This paper presents the second-place methodology in the Volvo Discovery Challenge at ECML-PKDD 2024, where we used Long Short-Term Memory networks and pseudo-labeling to predict maintenance needs for a component of Volvo trucks. We processed the training data to mirror the test set structure and applied a base LSTM model to label the test data iteratively. This approach refined our model's predictive capabilities and culminated in a macro-average F1-score of 0.879, demonstrating robust performance in predictive maintenance. This work provides valuable insights for applying machine learning techniques effectively in industrial settings.
@article{arxiv.2409.13877,
title = {Achieving Predictive Precision: Leveraging LSTM and Pseudo Labeling for Volvo's Discovery Challenge at ECML-PKDD 2024},
author = {Carlo Metta and Marco Gregnanin and Andrea Papini and Silvia Giulia Galfrè and Andrea Fois and Francesco Morandin and Marco Fantozzi and Maurizio Parton},
journal= {arXiv preprint arXiv:2409.13877},
year = {2024}
}
Comments
2nd place at ECML-PKDD Discovery Challenge https://www.hh.se/english/about-the-university/events/discovery-challenge-ecml-pkdd-2024.html