An Overview of the Current Classification Techniques in Intrusion Detection

Size: px
Start display at page:

Download "An Overview of the Current Classification Techniques in Intrusion Detection"

Transcription

1 82 Int'l Conf. Security and Management SAM'15 An Overview of the Current Classification Techniques in Intrusion Detection Buthina Al-Dhafian, Iftikhar Ahmad, Abdullah Al-Ghamid Software Engineering Department, King Saud University Riyadh, Saudi Arabia Abstract During the last decade, a lot of attention has been given to intrusion detection systems (IDSs) as another security tools used to detect attacks and make working in computer systems and network more efficient and stable. However, the current challenge in these systems is consider into which is an optimal classification technique that must be used to detect intrusion in high level of accuracy. Many classification techniques have been designed in IDSs to detect attacks, where the accuracy of IDSs depends mainly on them. Numbers of studies have been proposed to enhance the performance of IDSs by increased the detection rates (DR) and decreased the false alarms rats (FAR). In this paper, we present a review of the current classification techniques that are used during designing IDSs. We also provide a review of the current dataset that are used to train and test selected classifier. The main goal of our research is to provide a review of the current classification techniques in intrusion detection in order to enhance the performance of classifier by highlighting different issues, which need to be solved. This paper seeks to help the researchers to develop an optimal classification technique by eliminating the issues that reduces the accuracy of IDSs. Keywords Intrusion detection; Intrusion detection systems; Classification techniques; Datasets I. INTRODUCTION Nowadays, the basic design of security in computer systems and network has been changed due to the huge number of attacks, which appeared because the increased in the numbers of internet users. Despite the existing tools of security systems, which protect these attacks such as firewalls, antivirus, and data encryption, it still hard to ensur that computer systems and network will be free of security flaws. IDSs are emerged as another technique which have increased the tools of security systems to monitor, identify, and detect intrusions in high level of accurecy [1]. The machanism of IDSs depends on the observation to classfiy data into normal or abnoraml behavoir. When an intrusion or suspect pattern is observed, an alarm is activated in order to take measures to maintain the integrity of the system [2]. Many IDSs have been designed to detect intrusions, although, maintaining the accuracy considered the main issue, which is mainly depends on optimal classifier selection [3]. Find the optimal classification techniques in IDSs considered as a critical issue as well, such that each classifier required to be trained with sample data in order to recognize the patterns and then it is tested with other samples in order to be efficient enough to perform well. Moreover, select the appropriate datasets, which are used in the testing and the training process considered as a dilemma [4], which is used to evaluate IDSs. Therefore, a number of studies have been proposed using multiple of classification techniques in order to design IDSs with high level of accuracy. The main goal of our research is to present a review of the current studies that aim to improve classification in IDSs by increasing the DR and decreasing the FA using optimal classifier technique. Moreover, to highlight different issues that need to be solved during develop an optimal classification technique in IDSs. Besides this introductory section, the remaining of this paper is organized as follow. A background study in IDSs is given in section II. A review of current classification techniques in IDSs is presented in section III. An overview of the most popular standard datasets used in IDSs are explore in section IV. A comparative analysis is discussed in section V. Finally, conclusion is drown in section VI. II. BACKGROUND An intrusion can be described as any event that violates systems security, occurs by an intruder. An intrusion detection (ID) considered as one of the security systems tools, which is used to detect intrusion in computer systems and network based on the hypothesis that the behavior of an intruder and a legitimate user vary from each other. IDSs have appeared to deal with vulnerabilities of systems security, where they are designed as complementary rather than alternative tools to these systems. The first concept of ID appeared early 1980 [5], which emphasized on a single computer system followed by the actual work done by Denning in 1987 [6] at SRI International, where intrusion detection is extended to address multiple computers in a distributed system. Few years later, many IDSs ware proposed to serve on both sides of researches and commercial world. However, they have operated based on general architectural framework [1], [2]. The main component of this architecture is a detector (also known as analysis engine), which is responsible for classifying data to normal or abnormal behavior, and considered the basis to determine the accuracy of IDSs. IDSs work to identify the suspect behavior based on three types of detection methodologies [2], [7]: (i) Misuse-based detection (MD), which is used mainly to detect intrusions according to the predefined pattern of known attacks, the accuracy of this type considered good and theoretically, it has a very low FAR, however it cannot detect new attacks. (ii) Anomaly-based detection (AD), where detect intrusions based on a reference model of the normal behavior of the monitored

2 Int'l Conf. Security and Management SAM'15 83 system, it has the capability to detect unknown attacks, but the FAR it very high. (iii) Stateful protocol analysis (SPA), which detects intrusion based on predetermined profiles of accepted definitions of normal protocol activity for each protocol state, it differs from AD as it depends on vendor-developed universal profiles that determine how particular protocols work. These types of detection methodologies can perform separately or integrated into one system, known as hybrid IDS [8], which constructed to avail from multiple approaches and overcame many of the issues by producing a much stronger IDS. On the other hand, IDSs have many types of technologies, which can be categorized based on the scope of detection. These categories are [7], [9], [10]: (i) Host-based IDSs which analyze the activities that flow into the host to identify attacks, (ii) Network-based IDSs which analyze network packets that come from the outside to detect attacks, (iii) Application-based IDSs a partial set of HIDS, responsible for monitoring and analyzing the activities that have took place inside a software application. (iv) Wireless IDSs [11] which monitor and analyze the protocols of wireless networks to identify shady activities, and (v) Network Behavior Analysis (NBA) IDSs which inspect network traffic or statistics on network traffic to identify suspicious behavire. Some of these technologies can be adopted with each other to be known as Mixed IDS (MIDS) with a view to improve DR and make systems as much as possible free from attacks. III. CURRENT CLASSIFICATION TECHNIQUES IN INTRUSION DETECTION In literature, numerous of studies have applied different classification techniques to design IDSs; some studies have deigned IDSs by using single techniques (such as neural network, fuzzy techniques, support vector machines, etc...), and the other hand, some studies have deigned IDSs based on combining different techniques (such as hybrid or ensemble techniques). A brief overview of the current classification techniques in IDSs is listed below: A. Approach-1 Tong et al. [8] have been proposed a hybrid RBF/Elman neural network model to be used for both AD and MD, which can efficiently detect temporally dispersed and collaborative attacks. They used a radial basis function (RBF) network as a real-time pattern classification and they applied the Elman network to restore the memory of past events. Their model takes an output of RBF as input of Elman network, while an Elman network restore each output of RBF network by keep memory of past misuse events. For their experiments, they have used DARBA dataset. The results showed that, their model can detect intrusions with higher RD and lower FPR compared with other IDSs that used neural network techniques. Additionally, the ability to determined DOS and probing attacks in IDSs is enhanced. B. Approach-2 A new approach, which called FC-ANN have been proposed by Wang et al. [3] with the aim to enhance detecting precision for low-frequent attacks, detecting stability as well as achieving higher DR and lower FPR. Feed-Forward neural network (FFNN) and Fuzzy c-means clustering are used for designing their approach, which designed based on the following three phases. In the first phase, a fuzzy clustering technique is used to generate different training subsets to reduce the size and complexity, while in the second phase different ANNs are trained based on different training sets. Finally, in the last phase, a meta-learner, fuzzy aggregation module, is introduced to learn and combine the different ANN s results in order to eliminate the errors of different ANNs. For their experiment, CUP dataset were used for the evaluation purpose. The results have demonstrated that, their approach has the effectiveness especially for low-frequent attacks, i.e., R2L and U2R attacks in terms of detection precision and detection stability. C. Approach-3 A -based intrusion detection system has been proposed by Horng et al. [12] with the aim of shorten the training time as well as to improve the performance of classifier. Their approach combines three methods, which are a hierarchical clustering algorithm, a simple feature selection procedure, and the technique. dataset is used for their experiments for evaluation purpose. First, they used BIRCH clustering algorithm to transform the dataset to a smaller sized dataset. Then, they trained classifiers based on the reduced dataset with abstract data points. Finally, they used leave-one-out procedure in order to remove unimportant features from training set. The results have showed that the proposed system had the best performance to detect intrusion, practically; it had superior performance in the detection DoS and Probe attacks. Approach-4 Ahmad et al. [13] have been proposed an intrusion detection system in order to overcome performance issues by using feature subset selection based on multilayer Perceptron (MLP). In their approach, Principle Components Analysis (PCA) are used for features transformation, while Genetic Algorithms (GA) are applied for search the principal feature space for a subset of features. For classification purpose, they used MLP. cup was used for their experiments. Their approach was an initial effort for features subset selection in intrusion detection, which aims to find a subset of principle components by using GA to search the PCA space. The results showed that, the proposed approach have improved accuracy, simplified the architecture of intrusion detection, as well as decreased training time and computational overheads. D. Approach-5 Chitrakar and Chuanhe [14] have proposed a new hybrid approach, which aims to make the classification operate in anomaly-based IDSs more accurate and efficient. They used K- Medoids Clustering approach to gathering similar data instances based on their behavior and applied to classify data to normal or abnormal behavior. Their experiments have been evaluated based on Kyoto2006+ datasets. The main procedure in their approach is converted the attribute of datasets into suitable types, and then normalized

