Jens E. d'Hondt, PhD

Jens E. d'Hondt, PhD

Postdoctoral Scientist

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Page Not Found

This page does not exist. Try the publications, talks or CV pages instead.

Jens E. d'Hondt

Jens d’Hondt is a postdoctoral scientist at the Barcelona Supercomputing Center working on deep-learning downscaling of atmospheric simulations. PhD from TU Eindhoven on multivariate similarity search; papers at VLDB, SIGMOD and ICDE.

Blog

Plain-language explainers by Jens d’Hondt on multivariate time-series distance measures, similarity search, correlation discovery and deep-learning downscaling.

CV

Curriculum vitae of Jens d’Hondt: PhD in computer science from TU Eindhoven, postdoc at the Barcelona Supercomputing Center, publications, talks and service.

Publications

Papers by Jens d’Hondt on multivariate time-series similarity search, distance measures, correlation discovery and deep-learning downscaling of air-quality simulations, at VLDB, SIGMOD, ICDE and beyond.

Sitemap

Every page on this site, in one list.

Talks and presentations

Conference and seminar talks by Jens d’Hondt on multivariate time-series similarity search, correlation discovery and distance measures, with slides and recordings.

Posts

publications

Evaluation of computer-tailored motivational messaging in a health promotion context

Published in CONTEXT 2019 (Modeling and Using Context, LNCS), 2019

A randomized controlled trial evaluating computer-tailored motivational messaging in a health promotion intervention, showing improved attitudes toward persuasive attempts but limited longer-term behavior change.

Recommended citation: d'Hondt, J.E., Nuijten, R.C.Y., Van Gorp, P.M.E. (2019). Evaluation of Computer-Tailored Motivational Messaging in a Health Promotion Context. In: Bella, G., Bouquet, P. (eds) Modeling and Using Context. CONTEXT 2019. LNCS 11939, pp. 120–133.
Download Paper | Download Slides

Multivariate correlations discovery in static and streaming data

Published in VLDB 2022 (Proceedings of the VLDB Endowment 15(6)), 2022

VLDB 2022 paper introducing efficient algorithms (Correlation Detective) for discovering strong multivariate correlations in static and streaming data, outperforming the state-of-the-art typically by an order of magnitude.

Recommended citation: Koen Minartz, Jens E. d'Hondt, and Odysseas Papapetrou. 2022. Multivariate correlations discovery in static and streaming data. Proc. VLDB Endow. 15, 6 (February 2022), 1266–1278.
Download Paper | Download Slides

Efficient detection of multivariate correlations with different correlation measures

Published in The VLDB Journal, 2023

VLDB Journal article on efficient detection of multivariate correlations, supporting four different correlation measures and additional constraints for static and streaming data.

Recommended citation: d'Hondt, J.E., Minartz, K. & Papapetrou, O. Efficient detection of multivariate correlations with different correlation measures. The VLDB Journal 33, 481–505 (2024).
Download Paper

Beyond the Dimensions: A Structured Evaluation of Multivariate Time Series Distance Measures

Published in ICDE 2024 Workshops (MulTiSA), 2024

ICDE 2024 workshop paper presenting a structured evaluation and novel taxonomy of 12 multivariate time-series distance measures, showing that the optimal choice depends on the data and task at hand.

Recommended citation: J. E. d'Hondt, O. Papapetrou and J. Paparrizos, "Beyond the Dimensions: A Structured Evaluation of Multivariate Time Series Distance Measures," 2024 IEEE 40th International Conference on Data Engineering Workshops (ICDEW), Utrecht, Netherlands, 2024, pp. 107–112.
Download Paper | Download Slides

Multivariate Similarity Search - A Call for a New Breed of Similarity Search Algorithms

Published in ICDE 2024, 2024

ICDE 2024 talk revisiting similarity search under the lens of multivariate similarity measures, calling for a new breed of similarity search algorithms.

Recommended citation: O. Papapetrou and J. E. d'Hondt, "Multivariate Similarity Search - A Call for a New Breed of Similarity Search Algorithms," 2024 IEEE 40th International Conference on Data Engineering (ICDE), Utrecht, Netherlands, 2024, pp. 5662-5662.
Download Paper

A Survey on Time-Series Distance Measures

Published in arXiv, 2024

A comprehensive survey of over 100 time-series distance measures across 7 categories, covering distinctions and applications for both univariate and multivariate time series.

