Visual-Interactive Preprocessing of Time Series Data

Time series data is an important data type in many different application scenarios. Consequently, there are a great variety of approaches for analyzing time series data. Within these approaches different strategies for cleaning, segmenting, representing, normalizing, comparing, and aggregating time series data can be found. When combining these operations, the time series analysis preprocessing workflow has many degrees of freedom. To define an appropriate preprocessing pipeline, the knowledge of experts coming from the application domain has to be included into the design process. Unfortunately, these experts often cannot estimate the effects of the chosen preprocessing algorithms and their parameterizations on the time series. We introduce a system for the visual-interactive exploitation of the preprocessing parameter space. In contrast to 'black box'-driven approaches designed by computer scientists based on the requirements of domain experts, our system allows these experts to visual-interactively compose time series preprocessing pipelines by themselves. Visual support is provided to choose the right order and parameterization of the preprocessing steps. We demonstrate the usability of our approach with a case study from the digital library domain, in which time-oriented scientific research data has to be preprocessed to realize a visual search and analysis application