3 84 Int'l Conf. Security and Management SAM'15 them based on the need of the kernel. The entire selected data are classified into [-1, 1], where -1 represents each sample with both known and unknown attack and 1 represents normal class. The experimental result referred to that, their approach has increased DR as well as decreased FPR in in superior level compared to the other hybrid approach. E. Approach-6 A hybrid intrusion detection system for MD and AD have been proposed by Om and Kundu [15], which combines K- Means and two classifiers K-Nearest Neighbor (K-NN) and Naïve Bayes. - dataset are used for evaluation purpose. First, the entropy based feature selection algorithm is used to select the appropriate features. Then, k-means clustering algorithm is applied on the selected features to split the data records into normal and abnormal clusters. After that, the obtained data are classified into normal or abnormal clusters by using the hybrid classifier. The main goals in their approach were to reduce the FAR, detect the intrusions, and further classify them into four categories: DoS, U2R, R2L, and probe. As a result, they have found that the proposed approach is better than the other conventional approaches such as kmeans, knn, and Naïve Bayes in terms of accuracy, DR, and FAR. F. Approach-7 An optimized intrusion detection using soft computing techniques has been proposed by Ahmad et al. [16], with the aim to provide an optimal intrusion detection system that has ability to minimize amount of features and maximize DR. In their approach, dataset is used for evaluation purpose and PCA is applied to convert the input samples into a new feature space. Moreover, GA is used to find a suitable number of principal components, and for classification purpose, they have applied. They have focused on comparing performance on feature sets. First, they obtained 12 features from PCA and GA and classified them with. Second, they collected 22 features directly from PCA output using the traditional method and classified them with. The experimental results referred to that, the proposed method has provided an optimal intrusion detection, which is able to minimize amount of features and maximize the DR. G. Approach-8 Kim et al. [17] have proposed a new hybrid intrusion detection which integrating hierarchically MD model and AD model to overcome performance issue. In their proposed approach, the MD has used the information of known attack to build a classifier while the AD has used information of normal traffic to build a classifier. First, the MD model is decomposed normal training data into disjoint subsets. Then MD model is applied for each separate subset of normal training data. The techniques that used for each model were C4.5 decision tree (DT) for MD model, and 1-class to construct multiple AD models. For their experiments, they have used NSL- data set. It has been found that the result of the proposed method is superior to the conventional methods with respect to performance for detecting unknown attacks, training and testing time. H. Approach-9 Ahmad et al. [18] have proposed a novel method in intrusion detection to enhance the performance of the classifier, where PCA is applied for feature transformation. Moreover, GA is used to find the genetic principle components, which offer a subset of features with optimal sensitivity and the highest discriminatory power. For classification purpose, they have applied, where - dataset is used for evaluation purpose. Their work has extended the previous work [16], and the results have showed that, the proposed method has enhanced the performance of classifier in intrusion detection by minimizing the number of features (up to 10) and maximizing the DR (up to %). I. Approach-10 Chaunhan et al. [19] have presented a comparison between different classification techniques, which are worked to detect intrusions and classify them into normal and abnormal behaviors. The algorithms that have been selected are J48, Naive Bayes, RIPPER (JRip), and One Rule (OneR). Their experiments were performed by using NSL- dataset. WEKA platform was selected for the implementation of the selected algorithms. The results have showed that the best algorithm for classification purpose is OneR classifier, where it required the shortest time, which is around 0.45 s with 10-fold cross-validation, and 0.32 s with supplied test set compared with others classifiers. IV. STANDARD DATASETS One of the most important parameters which can affect the capability of the intrusion detection mechanism is dataset, where the performance of IDSs depends on its accuracy and vice versa. When the training dataset is optimally accurate with a rich content then, the efficiency of the trained system is improved. Thus, the collection of the data in order to train and test the different classfication techniques is a critical dilemma. There are three different methods for collecting data to be used for experiments in the IDSs, which are [4]: (i) real traffic, (ii) sanitized traffic, and (iii) simulated traffic. However, these methods still inefficient for training and testing classification techniques, where using the real traffic to collecting data can be very costly, sanitized traffic is more risky, while generated simulation traffic required a hard work that can make the standard constructed simulated datasets popular for evaluating IDSs. There is a number of standard datasets, which can be classified based on the network traffic such as DARBA [20], - [21], NSL- [22], CAIDA [23], and Kyoto2006+ [24]. Accordingly, these data can be used for experiments in order to evaluate classification techniques in the field of IDSs. A. DARBA Dataset DARPA dataset [20], [25] is the first standard corpora for evaluating computer network IDSs, which has been collected and distributed by MIT Lincoln Laboratory. Each evaluation effort built to measure the possibility of detectiton and FA for each system under test using many types of attacks. DARPA dataset was collected by set up a test bed that simulated the operation of a typical US Air Force LAN for over two months

4 Int'l Conf. Security and Management SAM'15 85 to structure audit data to be used for evaluating algorithms in IDSs. B. CUP Dataset CUP dataset [21] is a connection of data transfer collected from a virtual environment to be used for the Competition of the Third International Knowledge Discovery and Data Mining Tools. This standard dataset is gathered by Stolfo et al. [26], based on the pre-processing version of data built in DARPA 1998 [20]. Each connection record is about 100 bytes, consists of 41 features, and labeled as normal or as an attack. CUP dataset separated into two sets, which are training set and testing set. C. NSL- Dataset NSL- dataset appeared to fix the issues occurred in CUP dataset, which has highly affected the performance of the evaluated systems. It proposed by Tavallaee et al. [27] as a new revised version of CUP dataset, and it publicly available online on [22]. The main contribution of NSL- dataset, it does not include redundant records whether in train or test sets. D. CAIDA Datasets CAIDA datasets [23] are a collection of several different types of data, resulting from both active and passive measurement of the internet. These datasets are available to the research community with retention the privacy of individuals and organizations who donate data or network access. Established in 1997 by Dr. Kc Claffy and Tracie Monk, and located in San Diego Supercomputing Center (SDSC). The data collection for each dataset is still active and has continuing, regularly scheduled groups, or terminated and will not be resumed. E. Kyoto Dataset Kyoto dataset [24], [28] is a connection of data transfer collected from honeypots and darknets data published by Kyoto University. It appeared to fill the gap in the existing evaluation datasets, such as CUP dataset. Kyoto2006+ dataset built from diverse types of honeypots over three years of real traffic data from 2006 until It consists of 14 statistical features, which are derived from CUP dataset ignoring other features that contain redundant. As well, it includes additional 10 features for more analysis and evaluation of NIDSs. V. COMPARATIVE ANALYSIS A. Datasets Comparison Table I presents a comparison between five types of standard datasets, which are mentioned in section V. DARPA dataset considered as a popular dataset used in IDSs to measure DR and FA for any network traffic, which consists of four types of attacks (DoS, R2L, U2R, and Probing attacks). However, it faced a set of critiques [29], where it appeared early in 1998 and, using very simple models to create background traffic, and the synthesized data it does not look like to be similar background traffic in real networks. Moreover, traffic collectors that used to collect data from network traffic (i.e. TCPdump) are extremely probable ignore packet during intensive traffic load [27]. CUP dataset also appeared early in as a preprocessing version for data in DARPA 1998 dataset, which classified records into 41 features that are not related to any critiques to DARBA dataset. In spite of CUP dataset including a huge number of attacks where the attack types in training set are not the same in testing set, it includes redundant and duplicate record, which cause overhead during the evaluation process [27]. NSL- dataset also appeared as a new version from dataset, which has removed redundant or duplicate records in - dataset and represent the records in way that is more reasonable. However, it does not considered as the ideal way for representing the existing real networks [28]. The issue is not just exclusive on NSL- dataset but also includes old version from DARBA 1998 and datasets. CAIDA datasets also appeared for evaluating IDSs, which consists of different types of datasets that are collected from the internet, which considered as a perfect resource for representing the real existing networks. Kyoto dataset also appeared for evaluating NIDSs, and it is built by through ignoring features that contain redundant, as the previous mentioned datasets, this dataset is also the comprehensive representation for the real current networks. Although it is recently emerged, it does not mention information on particular attack types. Standards datasets are not limited to these five types of evaluation. VELOS dataset [30] also appeared to evaluate the performance of IDSs, which includes approximately 10 gigabytes of normal and malicious traffic with nearly different kinds of the potential attacks, mainly web access attacks. This traffic is primarily in several pcap files and tcpdump text files. B. Approaches Comparsion Table II presents a comparison for the approaches that were proposed and applied for intrusion detection, which are mentioned in section IV. It provides details regarding the techniques that are combined with a view to strengthen the performance of the IDSs. Moreover, it highlights different issues that need to be solved. In the hybrid approach for Tong et al. [8], despite the ability of using, MPM, softcomputing, and other pattern classification technology as a pattern classification module instead of RBF neural network, it considered not suitable for classification. Results have showed its weakness compared with the other hybrid approaches. This can be due to used full features, which caused redundant and introducing overheads. Moreover, the raw feature set can confounded the classifier due to the redundancy and results FA. Further, it increases training and testing overheads, reduces accurate DR, consumes more memory and computational resources, increases architectural complexity and malfunctioning the system. The same issue appeared in [3] where they used full features, which caused overhead, and found that redundancy have reduced accuracy of DR, as well as it considered a time consuming for the training process. Moreover, they have found that the used classifier it often converges to the local minima, which considered unsettled, it also considered unsuitable for nonlinearly separable problem, and slow for overtraining. Despite using clustering approach for

