Research Article
Classification of UNSW-NB15 dataset using Exploratory Data Analysis using Ensemble Learning
@ARTICLE{10.4108/eai.13-10-2021.171319, author={Neha Sharma and Narendra Singh Yadav and Saurabh Sharma}, title={Classification of UNSW-NB15 dataset using Exploratory Data Analysis using Ensemble Learning}, journal={EAI Endorsed Transactions on Industrial Networks and Intelligent Systems}, volume={8}, number={29}, publisher={EAI}, journal_a={INIS}, year={2021}, month={10}, keywords={KDD’99, UNSW-NB15, Ensemble algorithms, XGBoost, AdaBoost, Random Forest, Extra trees}, doi={10.4108/eai.13-10-2021.171319} }- Neha Sharma
Narendra Singh Yadav
Saurabh Sharma
Year: 2021
Classification of UNSW-NB15 dataset using Exploratory Data Analysis using Ensemble Learning
INIS
EAI
DOI: 10.4108/eai.13-10-2021.171319
Abstract
Recent advancements in machine learning have made it a tool of choice for different classification and analytical problems. This paper deals with a critical field of computer networking: network security and the possibilities of machine learning automation in this field. We will be doing exploratory data analysis on the benchmark UNSW-NB15 dataset. This dataset is a modern substitute for the outdated KDD’99 dataset as it has greater uniformity of pattern distribution. We will also implement several ensemble algorithms like Random Forest, Extra trees, AdaBoost, and XGBoost to derive insights from the data and make useful predictions. We calculated all the standard evaluation parameters for comparative analysis among all the classifiers used. This analysis gives knowledge, investigates difficulties, and future opportunities to propel machine learning in networking. This paper can give a basic understanding of data analytics in terms of security using Machine Learning techniques.
Copyright © 2021 Neha Sharma et al., licensed to EAI. This is an open access article distributed under the terms of the Creative Commons Attribution license, which permits unlimited use, distribution and reproduction in any medium so long as the original work is properly cited.