Network Anomaly Detector using Machine Learning
DOI:
https://doi.org/10.14419/ijet.v7i3.12.15914Published:
2018-07-20Keywords:
syslogs, detector, predictor, random forest, supervisedAbstract
The 4G network consists of a network of routers on each tower that decides where a certain packet must be switched to. These routers like any other hardware device is subject to failure due to number of factors such as threshold violations and problems with its tuning. The routers and other relevant hardware devices undergo various maintenance cycles that can sometimes be wasteful as the hardware may be replaced even when in complete working condition. This is a measure taken to ensure the network is always up and running. This measures has proven to be expensive and alternative solutions have been looked for. To alleviate the costs involved in the maintenance of these routers, a system will be developed to perform applications such as report failures, find the root cause and implement a remedial action automatically.The prediction of failures in the routers is achieved by unsupervised machine learning while will be trained to pick up anomalies from a continuous stream of log messages sent to system which is then analyzed. The anomaly data is then used to schedule maintenance runs more effectively.
References
[1] https://nishanthu.github.io/articles/ClusteringUsingRandomForest.html
[2] https://www.stat.berkeley.edu/~breiman/RandomForests/cc_home.htm#intro
[3] https://mapr.com/ebooks/spark/08-unsupervised-anomaly-detection-apache-spark.html
[4] https://labs.genetics.ucla.edu/horvath/RFclustering/RFclustering.htm
[5] https://labs.genetics.ucla.edu/horvath/RFclustering/RFclustering/RandomForestHorvath.pdf
[6] http://pages.cs.wisc.edu/~matthewb/pages/notes/pdf/ensembles/RandomForests.pdf
How to Cite
License
Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution Licensethat allows others to share the work with an acknowledgement of the work''s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal''s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Accepted 2018-07-20
Published 2018-07-20