EU GDPR.

No personal data is transferred to countries outside the European Economic Area and associated countries (no third country transfer). If this should become necessary, you will be informed.

5. Deletion deadlines

IP addresses are temporarily stored (for 24 hours) in the log files of the RAG chatbot.

All other data is considered not person-related.

6. Data publication

The presentation of results to third parties, be it in form of scientific publications or presentations, is only performed in a form that does not permit inferences to the identity of single persons.

7. Right to withdraw consent

You have the right to withdraw your consent at any point. The lawfulness of any processing until withdrawal is not impacted (Art. 7 GDPR).

You can withdraw by closing the browser window.

8. Further rights as an impacted person

As an impacted person you can always exercise your rights as defined by the GDPR:

You have the right to information about the data stored relating to your person (Art. 15 GDPR).

If any data is inaccurate, you have a right to correction (Art. 16 GDPR).

You can ask for deletion or less processing according to Art. 17, 18 and 20 GDPR.

If you have consented to data processing and processing is performed by automatic means, you have a right to data transmission (Art. 20 GDPR).

You can exercise these rights by contacting the contact person for the study. Objections or complained can also be transmitted to the data protection officer.

You further have the right to file a complaint with the higher data protection authority (Art. 77 GDPR), such as the state data protection officer (responsible for Bielefeld University):

Please note that any deletion of data except IP addresses is not possible because we will not be able to identify your data due to anonymization. From this point forward, your rights to information, correction, deletion, less processing or withdrawal according to Articles 15-20 GDPR will not apply anymore.