SoSe 2022

The lecture deals with the basics of pattern recognition in time series (e.g., sensor signals) and spatially distributed gathered data (e.g., in sensor networks). Among others, the following topics are covered:

• Fundamentals (e.g., segmentation of time series, correlation of data, features to describe temporal/spatial data)
• similarity measurement of time series
• clustering/classification
• forecasting
• motif recognition
• anomaly detection with various techniques (e.g., Nearest Neighbor, Support Vector Machines)
• various example applications (signature verification, collaborative hazard warning in vehicles, activity recognition, context recognition with smartphones, and others)

Moreover, we cover novel topics related to the deep-learned-based processing of time series. This includes the fundamentals of deep learning for time series, convolutional and recurrent neural networks to model time series, and other state-of-the-art techniques for processing time series with deep learning methods.