Unsupervised pattern recognition in time series

Job offer at the FZI Karlsruhe.

Unsupervised pattern recognition in time series

Lupe

Context

Data, especially time series, is becoming increasingly important in many areas of everyday life. Analyzing this data makes it possible to make predictions about future developments. For example, historical time series data can be used to make statements about the future purchasing behavior of customers or the failure rate of electronic devices. Pattern recognition is an important aspect of time series analysis. This can be used to identify both typical and deviating patterns of behavior. The challenge in pattern recognition is to design suitable methods that are not only able to reliably recognize different patterns, but also to classify and distinguish variations in the data appropriately.

Tasks

  • You will research the current state of science and derive solutions based on this.
  • You will select one or more methods and use them to develop a concept for pattern recognition in time series.
  • You will analyze and evaluate the implemented methods and assess their quality for pattern recognition in time series.

Requirements

  • You have very good programming skills in Python.
  • You work independently and in a structured manner, are motivated and committed.
  • You have a very good command of written and spoken German and English.