5 86 Int'l Conf. Security and Management SAM'15 Dataset DARPA Release Date 1998 Prepared Data MIT Lincoln Laboratory CUP Stolfo et al. NSL-CUP 2009 CAIDA 1998 Kyoto Mahbod et al. Dr. Kc Claffy and Tracie Monk Kyoto University TABLE I: STANDARDS s DATASETS COMPARISON IN IDS Data Source Features Availability Pros Cons Simulated the operation of a typical US Air Force LAN Preprocessing version for data in DARPA 1998 dataset An improved version of CUP dataset Depends on the dataset that is captured Real traffic data in honeypot and darknets - 41 features (basic features, content features, and traffic features): 32 numeric features, 9 categorical features 41 features (basic features, content features, and traffic features): 32 numeric features, 9 categorical features Depends on the dataset that is captured 24 features: 14 statistical features derived from CUP, 10 additional features First standard corpora for evaluating IDSs Consists of a broad range of attacks ( DoS, Probing, R2L, U2R) Measure DR and FA for any network traffic Used for evaluating AD systems Attacks types in training set are distinctive from the testing set. Dose not includes redundant or duplicate records The selected records of NSL- dataset are inversely proportional to the percentage of records in the CUP dataset The number of the records is reasonable in the training set and testing set. Collect different types of data from the online available sources Retention the privacy of individuals who donate the data Perfect in representing the existing real networks Used for evaluating NIDSs Ignored features that contain redundant Perfect in representing the existing real networks Models that used to generate background traffic were too simple [29] Workload of the synthesized data does not simulate the background traffic in real networks [29] TCPdump can cause overloaded and drop packets [27] Includes redundant and or duplicate records [27] Does not reflect the modern environment [28] Not perfect for representing the existing real networks [28] Some datasets are restricted by permission Does not mention information on particular attack types performance, still determine the suitable number of clustering remains an open problem. Although Horng et al. [12] have used leave-one-out procedure to ignore irrelevant features from dataset, it is much complicated and overheads. Despite the better results of K-NN and Naïve Bayes compared to the conventional as kmeans, knn, and Naïve Bayes in terms of accuracy, DR, and FAR., K-NN and Naïve Bayes can be further explored with other feature selection techniques, where the algorithm that is used is characterized by simplicity for dealing with redundant and irrelevant records in - dataset. The hybrid approach that was proposed by Chitrakar and Chuanhe [14] also showed a better accurate result for DR and FPR. It used Kyoto dataset for evaluation purpose, where it ignored features that contain redundant. However, classifier that was used in their approach can be more efficient and stable when applying multiple kernel based classification schemes. Additionally, the time complexity of k- Medoids clustering still needs to be decreased. On the other side, the approaches that presented by Ahmad et al. [13], [16], [18], which sought to improve the performance issues in IDSs, provided the better results for DR and FA. In [13], their proposed approach was an initial effort for features subset selection, which presented in order to override feature selection issues. Their experiments showed that, the proposed approach have improved accuracy, simplified the architecture of intrusion detection, as well as decreased training time and computational overheads. However, MLP classifier not adequately explained with more experiments, such as examining the ability of the classifier to execute well on the original dataset, or execute well on transformed dataset; this can be due to the problem for local minimal and required overtraining. Therefore, their method needs more experimentation to verify it. In [16], Ahmad et al. have used classifier for their method with the aim to provide an optimal intrusion detection system that has ability to minimize amount of features and maximize DR. However, their approach required more experimentation, which can verify it. Although Ahmed et al. [18] have provided a new approach to enhance the performance of classifier, and it results have showed that the proposed approach enhanced the performance of classifier in intrusion detection in terms of minimizing the number of features and maximizing the DR, classifier is not suitable for multiclass. Moreover, their work also needs more experimentation to verify it. Furthermore, despite the better results that have enhanced the detection accuracy which were resulted from NSL- dataset, in [17], decompose the normal data using C4.5 DT degrades the misuse detection performance, and unequal allocation of data instances hinders the reduction of the training and testing time. Moreover, in [19], the results of accuracy, and time complexity, can be enhanced

6 Int'l Conf. Security and Management SAM'15 87 TABLE II: APPROACHES COMPARISON IN IDS Authors Tong et al. [7] Wang et al. [3] Horng et al. [10] Ahmad et al. [11] Chitrakar and Chuanhe [12] Om and Kundu [13] Ahmad et al. [14] Kim et al. [15] Ahmad et al. [16] Chaunhan et al. [17] Year 2009 Detection Patterns Preprocessing Techniques 2010 AD AD 2012 AD Classification Techniques - RBF NN leave-one-out procedure PCA + GA Sampling and filtering Entropy based feature selection algorithm PCA + GA Feed-Forward NN MLP K-NN + Naïve Bayes - C4.5 DT + PCA + GA 2014 AD - Data source DARBA Kyoto NSL- Accuracy % DR % FP % < >99.0 < Issues/Problems Used full features that leading to redundancy and results FA, Increases training and testing overheads, Reduces accurate DR, consumes more memory and computational resources, Increases architectural complexity and malfunction of the system Used full features, which caused overhead and redundant, and required more time for training Identifying the suitable number of clustering still an open issue The used classifier it often converges to the local minima, which considered unsettled, unsuitable for nonlinearly separable problem, and slow for overtraining More complicated and had overheads on massive dataset due to the methods that are used to determine the significant feature. with other feature selection techniques MLP classifier not adequately explained with more experiments such as, its ability of executing it on original dataset, or its ability on executing along with a transforming dataset, due to the problem for minimal and overtraining. The approach needs more experimentation to verify it classifier can be more efficient and stable when applying multiple kernel based classification schemes Increased time complexity through applying every data samples, one by one, during the implemented process Hybrid Classifiers that used can be further explored with other feature selection techniques, where the algorithm that was used is very simple to deal with redundant and irrelevant records in - dataset. The approach required more experimentation to verify it with other feature selection techniques Decompose the normal data using C4.5 DT degrades the misuse detection performance. Unequal allocation of data instances hinders the reduction of the training and testing time with other feature selection techniques The approach needs more experimentation to verify it with other feature selection techniques, classifier being inappropriate for multiclass J The results can be enhanced and the timing Naïve Bayes for the training and the testing can be reduced NSL by ignoring redundant and irrelevant features JRip through using optimal technique for feature OneR selections by ignoring redundant and irrelevant features using optimal technique for feature selections. Therefore, the accuracy of IDSs depends on an optimum classification technique, which is mainly depends on optimal dataset selection. Many techniques for feature selection were applied on dataset with a view to enhance the accuracy of the classifier. A numbers of studies have been presented in order to enhance the performance of IDSs by increasing DR and decreasing FPR. Thus, finding the optimal classification technique to avoid current issues in the recent techniques, choosing suitable dataset, which includes rich types of recent attacks, and selecting a suitable features are the current important issues in the field of IDSs. VI. CONCLUSION Find the optimal classification technique to enhance the performance of IDSs by increasing the DR and decreasing the FA is still an ongoing area. In this paper, a review of the current classification techniques in IDSs is introduced. Moreover, a review of the most popular datasets used for train and test

