This is our implementation of Brauckhoff et. als article Anomaly extraction in backbone networks using association rules. A copy of our report is included in this repository as PDF.

The main modules used are

• detector.py: Main part of algorithm used to process data for anomalies
• elasticquery.py: Used to fetch data from ElasticSearch server or load file from disk
• run_all.py: Runs whole program and extracts data, see this first for basic usage

A config file settings.py has to be created with username and password to the ElasticSearch server for usage as follows.

server = 'https://es-elk-syslog.etc.com:1982'
index = 'index*'
username = 'username'
password = 'password'