On the designing of two grains levels network intrusion detection system

Al‐Mamory, Safaa O.; Jassim, Firas S.

doi:10.1016/j.kijoms.2015.07.002

Cited by 20 publications

(11 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SPLR-based FS performs better than VFDT [19] on the KDD '99 dataset because the VFDT only considers the 20 selected feature sets for classification, while the proposed approach selects descriptive features and classification simultaneously. This specifies that SPLR has an exceptional capability to extract essential and rich information for the IDS.…”

Section: Resultsmentioning

confidence: 99%

“…However, several issues that remain to be addressed in feature selection; first, redundancy in datasets and combination process for FS techniques through any learning algorithm; second, an isolation and redundant feature selection as well as ineffective features selection from datasets-these problems lead to difficult stages for any learning algorithm. Furthermore, the Very Fast Decision Tree (VFDT) based on Hoeffing tree [19], uses information gain or gaini index for feature selection; it includes many refinement processes while training the model. The VFDT separates the features that are not promising before training the model.…”

Section: Related Workmentioning

confidence: 99%

“…Table 1 shows the KDD '99 data sets. For our experiment, we have selected a number of random samples from the original KDD '99 dataset as used in [19]. An extracted 1,774,985 instances as a training set listed in Table 4 and 67,688 instances of the independent testing set.…”

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

“…An increase in λ corresponds to a decrease in non-zero coefficients (selected feature variables) while the degree of sparsity can be measured by Equation (12). Here, the proposed technique (SPLR) is assessed with the similar training and testing datasets (1,074,985) as were used in [19,30,31]. The proposed SPLR feature selection technique is compared against the KDD '99 in order to show the potential in classification with better performance than [19] as well as other classifier models.…”

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

“…Here, the proposed technique (SPLR) is assessed with the similar training and testing datasets (1,074,985) as were used in [19,30,31]. The proposed SPLR feature selection technique is compared against the KDD '99 in order to show the potential in classification with better performance than [19] as well as other classifier models. However in the experimental results, we provide the discriminative feature selection and classification accuracy by SPLR as shown in Figure 5a,b respectively.…”

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

See 4 more Smart Citations

Network Intrusion Detection through Discriminative Feature Selection by Using Sparse Logistic Regression

et al. 2017

View full text Add to dashboard Cite

Intrusion detection system (IDS) is a well-known and effective component of network security that provides transactions upon the network systems with security and safety. Most of earlier research has addressed difficulties such as overfitting, feature redundancy, high-dimensional features and a limited number of training samples but feature selection. We approach the problem of feature selection via sparse logistic regression (SPLR). In this paper, we propose a discriminative feature selection and intrusion classification based on SPLR for IDS. The SPLR is a recently developed technique for data analysis and processing via sparse regularized optimization that selects a small subset from the original feature variables to model the data for the purpose of classification. A linear SPLR model aims to select the discriminative features from the repository of datasets and learns the coefficients of the linear classifier. Compared with the feature selection approaches, like filter (ranking) and wrapper methods that separate the feature selection and classification problems, SPLR can combine feature selection and classification into a unified framework. The experiments in this correspondence demonstrate that the proposed method has better performance than most of the well-known techniques used for intrusion detection.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

Section: Kdd Cup 1999 Datasetmentioning

confidence: 99%

See 3 more Smart Citations