7 88 Int'l Conf. Security and Management SAM'15 selected classifier is also introduced. Each classification technique has its superiority and limitations during classify data into normal or abnormal, so that it is important to select an optimal one during intrusion detection process. Additionally, we have also discussed these reviews to conclude that, a set of issues must be take into consideration during development of classification techniques in IDSs, such as which is an optimal dataset that includes a rich types of recent attacks, and which features that must be selected without confused, overhead, and time-consuming selected classifier. ACKNOWLEDGMENT This research work is supported by Department of Software Engineering, CCIS, King Saud University, Riyadh, Saudi Arabia. REFERENCES [1] T. Verwoerd and R. Hunt, "Intrusion Detection Techniques and Approaches," Computer Communications, vol. 25, no. 15, pp , September [2] A. Lazarevic, V. Kumar and J. Srivastava, "Intrusion Detection: A Survey," in Managing Cyber Threats, vol. 5, Springer US, 2005, pp [3] G. Wang, J. Hao, J. Ma and L. Huang, "A new approach to intrusion detection using Artificial Neural Networks and fuzzy clustering," Expert Systems with Applications, vol. 37, no. 9, pp , September [4] I. Ahmad, A. Abdullah and A. Alghamdi, "Artificial neural network approaches to intrusion detection: a review," in Proceedings of the 8th International Conference on the World Scientific and Engineering Academy and Society, Istanbul, Turkey, [5] J. P. Anderson, "Computer security threat monitoring and surveillance," Fort Washington, Pennsylvania, April, [6] D. E. Dorothy, "An Intrusion-Detection Model," Software Engineering, IEEE Transactions on, vol. 13, no. 2, pp , February [7] H.-J. Liao, C.-H. R. Lin, Y.-C. Lin and K.-Y. Tung, "Intrusion Detection System: A Comprehensive Review," Journal of Network and Computer Applications, vol. 36, no. 1, pp , 1 January [8] X. Tong, Z. Wang and H. Yu, "A research using hybrid RBF/Elman neural networks for intrusion detection system secure model," Computer Physics Communications, vol. 180, no. 10, pp , Octobar [9] Y. Bai and K. Hidetsune, "Intrusion Detection Systems: technology and development," in Proceedings of the 17th International Conference on Advanced Information Networking and Applications, March, [10] K. Scarfone and P. Mell, "Guide to Intrusion Detection," National Institute of Standards and Technology, February, [11] R. Mirchell and I.-R. Chen, "A Survey of Intrusion Detection in Wireless Network Application," Computer Communications, vol. 42, pp. 1-23, 1 February [12] S.-J. Horng, M.-Y. Su, Y.-H. Chen, T.-W. Kao, R.-J. Chen,. J.-L. Lai and C. D. Perkasa, "A novel intrusion detection system based on hierarchical clustering and support vector machines," Expert Systems with Applications, vol. 38, no. 1, pp , January [13] I. Ahmad, A. Abdullah, A. Algamdi, K. Alnfajan and M. Hussain, "Intrusion detection using feature subset selection based on MLP," Scientific Research and Essays, vol. 6, no. 34, pp , December [14] C. Huang and R. Chitrakar, "Anomaly detection using Support Vector Machine classification with k-medoids clustering," in Proceedings of the 3rd International Conference on Asian Himalayas, Kathmandu. Nepal, November, [15] H. Om and A. Kundu, "A hybrid system for reducing the false alarm rate of anomaly intrusion detection system," in Proceedings of the 1st International Conference on Recent Advances in Information Technology (RAIT), March, [16] I. Ahmad, A. Abdullah and A. Alghamdi, "Optimized intrusion detection mechanism using soft computing techniques," Telecommunication Systems, vol. 52, no. 4, pp , April [17] G. Kim, S. Lee and S. Kim, "A novel hybrid intrusion detection method integrating anomaly detection with misuse detection," Expert Systems with Applications, vol. 41, no. 4, pp , March [18] I. Ahmad, M. Hussain, A. Alghamd and A. Alelaiwi, "Enhancing performance in intrusion detection using optimal feature subset selection based on genetic principal components," Neural Computing and Applications, vol. 24, no. 7-8, pp , June [19] H. Chauhan, V. Kuma, S. Pundir and E. S. Pilli, "Comparative Analysis and Research Issues in Classification Techniques for Intrusion Detection," in Proceedings of the International Conference on Advanced Computing, Networking, and Informatics,, India, [20] "DARPA Intrusion Detection Evaluation Program," MIT Lincoln Labs, [Online]. Available: val/index.html. [Accessed February 2015]. [21] " 1990 Data,". [Online]. Available: [Accessed February 2015]. [22] "NSL- Dataset," [Online]. Available: [Accessed February 2015]. [23] "CAIDA Dataset," CAIDA, [Online]. Available: [Accessed February 2015]. [24] "Kyoto2006+ Dataset," [Online]. Available: [Accessed February 2015]. [25] R. Lippmann, D. Fried, I. Graf, J. Haines, K. Kend, D. McClung, D. Weber, S. Webster, D. Wyschogrod, R. Cunningham and M. Zissman, "Evaluating intrusion detection systems: The 1998 DARPA off-line intrusion detection evaluation," in Proceedings of the International Conference on DARPA Information Survivability Conference and Exposition, [26] S. J. Stolfo, W. Fan, W. Lee, A. Prodromidis and P. K. Chan, "Costbased modeling for fraud and intrusion detection: results from the JAM project," in Proceedings of the International Conference on DARPA Information Survivability Conference and Exposition, [27] M. Tavallaee, E. Bagheri, W. Lu and A. A. Ghorbani, "A detailed analysis of the CUP 99 data set," in Proceedings of the Second IEEE international conference on Computational intelligence for security and defense applications, Ottawa, ON Canada, July, [28] H. Takakura, Y. Okabe, M. Eto, D. Inoue and K. Nakao, "Statistical analysis of honeypot data and building of Kyoto dataset for NIDS evaluation," in Proceedings of the First Workshop on Building Analysis Datasets and Gathering Experience Returns for Security, [29] J. McHugh, "Testing Intrusion detection systems: a critique of the 1998 and DARPA intrusion detection system evaluations as performed by Lincoln Laboratory," ACM Transactions on Information and System Security, vol. 3, no. 4, pp , November [30] " VELOS Dataset," [Online]. Available: [Accessed February 2015].

A survey on Data Mining based Intrusion Detection Systems

A survey on Data Mining based Intrusion Detection Systems International Journal of Computer Networks and Communications Security VOL. 2, NO. 12, DECEMBER 2014, 485 490 Available online at: www.ijcncs.org ISSN 2308-9830 A survey on Data Mining based Intrusion

More information

Efficient Security Alert Management System

Efficient Security Alert Management System Efficient Security Alert Management System Minoo Deljavan Anvary IT Department School of e-learning Shiraz University Shiraz, Fars, Iran Majid Ghonji Feshki Department of Computer Science Qzvin Branch,

More information

A Review of Anomaly Detection Techniques in Network Intrusion Detection System

A Review of Anomaly Detection Techniques in Network Intrusion Detection System A Review of Anomaly Detection Techniques in Network Intrusion Detection System Dr.D.V.S.S.Subrahmanyam Professor, Dept. of CSE, Sreyas Institute of Engineering & Technology, Hyderabad, India ABSTRACT:In

More information

Performance Evaluation of Intrusion Detection Systems using ANN

Performance Evaluation of Intrusion Detection Systems using ANN Performance Evaluation of Intrusion Detection Systems using ANN Khaled Ahmed Abood Omer 1, Fadwa Abdulbari Awn 2 1 Computer Science and Engineering Department, Faculty of Engineering, University of Aden,

More information

A new Approach for Intrusion Detection in Computer Networks Using Data Mining Technique

A new Approach for Intrusion Detection in Computer Networks Using Data Mining Technique A new Approach for Intrusion Detection in Computer Networks Using Data Mining Technique Aida Parbaleh 1, Dr. Heirsh Soltanpanah 2* 1 Department of Computer Engineering, Islamic Azad University, Sanandaj

More information

How To Prevent Network Attacks

How To Prevent Network Attacks Ali A. Ghorbani Wei Lu Mahbod Tavallaee Network Intrusion Detection and Prevention Concepts and Techniques )Spri inger Contents 1 Network Attacks 1 1.1 Attack Taxonomies 2 1.2 Probes 4 1.2.1 IPSweep and

More information

Detection. Perspective. Network Anomaly. Bhattacharyya. Jugal. A Machine Learning »C) Dhruba Kumar. Kumar KaKta. CRC Press J Taylor & Francis Croup

Detection. Perspective. Network Anomaly. Bhattacharyya. Jugal. A Machine Learning »C) Dhruba Kumar. Kumar KaKta. CRC Press J Taylor & Francis Croup Network Anomaly Detection A Machine Learning Perspective Dhruba Kumar Bhattacharyya Jugal Kumar KaKta»C) CRC Press J Taylor & Francis Croup Boca Raton London New York CRC Press is an imprint of the Taylor

More information

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015

International Journal of Computer Science Trends and Technology (IJCST) Volume 3 Issue 3, May-June 2015 RESEARCH ARTICLE OPEN ACCESS Data Mining Technology for Efficient Network Security Management Ankit Naik [1], S.W. Ahmad [2] Student [1], Assistant Professor [2] Department of Computer Science and Engineering

More information

A Dynamic Flooding Attack Detection System Based on Different Classification Techniques and Using SNMP MIB Data

A Dynamic Flooding Attack Detection System Based on Different Classification Techniques and Using SNMP MIB Data International Journal of Computer Networks and Communications Security VOL. 2, NO. 9, SEPTEMBER 2014, 279 284 Available online at: www.ijcncs.org ISSN 2308-9830 C N C S A Dynamic Flooding Attack Detection

