Elia.
van wolputte

CV portfolio academia contact
portfolio

Achievements that I am proud of. Concretely, publications and software.

Almost all of these achievements are the result of a team effort. This portfolio would be an empty wasteland if not for all my collaborators, to whom I wish to express my sincere gratitude.

Use the sidebar for quick and easy navigation.



Publications

Automatic Generation of Product Concepts from Positive Examples, with an Application to Music Streaming | 2022

  • Authors. K. Goyal, W. Meert, H. Blockeel, E. Van Wolputte, K. Vanderstraeten, W. Pijpops, K. Jaspers
  • Venue. The 34rd Benelux Conference on Artificial Intelligence and the 31th Belgian Dutch Conference on Machine Learning (BNAIC/BeNeLearn 22)
  • Resources. paper bibtex


avatar—Automated Feature Wrangling for Machine Learning | 2021

  • Authors. G. Verbruggen, E. Van Wolputte, S. Dumancic, L. De Raedt
  • Venue. International Symposium on Intelligent Data Analysis (IDA)
  • Resources. paper bibtex


The Effect of Hyperparameter Tuning on the Comparative Evaluation of Unsupervised Anomaly Detection Methods | 2021

  • Authors. J. Soenen, E. Van Wolputte, L. Perini, V. Vercruyssen, W. Meert, J. Davis, H. Blockeel
  • Venue. KDD'21 Workshop on Outlier Detection and Description (ODD 21)
  • Resources. paper code bibtex


VisualSynth: Democratizing Data Science in Spreadsheets | 2020

  • Authors. C. Gautrais, Y. Dauxais, S. Kolb, A. Jain, M. Kumar, S. Teso, E. Van Wolputte, G. Verbruggen, L. De Raedt
  • Venue. Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECMLPKDD-20)
  • Resources. paper demo bibtex


Missing value imputation with MERCS: a faster alternative to MissForest | 2020

  • Authors. E. Van Wolputte, H. Blockeel
  • Venue. International Conference on Discovery Science
  • Resources. paper slides poster code bibtex


SynthLog: A Language for Synthesising Inductive Data Models | 2019

  • Authors. Y. Dauxais, C. Gautrais, A. Dries, A. Jain, S. Kolb, M. Kumar, S. Teso, E. Van Wolputte, G. Verbruggen, L. De Raedt
  • Venue. Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECMLPKDD-19)
  • Resources. paper bibtex


Anomaly detection for cern beam transfer installations using machine learning | 2019

  • Authors. T. Dewitte, W. Meert, E. Van Wolputte, P. Van Trappen
  • Venue. The 17th International Conference on Accelerator and Large Experimental Control Systems (ICALEPCS 2019)
  • Resources. paper slides code bibtex


MERCS: multi-directional ensembles of regression and classification trees | 2018

  • Authors. E. Van Wolputte, E. Korneva, H. Blockeel
  • Venue. Thirty-Second AAAI Conference on Artificial Intelligence
  • Resources. paper code bibtex


COBRAS: Interactive clustering with pairwise queries | 2018

  • Authors. T. Van Craenendonck, S. Dumancic, E. Van Wolputte, H. Blockeel
  • Venue. International Symposium on Intelligent Data Analysis
  • Resources. paper code demo bibtex


Code-modulated visual evoked potentials using fast stimulus presentation and spatiotemporal beamformer decoding | 2017

  • Authors. B. Wittevrongel, E. Van Wolputte, MM Van Hulle
  • Journal. Scientific reports
  • Resources. paper bibtex



Software

Python package for the MERCS algorithm | code

  • Technologies. python, sklearn, xgboost, lightgbm, numpy, graphviz, pip, conda, networkx
  • Collaborators. E. Korneva, H. Blockeel
  • Context. Python package for the MERCS-algorithm, that builds multi-directional ensembles of decision trees. Includes tools for visualization and analysis after the fact.


Databases take-home project | code

  • Technologies. python, jupyter, mysql, mariadb, xampp
  • Collaborators. B. Berendt, A. Kimmig, B. Bogaerts, J. Bekker, S. Kolb
  • Context. Public repository of a take-home project for 100+ students in a university course on databases, auto-generated from a private 'backend' repository that also can run, correct, analyze and grade student solutions. The backend repo cannot be released, as it continues to be used at KU Leuven.


Anomaly detection pipeline for beam transfer equipment at CERN | code

  • Technologies. python, sklearn, numpy, pandas, spark, hdfs
  • Collaborators. P. Van Trappen, A. Halilovic, T. Dewitte, K. Bringmans, T. Claessens
  • Context. Proof-of-concept anomaly detection pipeline for beam transfer equipment at CERN. Developed by master thesis students, under supervision of P. Van Trappen (CERN) and myself.


Cookiecutter repository for datascience projects | code

  • Technologies. python, sklearn, numpy, pandas, pip, conda, jupyter, altair
  • Context. A parametrized template repository (cookiecutter is the template 'language') for datascience projects. Simplifies organization and configuration for common objectives such as pip release, docs, testing, CI and proper experimental methodology.


Python package for the ALSO algorithm | code

  • Technologies. python, sklearn, numpy, pandas
  • Context. Small python package that implements the ALSO anomaly detection algorithm. We were not involved in the development of this algorithm, but we realised no convenient opensource solution existed.


AFFE: Another Framework For Experiments | code

  • Technologies. python, sklearn, numpy, pandas, pip, conda, bash, git
  • Collaborators. J. Soenen, W. Meert, J. Bekker
  • Context. AFFE is a simple and modular python toolbox that makes it easier to properly organize and conduct reproducible experiments. Initially a loose collection of useful scripts, it organically grew into a (semi-)organized library that others have used and enjoyed.