Data Corpus

We included in this webpage the original data set used and the results obtained by applying track before detect. bellow is the description of the dataset. N.B:Data are ordered, timestamped, single-valued metrics. All data files contain anomalies, unless otherwise noted.

Real data

Artificial data


To facilitate the interpretation of the results after using track before detect, we used three types of figures (scatter, density_heatmap and line);A CSV file is generated for each analyzed using a case where you can find the calculated anomaly score

below is a mix of the results from existing methods like Gaussian Mode, Regression Model, KNN, LOF, STREAM, INFLO, Autoencoders