More information

A SURVEY ON GENETIC ALGORITHM FOR INTRUSION DETECTION SYSTEM

A SURVEY ON GENETIC ALGORITHM FOR INTRUSION DETECTION SYSTEM A SURVEY ON GENETIC ALGORITHM FOR INTRUSION DETECTION SYSTEM MS. DIMPI K PATEL Department of Computer Science and Engineering, Hasmukh Goswami college of Engineering, Ahmedabad, Gujarat ABSTRACT The Internet

More information

Taxonomy of Intrusion Detection System

Taxonomy of Intrusion Detection System Taxonomy of Intrusion Detection System Monika Sharma, Sumit Sharma Abstract During the past years, security of computer networks has become main stream in most of everyone's lives. Nowadays as the use

More information

A SYSTEM FOR DENIAL OF SERVICE ATTACK DETECTION BASED ON MULTIVARIATE CORRELATION ANALYSIS

A SYSTEM FOR DENIAL OF SERVICE ATTACK DETECTION BASED ON MULTIVARIATE CORRELATION ANALYSIS Journal homepage: www.mjret.in ISSN:2348-6953 A SYSTEM FOR DENIAL OF SERVICE ATTACK DETECTION BASED ON MULTIVARIATE CORRELATION ANALYSIS P.V.Sawant 1, M.P.Sable 2, P.V.Kore 3, S.R.Bhosale 4 Department

More information

NETWORK INTRUSION DETECTION SYSTEM USING HYBRID CLASSIFICATION MODEL

NETWORK INTRUSION DETECTION SYSTEM USING HYBRID CLASSIFICATION MODEL NETWORK INTRUSION DETECTION SYSTEM USING HYBRID CLASSIFICATION MODEL Prof. Santosh T. Waghmode 1, Prof. Vinod S. Wadne 2 Department of Computer Engineering, 1, 2 JSPM s Imperial College of Engineering

More information

Network Intrusion Detection Using a HNB Binary Classifier

Network Intrusion Detection Using a HNB Binary Classifier 2015 17th UKSIM-AMSS International Conference on Modelling and Simulation Network Intrusion Detection Using a HNB Binary Classifier Levent Koc and Alan D. Carswell Center for Security Studies, University

More information

A Review on Network Intrusion Detection System Using Open Source Snort

A Review on Network Intrusion Detection System Using Open Source Snort , pp.61-70 http://dx.doi.org/10.14257/ijdta.2016.9.4.05 A Review on Network Intrusion Detection System Using Open Source Snort Sakshi Sharma and Manish Dixit Department of CSE& IT MITS Gwalior, India Sharmasakshi1009@gmail.com,

More information

Data Mining for Network Intrusion Detection

Data Mining for Network Intrusion Detection Data Mining for Network Intrusion Detection S Terry Brugger UC Davis Department of Computer Science Data Mining for Network Intrusion Detection p.1/55 Overview This is important for defense in depth Much

More information

An Efficient Way of Denial of Service Attack Detection Based on Triangle Map Generation

An Efficient Way of Denial of Service Attack Detection Based on Triangle Map Generation An Efficient Way of Denial of Service Attack Detection Based on Triangle Map Generation Shanofer. S Master of Engineering, Department of Computer Science and Engineering, Veerammal Engineering College,

More information

STUDY OF IMPLEMENTATION OF INTRUSION DETECTION SYSTEM (IDS) VIA DIFFERENT APPROACHS

STUDY OF IMPLEMENTATION OF INTRUSION DETECTION SYSTEM (IDS) VIA DIFFERENT APPROACHS STUDY OF IMPLEMENTATION OF INTRUSION DETECTION SYSTEM (IDS) VIA DIFFERENT APPROACHS SACHIN MALVIYA Student, Department of Information Technology, Medicaps Institute of Science & Technology, INDORE (M.P.)

More information

HYBRID INTRUSION DETECTION FOR CLUSTER BASED WIRELESS SENSOR NETWORK

HYBRID INTRUSION DETECTION FOR CLUSTER BASED WIRELESS SENSOR NETWORK HYBRID INTRUSION DETECTION FOR CLUSTER BASED WIRELESS SENSOR NETWORK 1 K.RANJITH SINGH 1 Dept. of Computer Science, Periyar University, TamilNadu, India 2 T.HEMA 2 Dept. of Computer Science, Periyar University,

More information

An Anomaly-Based Method for DDoS Attacks Detection using RBF Neural Networks

An Anomaly-Based Method for DDoS Attacks Detection using RBF Neural Networks 2011 International Conference on Network and Electronics Engineering IPCSIT vol.11 (2011) (2011) IACSIT Press, Singapore An Anomaly-Based Method for DDoS Attacks Detection using RBF Neural Networks Reyhaneh

More information

Intrusion Detection for Mobile Ad Hoc Networks

Intrusion Detection for Mobile Ad Hoc Networks Intrusion Detection for Mobile Ad Hoc Networks Tom Chen SMU, Dept of Electrical Engineering tchen@engr.smu.edu http://www.engr.smu.edu/~tchen TC/Rockwell/5-20-04 SMU Engineering p. 1 Outline Security problems

More information

Keywords - Intrusion Detection System, Intrusion Prevention System, Artificial Neural Network, Multi Layer Perceptron, SYN_FLOOD, PING_FLOOD, JPCap

Keywords - Intrusion Detection System, Intrusion Prevention System, Artificial Neural Network, Multi Layer Perceptron, SYN_FLOOD, PING_FLOOD, JPCap Intelligent Monitoring System A network based IDS SONALI M. TIDKE, Dept. of Computer Science and Engineering, Shreeyash College of Engineering and Technology, Aurangabad (MS), India Abstract Network security

More information

Hybrid Intrusion Detection System Using K-Means Algorithm

Hybrid Intrusion Detection System Using K-Means Algorithm International Journal of Computer Sciences and Engineering Open Access Review Paper Volume-4, Issue-3 E-ISSN: 2347-2693 Hybrid Intrusion Detection System Using K-Means Algorithm Darshan K. Dagly 1*, Rohan

More information

A Neural Network Based System for Intrusion Detection and Classification of Attacks

A Neural Network Based System for Intrusion Detection and Classification of Attacks A Neural Network Based System for Intrusion Detection and Classification of Attacks Mehdi MORADI and Mohammad ZULKERNINE Abstract-- With the rapid expansion of computer networks during the past decade,

More information

STANDARDISATION AND CLASSIFICATION OF ALERTS GENERATED BY INTRUSION DETECTION SYSTEMS

STANDARDISATION AND CLASSIFICATION OF ALERTS GENERATED BY INTRUSION DETECTION SYSTEMS STANDARDISATION AND CLASSIFICATION OF ALERTS GENERATED BY INTRUSION DETECTION SYSTEMS Athira A B 1 and Vinod Pathari 2 1 Department of Computer Engineering,National Institute Of Technology Calicut, India

More information

An Evaluation of Machine Learning Method for Intrusion Detection System Using LOF on Jubatus

An Evaluation of Machine Learning Method for Intrusion Detection System Using LOF on Jubatus An Evaluation of Machine Learning Method for Intrusion Detection System Using LOF on Jubatus Tadashi Ogino* Okinawa National College of Technology, Okinawa, Japan. * Corresponding author. Email: ogino@okinawa-ct.ac.jp

More information

Hybrid Model For Intrusion Detection System Chapke Prajkta P., Raut A. B.

Hybrid Model For Intrusion Detection System Chapke Prajkta P., Raut A. B. www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume1 Issue 3 Dec 2012 Page No. 151-155 Hybrid Model For Intrusion Detection System Chapke Prajkta P., Raut A. B.

More information

Observation and Findings

Observation and Findings Chapter 6 Observation and Findings 6.1. Introduction This chapter discuss in detail about observation and findings based on survey performed. This research work is carried out in order to find out network

More information

Application of Data Mining Techniques in Intrusion Detection

Application of Data Mining Techniques in Intrusion Detection Application of Data Mining Techniques in Intrusion Detection LI Min An Yang Institute of Technology leiminxuan@sohu.com Abstract: The article introduced the importance of intrusion detection, as well as

More information

System for Denial-of-Service Attack Detection Based On Triangle Area Generation

System for Denial-of-Service Attack Detection Based On Triangle Area Generation System for Denial-of-Service Attack Detection Based On Triangle Area Generation 1, Heena Salim Shaikh, 2 N Pratik Pramod Shinde, 3 Prathamesh Ravindra Patil, 4 Parag Ramesh Kadam 1, 2, 3, 4 Student 1,

More information

Layered Approach of Intrusion Detection System with Efficient Alert Aggregation for Heterogeneous Networks

Layered Approach of Intrusion Detection System with Efficient Alert Aggregation for Heterogeneous Networks Layered Approach of Intrusion Detection System with Efficient Alert Aggregation for Heterogeneous Networks Lohith Raj S N, Shanthi M B, Jitendranath Mungara Abstract Protecting data from the intruders

More information

An Intrusion Detection System based on Support Vector Machine using Hierarchical Clustering and Genetic Algorithm

