Directly to

Data Processing for Engineers and Scientists (DaPro)

[LV; module ]


Course description

The course teaches basic knowledge on data acquisition, data preparation, data analysis and data visualization. This includes elementary knowledge in image processing. Additionally, data-based and data-assisted modeling, e.g. through interpolation or artificial neural networks, are addressed.

The course is accompanied by an extensive computer lab in Python. Additional material is available (templates, mini tutorials, digital course material, ...).

Content (subject to minor modifications)

  • motivation and introduction
  • notation
  • basics in linear algebra, stochastics and statistics
  • data acquisition
    • data sources and characterization thereof
    • basic data file formats (XML, HDF5, ...)
    • meta data
  • sampling techniques
    • Latin hypercube sampling
    • Design Of Experiments (DOE)
    • concept of greedy sampling
  • data analysis
    • basic statistic evaluation
    • denoising; filtering
    • data analysis in the frequency domain
    • Pearson correlation coefficient
    • empirical distribution function
  • feature extraction and dimensionality reduction
    • curse of dimensionality
    • Principal Component Analysis/Proper Orthogonal Decomposition
    • clustering
  • kernel methods
    • interpolation and regression; Gaussian process modeling
    • identification of the nugget parameter
  • image processing
    • filtering (via Fast Fourier Transform), e.g. Gauss blur
    • erosion, dilation
    • edge detection
  • introduction to feedforward neural networks
    • general formulation
    • conceptual ideas for training algorithms
    • input-/output-data preparation
  • visualization of scientific data with examples