Comparison of Machine Learning algorithms performance in detecting network intrusion

Jalil, Kamarularifin Abd; Kamarudin, Muhammad Hilmi; Masrek, Mohamad Noorman

doi:10.1109/icnit.2010.5508526

Cited by 21 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By comparing the experiments carried out by Jalill et al [38] and Katkar and Kulkarni [40], we have observed that SVM algorithm predict DOS-DDOS more accurately on the dataset UNSW_NB15 compared to the KDD'99 dataset (Acc_ SVM_UNSW_NB = 92.28% > Acc_ SVM_KDD = 62.5 %). This important difference according to W. Xingzhu [42] is caused by the redundant records on the KDD'99 dataset and SVM has slower training on high dimensional datasets.…”

Section: ) Use Model Evaluation Metricsmentioning

confidence: 91%

“…The accuracy can range from Acc=62.5% by using KDD'99 dataset and SVM algorithm to Acc=99.92% with Decision Tree J.48 algorithm and KDD'99 dataset. Indeed, according to the experiment of Jalill et al ( 2010) [38] based on the KDD'99 dataset, the Support Vector Machine (SVM) algorithm has a serious problem in accurately detecting DOS-DDOS attacks compared to the Decision Tree J.48 algorithm which shows high prediction accuracy that exceed 99%. The experiments based on the NB, C4.5, RF algorithms and UNSW_NB15 dataset realized by Bellouch et al (2018) [39], has shown that the prediction accuracy obtained by RF (Acc_ RF = 99.94%) is better than C4.5 (Acc_ C4..5 = 95.82%) and SVM (Acc_ SVM = 92.28%).…”

Section: ) Use Model Evaluation Metricsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting DOS-DDOS Attacks: Review and Evaluation Study of Feature Selection Methods based on Wrapper Process

Bouzoubaa¹,

Taher²,

Nsiri³

2021

IJACSA

View full text Add to dashboard Cite

Now-a-days, Cybersecurity attacks are becoming increasingly sophisticated and presenting a growing threat to individuals, private and public sectors, especially the Denial Of Service attack (DOS) and its variant Distributed Denial Of Service (DDOS). Dealing with these dangerous threats by using traditional mitigation solutions suffers from several limits and performance issues. To overcome these limitations, Machine Learning (ML) has become one of the key techniques to enrich, complement and enhance the traditional security experiences. In this context, we focus on one of the key processes that improve and optimize Machine Learning DOS-DDOS predicting models: DOS-DDOS feature selection process, particularly the wrapper process. By studying different DOS-DDOS datasets, algorithms and results of several research projects, we have reviewed and evaluated the impact on used wrapper strategies, number of DOS-DDOS features, and many commonly used metrics to evaluate DOS-DDOS prediction models based on the optimized DOS-DDOS features. In this paper, we present three important dashboards that are essential to understand the performance of three wrapper strategies commonly used in DOS-DDOS ML systems: heuristic search algorithms, meta-heuristic search and random search methods. Based on this review and evaluation study, we can observe some of wrapper strategies, algorithms, DOS-DDOS features with a relevant impact can be selected to improve the DOS-DDOS ML existing solutions.

show abstract

Section: ) Use Model Evaluation Metricsmentioning

confidence: 91%

Section: ) Use Model Evaluation Metricsmentioning

confidence: 99%

Predicting DOS-DDOS Attacks: Review and Evaluation Study of Feature Selection Methods based on Wrapper Process

Bouzoubaa¹,

Taher²,

Nsiri³

2021

IJACSA

View full text Add to dashboard Cite

show abstract

“…Also shows the resource performance by the algorithms. Jalil et al, (2010) have use three ML algorithms for his research namely decision tree (J48), support vector machine and neural network. He used KDD 99 dataset to analyse the accuracy, detection rate and false alarm rate and compared these three algorithms by their results and shown that decision tree algorithms suit best for his research.…”

