One of the major problems in Big Data is a large number of features or dimensions, which causes the issue of “the curse of dimensionality” when applying machine learning, especially classification algorithms. Feature selection is an important technique which selects small and informative feature subsets to improve the learning performance. Feature selection is not an easy task due to its large and complex search space. Recently, swarm intelligence techniques have gained much attention from the feature selection community because of their simplicity and potential global search ability. However, there has been no comprehensive surveys on swarm intelligence for feature selection in classification which is the most widely investigated area in feature selection. Only a few short surveys is this area are still lack of in-depth discussions on the state-of-the-art methods, and the strengths and limitations of existing methods, particularly in terms of the representation and search mechanisms, which are two key components in adapting swarm intelligence to address feature selection problems. This paper presents a comprehensive survey on the state-of-the-art works applying swarm intelligence to achieve feature selection in classification, with a focus on the representation and search mechanisms. The expectation is to present an overview of different kinds of state-of-the-art approaches together with their advantages and disadvantages, encourage researchers to investigate more advanced methods, provide practitioners guidances for choosing the appropriate methods to be used in real-world scenarios, and discuss potential limitations and issues for future research.

A survey on swarm intelligence approaches to feature selection in data mining

As Internet access widens, IDS (Intrusion Detection System) is becoming a very important component of network security to prevent unauthorized use and misuse of data. An IDS routinely handles massive amounts of data traffc that contain redundant and irrelevant features, which impact the performance of the IDS negatively. Feature selection methods play an important role in eliminating unrelated and redundant features in IDS. Statistical analysis, neural networks, machine learning, data mining techniques, and support vector machine models are employed in some such methods. Good feature selection leads to better classification accuracy. Recently, bio-inspired optimization algorithms have been used for feature selection. This work provides a survey of feature selection techniques for IDS, including bio-inspired algorithms.

/pdf/feature-selection-techniques-for-intrusion-detection-using-24wex1yiit.pdf

Feature selection techniques for intrusion detection using non-bio-inspired and bio-inspired optimization algorithms

The rapid development of information technology leads to increasing the number of devices connected to the Internet. Besides, the amount of network attacks also increased. Accordingly, there is an urgent demand to design a defence system proficient in discovering new kinds of attacks. One of the most effective protection systems is intrusion detection system (IDS). The IDS is an intelligent system that monitors and inspects the network packets to identify the abnormal behavior. In addition, the network packets comprise many attributes and there are many attributes that are irrelevant and repetitive which degrade the performance of the IDS system and overwhelm the system resources. A feature selection technique helps to reduce the computation time and complexity by selecting the optimum subset of features. In this paper, an enhanced anomaly-based IDS model based on multi-objective grey wolf optimisation (GWO) algorithm was proposed. The GWO algorithm was employed as a feature selection mechanism to identify the most relevant features from the dataset that contribute to high classification accuracy. Furthermore, support vector machine was used to estimate the capability of selected features in predicting the attacks accurately. Moreover, 20% of NSL–KDD dataset was used to demonstrate effectiveness of the proposed approach through different attack scenarios. The experimental result revealed that the proposed approach obtains classification accuracy of (93.64%, 91.01%, 57.72%, 53.7%) for DoS, Probe, R2L, and U2R attack respectively. Finally, the proposed approach was compared with other existing approaches and achieves significant result.

Anomaly-based intrusion detection system using multi-objective grey wolf optimisation algorithm

Network intrusion is a critical challenge in information and communication systems amongst other forms of fraud perpetrated over the Internet. Despite the various traditional techniques proposed to prevent this intrusion, the threat persists. These days, intrusion detection systems (IDS) are faced with detecting attacks in large streams of connections due to the sporadic increase in network traffics. Although machine learning (ML) has been introduced in IDS to deal with finding patterns in big data, the irrelevant features in the data tend to degrade both the speed and accuracy of detection of attacks. Also, it increases the computational resource needed during training and testing of IDS models. Therefore, in this paper, we seek to find the optimal feature set using discretized differential evolution (DDE) and C4.5 ML algorithm from NSL-KDD standard intrusion dataset. The result obtained shows a significant improvement in detection accuracy, a reduction in training and testing time using the reduced feature set. The method also buttresses the fact that differential evolution (DE) is not limited to optimization of continuous problems but work well for discrete optimization.

/pdf/efficient-feature-selection-technique-for-network-intrusion-jx0ffi23i4.pdf

Efficient Feature Selection Technique for Network Intrusion Detection System Using Discrete Differential Evolution and Decision.

In the process of detecting different kinds of attacks in anomaly-based intrusion detection system (IDS), both normal and attack data are profiled with the help of selected attributes. Various types of attributes are collected to create the attack and normal traffic patterns. Some of the attributes are derived from protocol header fields, and some of them represent continuous information profiled over a period. “Curse of Dimensionality” is one of the major issues in IDS. The computational complexity of the model generation and classification time of IDS is directly proportional to the number of attributes of the profile. In a typical IDS preprocessing stage, more significant features among the available features are selected. This paper presents a brief taxonomy of several feature selection methods with emphasis on soft computing techniques, viz., rough sets, fuzzy rough sets, and ant colony optimization.

A Survey of Feature Selection Techniques in Intrusion Detection System: A Soft Computing Perspective

Feature selection and classifier design is the key to network intrusion detection. In order to improve network intrusion detection rate for feature selection problem, this paper proposed a network intrusion detection method (ACO-FS -SVM) combining ant colony algorithm to select the features with a feature weighting SVM. First, the use of support vector machine classification accuracy and feature subset dimension construct a comprehensive fitness weighting index. Then use the ant colony algorithm for global optimization and multiple search capabilities to achieve optimal solutions feature subset search feature. And then selected the key feature of network data and calculated information gain access to various features weights and heavy weights to build support vector machine classifier based on the characteristics of network attacks right. At last, refine the final design of the local search methods to make the feature selection results without redundant features while improve the convergence resistance, and verify the data set by KDD1999 effectiveness of the algorithm. The results show that ACO-FS-SVM can effectively reduce the dimension of features, and have improved network intrusion detection accuracy and detection speed.

http://www.sersc.org/journals/IJSIA/vol9_no4_2015/24.pdf

Hunan Changde

Papers

ACO and SVM Selection Feature Weighting of Network Intrusion Detection Method