Recommended citation: Paparrizos, J., Li, H., Yang, F., Wu, K., d'Hondt, J.E., Papapetrou, O. (2024). A Survey on Time-Series Distance Measures. arXiv preprint arXiv:2412.20574.
Download Paper

MULISSE: Variable-Length Similarity Search for Multivariate Time Series

Published in ICDE 2025 Workshops (MulTiSA), 2025

Workshop paper (MulTiSa @ ICDE 2025) introducing MULISSE, an algorithm for exact variable-length k-NN subsequence search on multivariate time series, achieving up to two orders of magnitude speedup on synthetic datasets.

Recommended citation: Bart Pelok and Jens E. d'Hondt. 2025. MULISSE: Variable-Length Similarity Search for Multivariate Time Series. In Proceedings of the MulTiSA Workshop at the 41st IEEE International Conference on Data Engineering (ICDE 2025), Hong Kong.
Download Paper

A Structured Study of Multivariate Time-Series Distance Measures

Published in SIGMOD 2025 (Proceedings of the ACM on Management of Data 3(3)), 2025

SIGMOD 2025 paper presenting the most complete evaluation of multivariate time-series distance measures to date: 30 measures across 8 categories, 13 normalizations, 30 datasets, and 3 downstream tasks with rigorous statistical analysis.

Recommended citation: Jens E. d'Hondt, Haojun Li, Fan Yang, Odysseas Papapetrou, and John Paparrizos. 2025. A Structured Study of Multivariate Time-Series Distance Measures. Proc. ACM Manag. Data 3, 3 (SIGMOD), 1–29.
Download Paper

Generative Correlation Manifolds: Generating Synthetic Data with Preserved Higher-Order Correlations

Published in arXiv, 2025

White paper introducing Generative Correlation Manifolds (GCM), a computationally efficient synthetic data generation method that provably preserves the full correlation structure of a source dataset, from pairwise to higher-order interactions.

Recommended citation: d'Hondt, J.E., Punter, W.R., Papapetrou, O. (2025). Generative Correlation Manifolds: Generating Synthetic Data with Preserved Higher-Order Correlations. arXiv preprint arXiv:2510.21610.
Download Paper

MS-Index: Fast Top-k Subsequence Search for Multivariate Time Series under Euclidean Distance

Published in VLDB 2026 (Proceedings of the VLDB Endowment 19(2)), 2025

MS-Index is an exact index for top-k subsequence search on multivariate time series under Euclidean distance, supporting ad-hoc query channel selection and outperforming the state-of-the-art by one to two orders of magnitude.

Recommended citation: Jens E. d'Hondt, Teun Kortekaas, Odysseas Papapetrou, and Themis Palpanas. 2025. MS-Index: Fast Top-k Subsequence Search for Multivariate Time Series under Euclidean Distance. Proc. VLDB Endow. 19, 2, 99–112.
Download Paper

Physics-Constrained Deep Learning for Downscaling Atmospheric Chemistry Simulations: The Role of Auxiliary Forcings and Model Architecture

Published in EGU General Assembly 2026, 2026

EGU 2026 presentation on physics-constrained deep learning for downscaling MONARCH atmospheric chemistry simulations over the Iberian Peninsula, comparing CNN, RCAN, and EDSR architectures with high-resolution auxiliary forcings.

Recommended citation: d'Hondt, J. E. and Petetin, H.: Physics-Constrained Deep Learning for Downscaling Atmospheric Chemistry Simulations: The Role of Auxiliary Forcings and Model Architecture, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-9829.
Download Paper

Neural downscaling of air-quality simulations requires structural correction before spatial refinement

Published in Research Square (preprint), under review at Nature Communications, 2026

Preprint (under review at Nature Communications) showing that resolution-dependent model divergence, not missing spatial detail, dominates neural air-quality downscaling error, motivating structural correction before spatial refinement.

Recommended citation: d'Hondt, J., Petetin, H., Pérez García-Pando, C., Jorba, O., Guevara, M. Neural downscaling of air-quality simulations requires structural correction before spatial refinement. Preprint, Research Square (2026), under review at Nature Communications.
Download Paper

talks

VLDB 2022 - Correlation Detective

Published:

VLDB 2022 talk introducing Correlation Detective, a library for finding multivariate correlations in static and streaming data, with slides and recording.