scholarly journals Research on hybrid intrusion detection method based on the ADASYN and ID3 algorithms

2021 ◽  
Vol 19 (2) ◽  
pp. 2030-2042
Author(s):  
Yue Li ◽  
◽  
Wusheng Xu ◽  
Wei Li ◽  
Ang Li ◽  
...  
Keyword(s):  
Intrusion Detection ◽  
Decision Tree ◽  
False Alarm ◽  
False Alarm Rate ◽  
Detection Method ◽  
Detection System ◽  
Performance Comparison ◽  
Tree Model ◽  
Network Intrusion ◽  
Id3 Algorithm

<abstract> <p>Intrusion detection system plays an important role in network security. Early detection of the potential attacks can prevent the further network intrusion from adversaries. To improve the effectiveness of the intrusion detection rate, this paper proposes a hybrid intrusion detection method that utilizes ADASYN (Adaptive Synthetic) and the decision tree based on ID3 algorithm. At first, the intrusion detection dataset is transformed by coding technology and normalized. Subsequently, the ADASYN algorithm is applied to implement oversampling on the training set, and the ID3 algorithm is employed to build a decision tree model. In addition, the model proposed by the research is evaluated by accuracy, precision, recall, and false alarm rate. Besides, a performance comparison is conducted with other models. Consequently, it is found that the combined model based on ADASYN and ID3 decision tree proposed in this research possesses higher accuracy as well as lower false alarm rate, which is more suitable for intrusion detection tasks.</p> </abstract>

Network Intrusion Detection System Using C4.5 Algorithm

2017 ◽  
Vol 10 ◽  
pp. 15
Author(s):  
Saurabh Fegade ◽  
Amey Bhadkamka ◽  
Kamlesh Karekar ◽  
Jaikishan Jeshnani ◽  
Vinayak Kachare
Keyword(s):  
Intrusion Detection ◽  
False Alarm ◽  
False Alarm Rate ◽  
Intrusion Detection System ◽  
Detection System ◽  
Network Intrusion Detection ◽  
Support Vector ◽  
Network Intrusion ◽  
C4.5 Algorithm ◽  
Source Of Information

There is a great concern about the security of computer these days. The number of attacks has increased in a great number in the last few years, intrusion detection is the main source of information assurance. While firewalls can provide some protection, they fail to provide protection fully and they even need to be complemented with an intrusion detection system (IDS). A newer approach for Intrusion detection is data mining techniques.IDS system can be developed using individual algorithms like neural networks, clustering, classification, etc. The result of these systems is good detection rate and low false alarm rate. According to a recent study, cascading of multiple algorithms gives a way better performance than single algorithm. Single algorithm systems have a high alarm rate. Therefore, to solve this problem, a combination of different algorithms are required. In this research paper, we use the hybrid algorithm for developing the intrusion detection system. C4.5 Support Vector Machine (SVM) and Decision Tree combined to achieve high accuracy and diminish the false alarm rate. Intrusions can be classified into types like Normal, DOS, R2L and U2R.Intrusion detection with Decision trees and SVM were tested with benchmark standard NSL- KDD, which is the extended version of  KDD Cup 1999 for intrusion detection (ID).

Proposed Hybrid Classifier to Improve Network Intrusion Detection System using Data Mining Techniques

2020 ◽  
Vol 38 (1B) ◽  
pp. 6-14
Author(s):  
ٍٍSarah M. Shareef ◽  
Soukaena H. Hashim
Keyword(s):  
Intrusion Detection ◽  
False Alarm ◽  
Intrusion Detection System ◽  
Detection System ◽  
Multinomial Logistic Regression ◽  
Network Intrusion Detection ◽  
Limiting Factor ◽  
Network Intrusion ◽  
Network Intrusion Detection System ◽  
Normal Behavior

Network intrusion detection system (NIDS) is a software system which plays an important role to protect network system and can be used to monitor network activities to detect different kinds of attacks from normal behavior in network traffics. A false alarm is one of the most identified problems in relation to the intrusion detection system which can be a limiting factor for the performance and accuracy of the intrusion detection system. The proposed system involves mining techniques at two sequential levels, which are: at the first level Naïve Bayes algorithm is used to detect abnormal activity from normal behavior. The second level is the multinomial logistic regression algorithm of which is used to classify abnormal activity into main four attack types in addition to a normal class. To evaluate the proposed system, the KDDCUP99 dataset of the intrusion detection system was used and K-fold cross-validation was performed. The experimental results show that the performance of the proposed system is improved with less false alarm rate.

