In the EGAN (Zenati, Houssam, et al. "Efficient gan-based anomaly detection." ICLR WorkShop (2018). ), the KDDCUP99 ten percent dataset is actual a multi-dimensional point dataset instead of a multivariate time series one, which is the same as the website describes (http://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html). However, this work campare with EGAN on this dataset but without any dataset describtion on how can it be regarded as a time series one. Is it wrong?
In the EGAN (Zenati, Houssam, et al. "Efficient gan-based anomaly detection." ICLR WorkShop (2018). ), the KDDCUP99 ten percent dataset is actual a multi-dimensional point dataset instead of a multivariate time series one, which is the same as the website describes (http://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html). However, this work campare with EGAN on this dataset but without any dataset describtion on how can it be regarded as a time series one. Is it wrong?