From f85db086f0bec2bc8d549dad4028c851ddaccd78 Mon Sep 17 00:00:00 2001 From: Abd Errahmane Kiouche <abd-errahmane.kiouche@etu.univ-lyon1.fr> Date: Thu, 25 Jun 2020 18:23:53 +0200 Subject: [PATCH] Add new file --- src/Training_Phase/Clustering/README.md | 12 ++++++++++++ 1 file changed, 12 insertions(+) create mode 100644 src/Training_Phase/Clustering/README.md diff --git a/src/Training_Phase/Clustering/README.md b/src/Training_Phase/Clustering/README.md new file mode 100644 index 0000000..37ababc --- /dev/null +++ b/src/Training_Phase/Clustering/README.md @@ -0,0 +1,12 @@ +I don't own this part of code. This code is originally bases of Streamspot clustering code with slight modificiations. + +Original source : https://github.com/sbustreamspot/sbustreamspot-train + + +For the clustering part we've used k-medoids algorithm proposed by (Emaad manzoor et al 2016) with slight modifications. + + +References : + +MANZOOR, Emaad, MILAJERDI, Sadegh M., et AKOGLU, Leman. Fast memory-efficient anomaly detection in streaming heterogeneous graphs. +In : Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016. p. 1035-1044. \ No newline at end of file -- GitLab