Explore the test data and brainstorm RTDIP component ideas

[luccalb]

Explore the test data provided by shell and brainstorm ideas for RTDIP components that ensure better data quality or identify trends/anomalies

[Timm638]

Some Brainstorming done with scitkit-learn as inspiration:

• Dimensionality Reduction (Reduce redundant data, e. g. which sources correlate strongly with each other?)
• Normalization of Data (By Z-Mean, Min-Max-Scaling, ...)
• Other Preprocessing Methods: Map scalar data into bins, One-hot encoding
• Trend Identification: Linear Regression, ARIMA

Other notes:

• When we implement these functions, in which format should be work with the data? Convert everything into a pandas Dataframe and then back to the original format?