subsequence_fast_mcd

Subsequence Fast-MCD

Citekey	-
Source	own
Learning type	semi-supervised
Input dimensionality	univariate

Dependencies

• python 3
• sklearn

Notes

• We first split the univariate timeseries into smaller subsequences using sliding windows.
• Each window is then a multidimensional object fed into the Fast-MCD algorithm.
• Afterward, Fast-MCD works on the subsequences.

Subsequence Fast-MCD therefore outputs anomaly scores for windows. The results require post-processing. The scores for each point can be assigned by aggregating the anomaly scores for each window the point is included in.

U can use the following code snippet for the post-processing step in TimeEval (default parameters directly filled in from the source code):

from timeeval.utils.window import ReverseWindowing
# post-processing for Subsequence Fast-MCD
def post_sfmcd(scores: np.ndarray, args: dict) -> np.ndarray:
    window_size = args.get("hyper_params", {}).get("window_size", 100)
    return ReverseWindowing(window_size=window_size).fit_transform(scores)