Navigation auf uzh.ch
Navigation auf uzh.ch
This pages lists the open BSc. and MSc. thesis descriptions, as well as the master projects opportunities currently available in the DDIS research group.
If you are interested in any of the listed projects, please do not hesitate to contact the person mentioned in the open topic description.
If there are currently no open topics but you are generally interested in our research (see https://www.ifi.uzh.ch/en/ddis/research.html), or if you would like to propose a thesis about your own idea, you can send us an email to ddis-theses@ifi.uzh.ch.
Voting Advice Applications (VAA) such as Smartvote or Wahl-O-Mat depend on long questionnaires to recommend parties or candidates to a user. Recently, adaptive questionnaires have been introduced to optimize the data collection process and speed up recommendations in such applications. These adaptive questionnaires select the subsequent question based on the individual response profile of a user and, therefore, avoid redundancies.
The goal of this work is to conduct an experiment on users' behaviour and trust in such adaptive questionnaires. To this end, a self-hosted, adaptive VAA was built based on Smartvote. The student(s) should extend the front-end and back-end of this application to allow the test of a small number of hypotheses in an experiment with paid participants.
If interested, please contact us at the email address below. We can provide a more detailed description during a meeting.
Note: This project can be offered as a Bachelor or Master thesis, as well as a Master project. The hypotheses will be adjusted accordingly.
Requirements: Proficiency in Python for backend algorithm development, knowledge of PostgreSQL and Redis for database management and caching, and expertise in Angular, NestJS, and Nginx for front-end integration and deployment.
Contact: Fynn Bachmann
Goal: Framework for experimenting with the diagnosis of patients with rare diseases.
The RareSim project aims at developing novel AI-based approaches to support physicians in their tasks of helping rare disease patient (in particular for diagnosis). Whilst these more than 7000 rare diseases each afflict less than 2000 people, it is estimated that for all these diseases together affect around 500’000 people in Switzerland.
Tasks during this project:
For more information, contact us!
Start date: after November 2024
Contact: Oana Inel, Pascal Andermatt
Explainable AI has become a popular research topic in many domains including recommender systems. Despite the popularity, the evaluation of explanation methods is still an open research topic which is why it is important to provide a research infrastructure that allows for a reproducible and comparative evaluation.
In this project, the students will improve and extend the first version of such an evaluation framework for recommender systems explanations that is based on the Cornac framework.
Tasks:
The project is suitable for a group of 3-4 students and the scope will be adapted depending on the group size. If you have any questions or want to learn more about the project, send me an email. I'm happy to arrange a meeting with you.
Requirements: The existing code base is written in Python and Cython, so good programming and code comprehension skills in these languages are required. It is further helpful to have prior knowledge about recommender systems.
Start date: February 2025
Contact: Kathrin Wardatzky
Explanations for recommendations often tell a user why an item was recommended, but recent work suggests that this is not the optimal method to help a user make an informed decision and improve their understanding of why they received the recommendations.
In this thesis, the student will explore different methods to explain recommendations by giving the user control over the recommendations.
The tasks include:
If you are interested in this topic and/or want to learn more about it, don't hesitate to email me.
Contact: Kathrin Wardatzky
Explanations for recommendations often justify to a user why one recommended item has been recommended but there is little work on how other explanation methods compare to this explanation type. The goal of this thesis is to explore two alternative methods to generate explanations, debate dynamics and counterfactuals.
Reasoning with debate dynamics has been an effective method for link prediction tasks. Given the proximity to the recommendation problem, it can be a promising method to generate and explain recommendations.
Counterfactual explanations have become a popular method to extract possible actions that change the predicted outcome of a model and allow a user to take control over the recommendations.
In this thesis, the student will:
If you are interested in this thesis and would like to learn more, please reach out via email.
Contact: Kathrin Wardatzky
