Benchmarking uni- and multivariate machine learning based classification of time series

Benchmarking uni- and multivariate machine learning based classification of time series

Zeitserien

Context

In the future, the vehicle will be understood as part of a networked vehicle and system environment. In order to enable effective and innovative data exploitation over the entire vehicle lifecycle, intelligent methods for processing time series data such as V2X vehicle data are required. Due to their challenging characteristics, knowledge extraction from time-series data is only possible with expensive approaches and by considering the temporal context. The work is supported in the context of the SofDCar project, for more information please visit https://sofdcar.de/language/de/.

Tasks

  • Understanding of current state of the art methods for the classification of time series
  • Define evaluation criteria for the evaluation of the different procedures
  • Implementation and comparison of uni- and multivariate methods for the classification of time series

Requirements

  • Programming skills in Python or R
  • Basic knowledge in the field of machine learning
  • Independent and solution oriented way of working