An Intrusion Detection System based on Support Vector Machine using Hierarchical Clustering and Genetic Algorithm An Intrusion Detection System based on Support Vector Machine using Hierarchical Clustering and Genetic Algorithm Minakshi Bisen* & Amit Dubey** *M.Tech Scholar, Department of Computer Science & Engineering,

More information

ALGORITHM FOR DISTRIBUTED AGENT BASED NETWORK INTRUSION DETECTION SYSTEM (NIDS)

ALGORITHM FOR DISTRIBUTED AGENT BASED NETWORK INTRUSION DETECTION SYSTEM (NIDS) ALGORITHM FOR DISTRIBUTED AGENT BASED NETWORK INTRUSION DETECTION SYSTEM (NIDS) Aleksandar Sokolovski Faculty Of Informatics, European University Skopje, Macedonia Saso Gelev Faculty Of Informatics, European

More information

International Journal of Computer Science Trends and Technology (IJCST) Volume 2 Issue 3, May-Jun 2014

International Journal of Computer Science Trends and Technology (IJCST) Volume 2 Issue 3, May-Jun 2014 RESEARCH ARTICLE OPEN ACCESS A Survey of Data Mining: Concepts with Applications and its Future Scope Dr. Zubair Khan 1, Ashish Kumar 2, Sunny Kumar 3 M.Tech Research Scholar 2. Department of Computer

More information

Index Terms: Intrusion Detection System (IDS), Training, Neural Network, anomaly detection, misuse detection.

Index Terms: Intrusion Detection System (IDS), Training, Neural Network, anomaly detection, misuse detection. Survey: Learning Techniques for Intrusion Detection System (IDS) Roshani Gaidhane, Student*, Prof. C. Vaidya, Dr. M. Raghuwanshi RGCER, Computer Science and Engineering Department, RTMNU University Nagpur,

More information

Two State Intrusion Detection System Against DDos Attack in Wireless Network

Two State Intrusion Detection System Against DDos Attack in Wireless Network Two State Intrusion Detection System Against DDos Attack in Wireless Network 1 Pintu Vasani, 2 Parikh Dhaval 1 M.E Student, 2 Head of Department (LDCE-CSE) L.D. College of Engineering, Ahmedabad, India.

More information

Denial of Service Attack Detection Using Multivariate Correlation Information and Support Vector Machine Classification

Denial of Service Attack Detection Using Multivariate Correlation Information and Support Vector Machine Classification International Journal of Computer Sciences and Engineering Open Access Research Paper Volume-4, Issue-3 E-ISSN: 2347-2693 Denial of Service Attack Detection Using Multivariate Correlation Information and

More information

Intrusion Detection Using PCA Based Modular Neural Network

Intrusion Detection Using PCA Based Modular Neural Network International Journal of Machine Learning and Computing, Vol. 2, No. 5, October 2012 Intrusion Detection Using PCA Based Modular Neural Network Khaled Al-Nafjan, Musaed A. Al-Hussein, Abdullah S. Alghamdi,

More information

Intrusion Detection System for Cloud Network Using FC-ANN Algorithm

Intrusion Detection System for Cloud Network Using FC-ANN Algorithm Intrusion Detection System for Cloud Network Using FC-ANN Algorithm Swati Ramteke 1, Rajesh Dongare 2, Komal Ramteke 3 Student, Department of Information Technology, VIIT, Pune, India 1 Student, Department

More information

How To Detect Denial Of Service Attack On A Network With A Network Traffic Characterization Scheme

How To Detect Denial Of Service Attack On A Network With A Network Traffic Characterization Scheme Efficient Detection for DOS Attacks by Multivariate Correlation Analysis and Trace Back Method for Prevention Thivya. T 1, Karthika.M 2 Student, Department of computer science and engineering, Dhanalakshmi

More information

Impact of Feature Selection on the Performance of Wireless Intrusion Detection Systems

Impact of Feature Selection on the Performance of Wireless Intrusion Detection Systems 2009 International Conference on Computer Engineering and Applications IPCSIT vol.2 (2011) (2011) IACSIT Press, Singapore Impact of Feature Selection on the Performance of ireless Intrusion Detection Systems

More information

An analysis of suitable parameters for efficiently applying K-means clustering to large TCPdump data set using Hadoop framework

An analysis of suitable parameters for efficiently applying K-means clustering to large TCPdump data set using Hadoop framework An analysis of suitable parameters for efficiently applying K-means clustering to large TCPdump data set using Hadoop framework Jakrarin Therdphapiyanak Dept. of Computer Engineering Chulalongkorn University

More information

A Proposed Architecture of Intrusion Detection Systems for Internet Banking

A Proposed Architecture of Intrusion Detection Systems for Internet Banking A Proposed Architecture of Intrusion Detection Systems for Internet Banking A B S T R A C T Pritika Mehra Post Graduate Department of Computer Science, Khalsa College for Women Amritsar, India Mehra_priti@yahoo.com

More information

The Integration of SNORT with K-Means Clustering Algorithm to Detect New Attack

The Integration of SNORT with K-Means Clustering Algorithm to Detect New Attack The Integration of SNORT with K-Means Clustering Algorithm to Detect New Attack Asnita Hashim, University of Technology MARA, Malaysia April 14-15, 2011 The Integration of SNORT with K-Means Clustering

More information

Classification Algorithms in Intrusion Detection System: A Survey

Classification Algorithms in Intrusion Detection System: A Survey Classification Algorithms in Intrusion Detection System: A Survey V. Jaiganesh 1 Dr. P. Sumathi 2 A.Vinitha 3 1 Doctoral Research Scholar, Department of Computer Science, Manonmaniam Sundaranar University,

More information

Combining Heterogeneous Classifiers for Network Intrusion Detection

Combining Heterogeneous Classifiers for Network Intrusion Detection Combining Heterogeneous Classifiers for Network Intrusion Detection Ali Borji School of Cognitive Sciences, Institute for Studies in Theoretical Physics and Mathematics, Niavaran Bldg. P.O.Box 19395-5746,

More information

Hybrid Intrusion Detection System Model using Clustering, Classification and Decision Table

Hybrid Intrusion Detection System Model using Clustering, Classification and Decision Table IOSR Journal of Computer Engineering (IOSR-JCE) e-issn: 2278-0661, p- ISSN: 2278-8727Volume 9, Issue 4 (Mar. - Apr. 2013), PP 103-107 Hybrid Intrusion Detection System Model using Clustering, Classification

More information

IDS IN TELECOMMUNICATION NETWORK USING PCA

IDS IN TELECOMMUNICATION NETWORK USING PCA IDS IN TELECOMMUNICATION NETWORK USING PCA Mohamed Faisal Elrawy 1, T. K. Abdelhamid 2 and A. M. Mohamed 3 1 Faculty of engineering, MUST University, 6th Of October, Egypt eng_faisal1989@yahoo.com 2,3

More information

A Review on Hybrid Intrusion Detection System using TAN & SVM

A Review on Hybrid Intrusion Detection System using TAN & SVM A Review on Hybrid Intrusion Detection System using TAN & SVM Sumalatha Potteti 1, Namita Parati 2 1 Assistant Professor, Department of CSE,BRECW,Hyderabad,India 2 Assistant Professor, Department of CSE,BRECW,Hyderabad,India

More information

Intrusion Detection. Jeffrey J.P. Tsai. Imperial College Press. A Machine Learning Approach. Zhenwei Yu. University of Illinois, Chicago, USA

Intrusion Detection. Jeffrey J.P. Tsai. Imperial College Press. A Machine Learning Approach. Zhenwei Yu. University of Illinois, Chicago, USA SERIES IN ELECTRICAL AND COMPUTER ENGINEERING Intrusion Detection A Machine Learning Approach Zhenwei Yu University of Illinois, Chicago, USA Jeffrey J.P. Tsai Asia University, University of Illinois,

More information

Customer Classification And Prediction Based On Data Mining Technique

Customer Classification And Prediction Based On Data Mining Technique Customer Classification And Prediction Based On Data Mining Technique Ms. Neethu Baby 1, Mrs. Priyanka L.T 2 1 M.E CSE, Sri Shakthi Institute of Engineering and Technology, Coimbatore 2 Assistant Professor

More information

A New Model for Pre-analysis of Network Traffic Using Similarity Measurement

A New Model for Pre-analysis of Network Traffic Using Similarity Measurement A New Model for Pre-analysis of Network Traffic Using Similarity Measurement Enas Ayman Al-Utrakchi Zarqa University/Department of Computer Science, Zarqa, 13132, Jordan e_utrakchi@yahoo.com Mohammad Rasmi

More information

Social Media Mining. Data Mining Essentials

Social Media Mining. Data Mining Essentials Introduction Data production rate has been increased dramatically (Big Data) and we are able store much more data than before E.g., purchase data, social media data, mobile phone data Businesses and customers

More information

CHAPTER 1 INTRODUCTION

CHAPTER 1 INTRODUCTION 21 CHAPTER 1 INTRODUCTION 1.1 PREAMBLE Wireless ad-hoc network is an autonomous system of wireless nodes connected by wireless links. Wireless ad-hoc network provides a communication over the shared wireless

More information

Intrusion Detection via Machine Learning for SCADA System Protection

Intrusion Detection via Machine Learning for SCADA System Protection Intrusion Detection via Machine Learning for SCADA System Protection S.L.P. Yasakethu Department of Computing, University of Surrey, Guildford, GU2 7XH, UK. s.l.yasakethu@surrey.ac.uk J. Jiang Department

More information

Introduction... Error! Bookmark not defined. Intrusion detection & prevention principles... Error! Bookmark not defined.

