One of the most significant challenges currently faced by experts, researchers, and data scientists is to explore potentially large amounts of time series efficiently. Special care has been taken to analyze temporal data in the last century, especially in the industrial domain: attention that skyrocketed along with the ability to produce, collect and treat those data. Nevertheless, time series analysis remains an open research subject, due to the complexity of this data type and the great amount of information it contains. Developing general methods for large-scale time series algorithms and data preprocessing is a pressing demand from companies wishing to explore big data.

This workshop offers a meeting opportunity for academic and industry researchers in the fields of machine learning, deep learning, data visualization, data mining, data engineering, and big data to discuss new areas of learning methods dedicated to time series in industrial environments. We encourage researchers and practitioners to submit papers describing original research addressing time series and scalable machine learning challenges.

This includes but is not restricted to the following topics:

  • Clustering and unsupervised learning
  • Time series classification
  • Deep learning approaches
  • Time series and attention mechanisms
  • Tensor time series
  • Geospatial time series
  • Online learning algorithms
  • Mixture models and model-based methods
  • Methods of detecting changes in evolving data (streaming and time series)
  • Federated methods
  • Theoretical frameworks for time series mining
  • Scalable algorithms for big data
  • Parallel and distributed computing for time series analyses
  • Visualization and interactive mining techniques
  • Future research challenges of time series analysis

Check the schedule on this page!

Past workshops