Supervision

A summary of the available projects and the theses I have supervised or co-supervised. I am also open to discussing other topics.

Available Theses
MSc/BSc
  • Unified Evaluation Metrics for Time Series Imputation.
  • Early Detection of Diseases using Federated Learning.
Current Theses
PhD
  • Quentin Nater ( 2025-now): Scalable Data Cleaning Systems for Extreme Large Time Series. Co-supervised with Philippe Cudré-Mauroux.
  • Zakhar Tymchenko (2023-now): Holistic Evaluation of Time Series Data Quality (On hold). Co-supervised with Philippe Cudré-Mauroux.
MSc
  • Flaminia Trinca: Integrating AutoML Model Selection into ImputeGAP.
  • Maurice Amon: Time Series Imputation using Large Language Models.
  • Flavien Buron: Extensive Comparison of Data Transformation Techniques for Graph Time Series Imputation.
Completed Theses
PhD
  • Abdelouahab Khelifati (2025): A Holistic Approach for Time Series Management: Unifying Data Storage and Data Processing. Co-supervised with Philippe Cudré-Mauroux.
  • Ines Arous (2022): Human-AI Collaborative Approaches for Open-Ended Data Curation. Dimitris N. Chorafas Award for Best Thesis in Computer Science. Co-supervised with Philippe Cudré-Mauroux and Jie Yang.
  • Artem Lutov (2019): Unsupervised and Parameter-Free Clustering of Large Graphs for Knowledge Exploration and Recommendation. Co-supervised with Philippe Cudré-Mauroux.
MSc/BSc
  • Matej Kutirov (2025, MSc): Automated Model Selection for Time Series Data Generation
  • Andy Keller (2025, MSc): Multimodal Unsupervised Streaming Event Detection
  • Dana Rim Ghousson (2025, MSc): Prediction of Environmental Hazards using Data Fusion
  • Quentin Nater (2024, MSc): ImputeGAP: A Python library for Time Series Imputation Techniques
  • Martin Poplawski (2024, MSc): Empirical Evaluation of Multimodal Data Fusion Approaches for Event Detection
  • Dardana Jaha (2024, MSc): A Comprehensive Analysis and Evaluation of Lossless Compression Techniques for Time Series Data
  • Lucien Gremaud (2024, BSc): Automatic Parametrization of Time Series Imputation Techniques
  • Mirko Bristle (2023, MSc): A Lightweight Meta-Learning Framework to Assess the Validity of Model Selection in Time Series Forecasting
  • Brian Schweigler (2023, MSc): Visualizing Time Series Recovery using ImputeBench
  • Manuel Mondal (2023, MSc): Evaluating Link Prediction for Emerging Nodes in Dynamic Networks
  • Flavien Buron (2023, BSc): Evaluation of Neural Networks Imputation using ImputeBench
  • Jonathan Bernhard (2023, BSc): An Empirical Comparison of Time Series Forecasting Techniques
  • Zakhar Tymchenko (2022, MSc): Evaluating the Impact of Imputation on Time Series Tasks
  • Jana Stojanovic (2022, BSc): Evaluating the Impact of Time Series Features on Downstream Tasks
  • Guillaume Chacun (2021, MSc): A Configuration-Free Repair of Time Series
  • Louis Müller (2021, MSc): Multiclass Classification of Open-ended Answers
  • Gabriela-Carmen Voroneanu (2021, MSc): Benchmark of Time Series Management Systems using Analytical Queries
  • Julia Eigenmann (2021, MSc): Evaluating Text Classification Models on Multilingual Documents
  • Jonas Fontana (2021, BSc): Comparison of Synthetic Time Series Data Generation Techniques
  • Adrian Hänni (2020, MSc): Correlation-based Anomaly Detection in Time Series
  • Mili Biswas (2019, MSc): Incremental Enrichment of Taxonomies using Transfer Learning
  • Ahana Malik (2018, MSc): Trend Prediction on Fashion Data
  • Manuel Mondal (2018, BSc): Reducing Polarization in Social Media by Balancing Content Exposure
  • Oliver Stapleton (2017, MSc): Real-Time Centroid Decomposition of Streams of Time Series
  • Leutrim Kaleci (2017, MSc): Modeling the Evolution of Fashion Trends using Matrix Factorization Techniques
  • Zakhar Tymchenko (2017, BSc): Empirical Comparison of Incremental Matrix Decomposition Techniques
  • Qian Liu (2016, MSc): Implementation of Centroid Decomposition Algorithm on Big Data Platforms - a Comparison Between Spark and Flink
  • Laura Rettig (2015, MSc): Online Anomaly Detection over Big Data Streams. JAACS Best Thesis Award and the Faculty Award for Best Thesis in Theoretical Sciences.
  • Jonathan Nagel (2014, MSc): Decomposition of Time Series Subsequences
  • Eszter Börzsönyi (2013, MSc): Evaluating the Optimality of Centroid Decomposition
  • Michał Kołtonik (2012, MSc): Visualization and Analysis of Hydrological Data