Introduction... Error! Bookmark not defined. Intrusion detection & prevention principles... Error! Bookmark not defined. Contents Introduction... Error! Bookmark not defined. Intrusion detection & prevention principles... Error! Bookmark not defined. Technical OverView... Error! Bookmark not defined. Network Intrusion Detection

More information

Network packet payload analysis for intrusion detection

Network packet payload analysis for intrusion detection Network packet payload analysis for intrusion detection Sasa Mrdovic Abstract This paper explores possibility of detecting intrusions into computer networks using network packet payload analysis. Quick

More information

Intrusion Detection System in Campus Network: SNORT the most powerful Open Source Network Security Tool

Intrusion Detection System in Campus Network: SNORT the most powerful Open Source Network Security Tool Intrusion Detection System in Campus Network: SNORT the most powerful Open Source Network Security Tool Mukta Garg Assistant Professor, Advanced Educational Institutions, Palwal Abstract Today s society

More information

Role of Anomaly IDS in Network

Role of Anomaly IDS in Network Role of Anomaly IDS in Network SumathyMurugan 1, Dr.M.Sundara Rajan 2 1 Asst. Prof, Department of Computer Science, Thiruthangal Nadar College, Chennai -51. 2 Asst. Prof, Department of Computer Science,

More information

Network TrafficBehaviorAnalysisby Decomposition into Control and Data Planes

Network TrafficBehaviorAnalysisby Decomposition into Control and Data Planes Network TrafficBehaviorAnalysisby Decomposition into Control and Data Planes Basil AsSadhan, Hyong Kim, José M. F. Moura, Xiaohui Wang Carnegie Mellon University Electrical and Computer Engineering Department

More information

False Positives Reduction Techniques in Intrusion Detection Systems-A Review

False Positives Reduction Techniques in Intrusion Detection Systems-A Review 128 False Positives Reduction Techniques in Intrusion Detection Systems-A Review Asieh Mokarian, Ahmad Faraahi, Arash Ghorbannia Delavar, Payame Noor University, Tehran, IRAN Summary During the last decade

More information

KEITH LEHNERT AND ERIC FRIEDRICH

KEITH LEHNERT AND ERIC FRIEDRICH MACHINE LEARNING CLASSIFICATION OF MALICIOUS NETWORK TRAFFIC KEITH LEHNERT AND ERIC FRIEDRICH 1. Introduction 1.1. Intrusion Detection Systems. In our society, information systems are everywhere. They

More information

Survey of Data Mining Approach using IDS

Survey of Data Mining Approach using IDS Survey of Data Mining Approach using IDS 1 Raman kamboj, 2 Kamal Kumar Research Scholar, Assistant Professor SDDIET, Department of Computer Science & Engineering, Kurukshetra Universty Abstract - In our

More information

Automated Malware Detection Based on Novel Network Behavioral Signatures

Automated Malware Detection Based on Novel Network Behavioral Signatures Automated Malware Detection Based on Novel Network Behavioral Signatures Maros Barabas, Ivan Homoliak, Michal Drozd, and Petr Hanacek 112 metrics divided into five categories according to their nature.

More information

Network Intrusion Detection Systems

Network Intrusion Detection Systems Network Intrusion Detection Systems False Positive Reduction Through Anomaly Detection Joint research by Emmanuele Zambon & Damiano Bolzoni 7/1/06 NIDS - False Positive reduction through Anomaly Detection

More information

Speedy Signature Based Intrusion Detection System Using Finite State Machine and Hashing Techniques

Speedy Signature Based Intrusion Detection System Using Finite State Machine and Hashing Techniques www.ijcsi.org 387 Speedy Signature Based Intrusion Detection System Using Finite State Machine and Hashing Techniques Utkarsh Dixit 1, Shivali Gupta 2 and Om Pal 3 1 School of Computer Science, Centre

More information

Intrusion Detection System Based Network Using SNORT Signatures And WINPCAP

Intrusion Detection System Based Network Using SNORT Signatures And WINPCAP Intrusion Detection System Based Network Using SNORT Signatures And WINPCAP Aakanksha Vijay M.tech, Department of Computer Science Suresh Gyan Vihar University Jaipur, India Mrs Savita Shiwani Head Of

More information

A Survey on Intrusion Detection System with Data Mining Techniques

A Survey on Intrusion Detection System with Data Mining Techniques A Survey on Intrusion Detection System with Data Mining Techniques Ms. Ruth D 1, Mrs. Lovelin Ponn Felciah M 2 1 M.Phil Scholar, Department of Computer Science, Bishop Heber College (Autonomous), Trichirappalli,

More information

Neural Networks for Intrusion Detection and Its Applications

Neural Networks for Intrusion Detection and Its Applications , July 3-5, 2013, London, U.K. Neural Networks for Intrusion Detection and Its Applications E.Kesavulu Reddy, Member IAENG Abstract: With rapid expansion of computer networks during the past decade, security

More information

Network Intrusion Detection using Semi Supervised Support Vector Machine

Network Intrusion Detection using Semi Supervised Support Vector Machine Network Intrusion Detection using Semi Supervised Support Vector Machine Jyoti Haweliya Department of Computer Engineering Institute of Engineering & Technology, Devi Ahilya University Indore, India ABSTRACT

More information

Review Article Intrusion Detection Systems Based on Artificial Intelligence Techniques in Wireless Sensor Networks

Review Article Intrusion Detection Systems Based on Artificial Intelligence Techniques in Wireless Sensor Networks Distributed Sensor Networks, Article ID 351047, 6 pages http://dx.doi.org/10.1155/2013/351047 Review Article Intrusion Detection Systems Based on Artificial Intelligence Techniques in Wireless Sensor Networks

More information

First Semester Computer Science Students Academic Performances Analysis by Using Data Mining Classification Algorithms

First Semester Computer Science Students Academic Performances Analysis by Using Data Mining Classification Algorithms First Semester Computer Science Students Academic Performances Analysis by Using Data Mining Classification Algorithms Azwa Abdul Aziz, Nor Hafieza IsmailandFadhilah Ahmad Faculty Informatics & Computing

More information

Data Mining For Intrusion Detection Systems. Monique Wooten. Professor Robila

Data Mining For Intrusion Detection Systems. Monique Wooten. Professor Robila Data Mining For Intrusion Detection Systems Monique Wooten Professor Robila December 15, 2008 Wooten 2 ABSTRACT The paper discusses the use of data mining techniques applied to intrusion detection systems.

More information

USING GENETIC ALGORITHM IN NETWORK SECURITY

USING GENETIC ALGORITHM IN NETWORK SECURITY USING GENETIC ALGORITHM IN NETWORK SECURITY Ehab Talal Abdel-Ra'of Bader 1 & Hebah H. O. Nasereddin 2 1 Amman Arab University. 2 Middle East University, P.O. Box: 144378, Code 11814, Amman-Jordan Email:

More information

Integration Misuse and Anomaly Detection Techniques on Distributed Sensors

Integration Misuse and Anomaly Detection Techniques on Distributed Sensors Integration Misuse and Anomaly Detection Techniques on Distributed Sensors Shih-Yi Tu Chung-Huang Yang Kouichi Sakurai Graduate Institute of Information and Computer Education, National Kaohsiung Normal

More information

Performance Comparison between Backpropagation Algorithms Applied to Intrusion Detection in Computer Network Systems

Performance Comparison between Backpropagation Algorithms Applied to Intrusion Detection in Computer Network Systems Performance Comparison between Backpropagation Algorithms Applied to Intrusion Detection in Computer Network Systems Iftikhar Ahmad, M.A Ansari, Sajjad Mohsin Department of Computer Sciences, Federal Urdu

More information

A Novel Datamining Based Approach for Remote Intrusion Detection

A Novel Datamining Based Approach for Remote Intrusion Detection A Novel Datamining Based Approach for Remote Intrusion Detection Renu Deepti.S, Loshma.G CSE, Sri Vasavi Engineering College(JNTUK) Tadepalligudem-India Abstract Today, as information systems are more

More information

International Journal of Innovative Research in Advanced Engineering (IJIRAE) ISSN: 2349-2163 Volume 1 Issue 11 (November 2014)

International Journal of Innovative Research in Advanced Engineering (IJIRAE) ISSN: 2349-2163 Volume 1 Issue 11 (November 2014) Denial-of-Service Attack Detection Mangesh D. Salunke * Prof. Ruhi Kabra G.H.Raisoni CEM, SPPU, Ahmednagar HOD, G.H.Raisoni CEM, SPPU,Ahmednagar Abstract: A DoS (Denial of Service) attack as name indicates

More information

INTERNATIONAL JOURNAL FOR ENGINEERING APPLICATIONS AND TECHNOLOGY DATA MINING IN HEALTHCARE SECTOR. ankitanandurkar2394@gmail.com

INTERNATIONAL JOURNAL FOR ENGINEERING APPLICATIONS AND TECHNOLOGY DATA MINING IN HEALTHCARE SECTOR. ankitanandurkar2394@gmail.com IJFEAT INTERNATIONAL JOURNAL FOR ENGINEERING APPLICATIONS AND TECHNOLOGY DATA MINING IN HEALTHCARE SECTOR Bharti S. Takey 1, Ankita N. Nandurkar 2,Ashwini A. Khobragade 3,Pooja G. Jaiswal 4,Swapnil R.