Section: Related Workmentioning

confidence: 99%

“…From his work, SVM showed a high performance of F-measure about 72%. (Ribeiro et al, 2015) 90% for spam - (Yang et al, 2013) 77% for twitter and yelp - (Alsudais et al, 2014) 90% for reporters and 84% for reportees - (Sinha et al, 2016) 96.07% - (Meda et al, 2014) 93.6% F-measure - (Chen et al, 2015c) Decision tree High 96.51% - (Ribeiro et al, 2015) 99% for IDS - (Jalil et al, 2010) 87.6% for spam - (Yang et al, 2013) 92% for spam reporters and 90% for spam reportees - (Sinha et al, 2016) 92% F-measure for C4.5 - (Chen et al, 2015c) Naïve Bayes High 86.63% - (Ribeiro et al, 2015) 70.9% F-measure - (Chen et al, 2015c) K-NN Average 84% for reporters and 89% for reportees - (Sinha et al, 2016) 90.5% F-measure - (Chen et al, 2015c) SVM Average 88.75% - (Du and Fang, 2004) 57% around for IDS - (Jalil et al, 2010) 79.9% F-measure - (Chen et al, 2015c) Bayes network Average 83.3% for spam - (Yang et al, 2013) 81.9% F-measure - (Chen et al, 2015c) Xu et al (2016) introduced a new point of view to efficiently detect spam in social networks. He collected two types of datasets from the Twitter and Facebook using the application programming interface, which contains both spam and non-spam contents.…”

Section: Related Workmentioning

confidence: 99%

Detecting spams in social networks using ML algorithms - a review

Murugan

Devi

2018

IJEWM

View full text Add to dashboard Cite

The social network, by the name which has popularised in today's world and growing rapidly at all times and controlling over mankind. The social networks like Twitter, Facebook, and LinkedIn, etc., have become a regular and daily usage of many people. It becomes a good mediator for the people who would like to share some posts, are some of their own videos, or some messages. But there has been major issues that the particular user of the social networks like Twitter and Facebook have the problem of indiscipline actions which we call as spam, by the third person who is knowingly doing this to spoil their intention and good opinion upon each other. Also, these spams help to steal information about the people who using social networks. In this paper, we study and analyse about the spam in social networks and machine learning algorithms to detect such kind of spams. This paper also focuses on the ML algorithms detection rate and false positive rate over different datasets.

show abstract

“…Their focus was on measuring the performance of classification algorithm based on True Positive rate and False Positive rate. Jalil and Masrek [12] in their paper evaluated the performance of J48 classification algorithm and compared its result to two other machine learning algorithm that are Neural Network and support Vector Machine based on the detection rate, false alarm rate and accuracy of classification based on attack type. This paper compares the performance of Naï ve Bayes and J48 algorithms on KDD"99 data set and studies the effects of removing redundancy from the dataset by applying preprocessing filter (i.e.…”

Section: Related Workmentioning

confidence: 99%

Comparative Analysis of Classification Algorithms on KDD'99 Data Set

Arora¹,

Bhatia²,

Singh³

2016

IJCNIS

View full text Add to dashboard Cite

Abstract-Due to the enormous growth of network based services and the need for secure communications over the network there is an increasing emphasis on improving intrusion detection systems so as to detect the growing network attacks. A lot of data mining techniques have been proposed to detect intrusions in the network. In this paper study of two different classification algorithms has been carried out: Naï ve Bayes and J48. Results obtained after applying these algorithms on 10% of the KDD"99 dataset and on 10% of the filtered KDD"99 dataset are compared and analyzed based on several performance metrics. Comparison between these two algorithms is also done on the basis of the percentage of correctly classified instances of different attack categories present in both the datasets as well as the time they take to build their classification models.Overall J48 is a better classifier compared to Naï ve Bayes on both the datasets but it is slow in building the classification model.

show abstract

Comparison of Machine Learning algorithms performance in detecting network intrusion

Cited by 21 publications

References 7 publications

Predicting DOS-DDOS Attacks: Review and Evaluation Study of Feature Selection Methods based on Wrapper Process

Predicting DOS-DDOS Attacks: Review and Evaluation Study of Feature Selection Methods based on Wrapper Process

Detecting spams in social networks using ML algorithms - a review

Comparative Analysis of Classification Algorithms on KDD'99 Data Set

Contact Info

Product

Resources

About