scholarly journals A determinant fuzzy apparoach for intrusion detection system

2018 ◽  
Vol 7 (1.9) ◽  
pp. 245
Author(s):  
S. Vimala ◽  
V. Khanna ◽  
C. Nalini
Keyword(s):  
Data Collection ◽  
Intrusion Detection ◽  
False Alarm ◽  
False Alarm Rate ◽  
Great Degree ◽  
Intrusion Detection System ◽  
Detection System ◽  
Information Mining ◽  
The Past ◽  
Recognition Systems

In MANETs, versatile hubs can impart transparently to each other without the need of predefined framework. Interruption location framework is a fundamental bit of security for MANETs. It is uncommonly convincing for identifying the Intrusions and for the most part used to supplement for other security segment. That is the reason Intrusion discovery framework (IDS) is known as the second mass of assurance for any survivable framework security. The proposed fluffy based IDSs for recognition of Intrusions in MANETs are not prepared to adjust up all sort of assaults. We have examined that all proposed fluffy based IDSs are seen as to a great degree obliged segments or qualities for data collection which is specific for a particular assault. So that these IDSs are simply recognize the particular assault in MANETs. The fluffy motor may perceive blockage from channel mistake conditions, and along these lines helps the TCP blunder discovery. Examination has been made on the issues for upgrading the steady quality and precision of the decisions in MANET. This approach offers a strategy for joining remote units' estimation comes to fruition with alliance information open or priori decided at conglomerating hubs. In our investigation work, the best need was to reduce the measure of information required for getting ready and the false alarm rate. We are chiefly endeavoring to improve the execution of a present framework rather than endeavoring to supplant current Intrusion recognition systems with an information mining approach. While current mark based Intrusion identification procedures have imperatives as communicated in the past region, they do even now give basic organizations and this normal us to choose how information mining could be used as a piece of a correlative way to deal with existing measures and improves it.

scholarly journals An Effective System of Intrusion Detection on Deep Neural Network by Hybrid Optimization in Cyber Security

2019 ◽  
Vol 9 (1) ◽  
pp. 1320-1327
Keyword(s):  
Intrusion Detection ◽  
False Alarm ◽  
False Alarm Rate ◽  
Search Algorithm ◽  
Detection System ◽  
Optimization Techniques ◽  
Hybrid Optimization ◽  
Experimental Result ◽  
Gray Wolf ◽  
High Detection

In present trends organizations are very much interested to protect data and prevent malware attack by using well flourished and excellent tools. Many algorithms are used for the intrusion detection system (IDS) and it has pros and cons. Here we proposed a novel method of intrusion detection using hybrid optimization techniques such as Gravity search algorithm with gray wolf optimization (GSGW). In this method the gray wolf technique has a leader for the continuous monitoring of the attacker and has a low false alarm rate and a high detection rate. The performance evaluation is done by the feature selection in NSL-KDD dataset. In the proposed method the experimental result reveals less false alarm rate, better accuracy and high Detection when compared to previous analysis.

scholarly journals The Use of Ensemble Models for Multiple Class and Binary Class Classification for Improving Intrusion Detection Systems

Sensors ◽  
2020 ◽  
Vol 20 (9) ◽  
pp. 2559 ◽  
Author(s):  
Celestine Iwendi ◽  
Suleman Khan ◽  
Joseph Henry Anajemba ◽  
Mohit Mittal ◽  
Mamdouh Alenezi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
False Alarm ◽  
False Alarm Rate ◽  
Detection Rate ◽  
Detection System ◽  
Denial Of Service ◽  
Intrusion Detection Systems ◽  
Ensemble Models ◽  
Detection Systems

The pursuit to spot abnormal behaviors in and out of a network system is what led to a system known as intrusion detection systems for soft computing besides many researchers have applied machine learning around this area. Obviously, a single classifier alone in the classifications seems impossible to control network intruders. This limitation is what led us to perform dimensionality reduction by means of correlation-based feature selection approach (CFS approach) in addition to a refined ensemble model. The paper aims to improve the Intrusion Detection System (IDS) by proposing a CFS + Ensemble Classifiers (Bagging and Adaboost) which has high accuracy, high packet detection rate, and low false alarm rate. Machine Learning Ensemble Models with base classifiers (J48, Random Forest, and Reptree) were built. Binary classification, as well as Multiclass classification for KDD99 and NSLKDD datasets, was done while all the attacks were named as an anomaly and normal traffic. Class labels consisted of five major attacks, namely Denial of Service (DoS), Probe, User-to-Root (U2R), Root to Local attacks (R2L), and Normal class attacks. Results from the experiment showed that our proposed model produces 0 false alarm rate (FAR) and 99.90% detection rate (DR) for the KDD99 dataset, and 0.5% FAR and 98.60% DR for NSLKDD dataset when working with 6 and 13 selected features.

Sign in / Sign up

Export Citation Format

Share Document