More information

Network Intrusion Detection using Random Forests

Network Intrusion Detection using Random Forests Network Intrusion Detection using Random Forests Jiong Zhang and Mohammad Zulkernine School of Computing Queen s University, Kingston Ontario, Canada K7L 3N6 {zhang, mzulker} @cs.queensu.ca Abstract Network

More information

A Content based Spam Filtering Using Optical Back Propagation Technique

A Content based Spam Filtering Using Optical Back Propagation Technique A Content based Spam Filtering Using Optical Back Propagation Technique Sarab M. Hameed 1, Noor Alhuda J. Mohammed 2 Department of Computer Science, College of Science, University of Baghdad - Iraq ABSTRACT

More information

Application of Data Mining to Network Intrusion Detection: Classifier Selection Model

Application of Data Mining to Network Intrusion Detection: Classifier Selection Model Application of Data Mining to Network Intrusion Detection: Classifier Selection Model Huy Anh Nguyen and Deokjai Choi Chonnam National University, Computer Science Department, 300 Yongbong-dong, Buk-ku,

More information

VHDL Modeling of Intrusion Detection & Prevention System (IDPS) A Neural Network Approach

VHDL Modeling of Intrusion Detection & Prevention System (IDPS) A Neural Network Approach VHDL Modeling of Intrusion Detection & Prevention System (IDPS) A Neural Network Approach Tanusree Chatterjee Department of Computer Science Regent Education and Research Foundation Abstract- The rapid

More information

NETWORK-BASED INTRUSION DETECTION USING NEURAL NETWORKS

NETWORK-BASED INTRUSION DETECTION USING NEURAL NETWORKS 1 NETWORK-BASED INTRUSION DETECTION USING NEURAL NETWORKS ALAN BIVENS biven@cs.rpi.edu RASHEDA SMITH smithr2@cs.rpi.edu CHANDRIKA PALAGIRI palgac@cs.rpi.edu BOLESLAW SZYMANSKI szymansk@cs.rpi.edu MARK

More information

IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 38, NO. 5, SEPTEMBER 2008 649

IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 38, NO. 5, SEPTEMBER 2008 649 IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS PART C: APPLICATIONS AND REVIEWS, VOL. 38, NO. 5, SEPTEMBER 2008 649 Random-Forests-Based Network Intrusion Detection Systems Jiong Zhang, Mohammad Zulkernine,

More information

Intrusion Detection Systems. Overview. Evolution of IDSs. Oussama El-Rawas. History and Concepts of IDSs

Intrusion Detection Systems. Overview. Evolution of IDSs. Oussama El-Rawas. History and Concepts of IDSs Intrusion Detection Systems Oussama El-Rawas History and Concepts of IDSs Overview A brief description about the history of Intrusion Detection Systems An introduction to Intrusion Detection Systems including:

More information

Performance Evaluation of Intrusion Detection Systems

Performance Evaluation of Intrusion Detection Systems Performance Evaluation of Intrusion Detection Systems Waleed Farag & Sanwar Ali Department of Computer Science at Indiana University of Pennsylvania ABIT 2006 Outline Introduction: Intrusion Detection

More information

Available online at www.sciencedirect.com. Procedia Engineering 30 (2012) 1 9

Available online at www.sciencedirect.com. Procedia Engineering 30 (2012) 1 9 Available online at www.sciencedirect.com Procedia Engineering 00 (2011) 000 000 Procedia Engineering 30 (2012) 1 9 Procedia Engineering www.elsevier.com/locate/procedia International Conference on Communication

More information

Robust Preprocessing and Random Forests Technique for Network Probe Anomaly Detection

Robust Preprocessing and Random Forests Technique for Network Probe Anomaly Detection International Journal of Soft Computing and Engineering (IJSCE) Robust Preprocessing and Random Forests Technique for Network Probe Anomaly Detection G. Sunil Kumar, C.V.K Sirisha, Kanaka Durga.R, A.Devi

More information

Feature Subset Selection in E-mail Spam Detection

Feature Subset Selection in E-mail Spam Detection Feature Subset Selection in E-mail Spam Detection Amir Rajabi Behjat, Universiti Technology MARA, Malaysia IT Security for the Next Generation Asia Pacific & MEA Cup, Hong Kong 14-16 March, 2012 Feature

More information

Intrusion Detection Systems

Intrusion Detection Systems Intrusion Detection Systems Assessment of the operation and usefulness of informatics tools for the detection of on-going computer attacks André Matos Luís Machado Work Topics 1. Definition 2. Characteristics

More information

Improving Intrusion Detection Performance Using Keyword Selection and Neural Networks. Abstract

Improving Intrusion Detection Performance Using Keyword Selection and Neural Networks. Abstract Improving Intrusion Detection Performance Using Keyword Selection and Neural Networks Richard P. Lippmann MIT Lincoln Laboratory, Rm S4-121 244 Wood Street Lexington, MA 02173-0073 rpl@sst.ll.mit.edu phone:

More information

Identifying At-Risk Students Using Machine Learning Techniques: A Case Study with IS 100

Identifying At-Risk Students Using Machine Learning Techniques: A Case Study with IS 100 Identifying At-Risk Students Using Machine Learning Techniques: A Case Study with IS 100 Erkan Er Abstract In this paper, a model for predicting students performance levels is proposed which employs three

More information

A Practical Approach to Anomaly based Intrusion Detection System by Outlier Mining in Network Traffic

A Practical Approach to Anomaly based Intrusion Detection System by Outlier Mining in Network Traffic A Practical Approach to Anomaly based Intrusion Detection System by Outlier Mining in Network Traffic By Prajowal Manandhar A Thesis Presented to the Masdar Institute of Science and Technology in Partial

More information

Spam detection with data mining method:

Spam detection with data mining method: Spam detection with data mining method: Ensemble learning with multiple SVM based classifiers to optimize generalization ability of email spam classification Keywords: ensemble learning, SVM classifier,

More information

An Alternative Model Of Virtualization Based Intrusion Detection System In Cloud Computing

An Alternative Model Of Virtualization Based Intrusion Detection System In Cloud Computing An Alternative Model Of Virtualization Based Intrusion Detection System In Cloud Computing Partha Ghosh, Ria Ghosh, Ruma Dutta Abstract: The massive jumps in technology led to the expansion of Cloud Computing

More information

FUZZY DATA MINING AND GENETIC ALGORITHMS APPLIED TO INTRUSION DETECTION

FUZZY DATA MINING AND GENETIC ALGORITHMS APPLIED TO INTRUSION DETECTION FUZZY DATA MINING AND GENETIC ALGORITHMS APPLIED TO INTRUSION DETECTION Susan M. Bridges Bridges@cs.msstate.edu Rayford B. Vaughn vaughn@cs.msstate.edu 23 rd National Information Systems Security Conference

More information

EFFICIENT DATA PRE-PROCESSING FOR DATA MINING

EFFICIENT DATA PRE-PROCESSING FOR DATA MINING EFFICIENT DATA PRE-PROCESSING FOR DATA MINING USING NEURAL NETWORKS JothiKumar.R 1, Sivabalan.R.V 2 1 Research scholar, Noorul Islam University, Nagercoil, India Assistant Professor, Adhiparasakthi College

More information

Adaptive Layered Approach using Machine Learning Techniques with Gain Ratio for Intrusion Detection Systems

Adaptive Layered Approach using Machine Learning Techniques with Gain Ratio for Intrusion Detection Systems Adaptive Layered Approach using Machine Learning Techniques with Gain for Intrusion Detection Systems Heba Ezzat Ibrahim Arab Academy for Science, Technology and Maritime Transport Cairo, Egypt Sherif

More information

Volume 3, Issue 3, March 2015 International Journal of Advance Research in Computer Science and Management Studies

Volume 3, Issue 3, March 2015 International Journal of Advance Research in Computer Science and Management Studies Volume 3, Issue 3, March 2015 International Journal of Advance Research in Computer Science and Management Studies Research Article / Survey Paper / Case Study Available online at: www.ijarcsms.com A Review

More information