Reconstructing Improper Values with Local Centered Moving Weighted Mean

Non-Causal Step

This step replaces any improper values in the data vector with the Weighted Mean of the neighbourhood values.

Input Parameters

1. Input data including improper values
2. Definition of improper values (‘nan’, ‘inf’, or ‘null’, etc.)
3. Weights for all available values
4. Local window size

Output Parameters

1. Output data excluding improper values

Workflow

Algorithm

Weighted Mean

References

• J. Han, M. Kamber and J. Pei, Data Mining - Concepts and Techniques, 3rd ed., Amsterdam: Morgan Kaufmann Publishers, 2012.