Intrusion Detection using NBHoeffding Rule based Decision Tree for Wireless Sensor Networks

Network Intrusion Detection Systems (NIDS) are utilized to find hostile network connections. This can be accomplished by looking at traffic network activity, but it takes a lot of work. The NIDS heavily utilizes approaches for data extraction and machine learning to find anomalies. In terms of feature selection, NIDS is far more effective. This is accurate since anomaly identification uses a number of time-consuming features. Because of this, the feature selection method influences how long it takes to analyze movement patterns and how clear it is. The goal of the study is to provide NIDS with an attribute selection approach. PSO has been used for that purpose. The Network Intrusion Detection System that is being developed will be able to identify any malicious activity in the network or any unusual behavior in the network, allowing the identification of the illegal activities and safeguarding the enormous amounts of confidential data belonging to the customers from being compromised. In the research, datasets were produced utilising both a network infrastructure and a simulation network. Wireshark is used to gather data packets whereas Cisco Packet Tracer is used to build a network in a simulated environment. Additionally, a physical network consisting of six node MCUs connected to a laptop and a mobile hotspot, has been built and communication packets are being recorded using the Wireshark tool. To train several machine learning models, all the datasets that were gathered—created datasets from our own studies as well as some common datasets like NSDL and UNSW acquired from Kaggle—were employed. Additionally, PSO, which is an optimization method, has been used with these ML algorithms for feature selection. In the research, KNN, decision trees, and ANN have all been combined with PSO for a specific case study. And it was found demonstrated the classification methods PSO + ANN outperformed PSO + KNN and PSO + DT in this case study.

Wireless Sensor Networks (WSNs) have become a hot research topic in recent years. They have many potential applications for both civil and military tasks. However, the unattended nature of WSNs and the limited computational and energy resources of their nodes make them susceptible to many types of attacks. Intrusion detection is one of the major and efficient defence methods against attacks in a network infrastructure. Intrusion Detection Systems can be seen as the second line of defence and they complement the security primitives that are adopted in order to prevent attacks against the computer network being protected. The peculiar features of a wireless sensor network pose stringent requirements to the design of intrusion detection systems. In this paper, we propose a hybrid, lightweight, distributed Intrusion Detection System (IDS) for wireless sensor networks. This IDS uses both misuse-based and anomaly-based detection techniques. It is composed of a Central Agent, which performs highly accurate intrusion detection by using data mining techniques, and a number of Local Agents running lighter anomaly-based detection techniques on the motes. Decision trees have been adopted as classification algorithm in the detection process of the Central Agent and their behaviour has been analysed in selected attacks scenarios. The accuracy of the proposed IDS has been measured and validated through an extensive experimental campaign. This paper presents the results of these experimental tests.

Over the years, the vulnerability of the network system has completely revolutionized the security system. Attackers simply exploit these exposures to gain undue access to network resources. It is essential to safeguard network resources with the Intrusion Prevention System (IPS) and Intrusion Detection System (IDS). Hence, to develop an optimized IDS model using ensemble modeling of machine learning technique is paramount. Therefore, this research attempts to build a network IDS using an ensemble of Decision Trees (DT) algorithms to optimize an IDS model. Furthermore, the C4.5 DT classifier used on the University of New South Wales Network-Based 2015 (UNSW-NB15) network data set. Then the ensemble techniques of bagging and AdaBoost are compared during the performance evaluation. The results of this work depict that performance increase as training size increases. Consequently, the results showed that adopting the ensemble model for the C4.5 decision tree classified improved network intrusion classification compared to using the machine learning algorithm in isolation. Moreover, the result also showed an improved classification model for network IDS model. The AdaBoost ensemble model of the C4.5 DT algorithm using partitions of 90% of training and 10% of testing data sets performs the best with 98% accuracy and precision.

Using decision tree algorithms for estimating ICU admission of COVID-19 patients

With the advancement of network technologies, new threats and attacks have started to gain space. These have put the safety of the network at stake. These threats and attacks could be significantly detected by a Network Intrusion Detection System (NIDS). The existing NIDS makes use of classical ML algorithms that fail to detect new attacks. This happens because the traditional systems are often built on selection-based intrusion detection techniques which only detect the attacks for which they are trained for. Another intrusion detection technique is anomaly-based detection which is capable of detecting new attacks if an unsupervised method is used. But this system suffers from high false-positive rates. Hence, this work proposes such a hybrid system where a combination of signature-based detection algorithm(Decision Tree, Naive Bayes, or its variants) with an unsupervised (clustered) anomaly-based detection algorithm (DBSCAN or Isolation Forest) is used to detect both studied as well as novel attacks. The proposed work is aimed to improve the detection rate and lowering the false alarm rate.

A scalable decision tree system and its application in pattern recognition and intrusion detection

Offline signature recognition has great importance in our day to day activities. Researchers are trying to use them as biometric identification in various areas like banks, security systems and for other identification purposes. Fingerprints, iris, thumb impression and face detection based biometrics are successfully used for identification of individuals because of their static nature. However, people’s signatures show variability that makes it difficult to recognize the original signatures correctly and to use them as biometrics. The handwritten signatures have importance in banks for cheque, credit card processing, legal and financial transactions, and the signatures are the main target of fraudulence. To deal with complex signatures, there should be a robust signature verification method in places such as banks that can correctly classify the signatures into genuine or forgery to avoid financial frauds. This paper, presents a pixels intensity level based offline signature verification model for the correct classification of signatures. To achieve the target, three statistical classifiers; Decision Tree (J48), probability based Naive Bayes (NB tree) and Euclidean distance based k-Nearest Neighbor (IBk), are used. For comparison of the accuracy rates of offline signatures with online signatures, three classifiers were applied on online signature database and achieved a 99.90% accuracy rate with decision tree (J48), 99.82% with Naive Bayes Tree and 98.11% with K-Nearest Neighbor (with 10 fold cross validation). The results of offline signatures were 64.97% accuracy rate with decision tree (J48), 76.16% with Naive Bayes Tree and 91.91% with k-Nearest Neighbor (IBk) (without forgeries). The accuracy rate dropped with the inclusion of forgery signatures as, 55.63% accuracy rate with decision tree (J48), 67.02% with Naive Bayes Tree and 88.12% (with forgeries).

Enhancing cyber security in WSN using optimized self-attention-based provisional variational auto-encoder generative adversarial network

Augmentation in performance and security of WSNs for IoT applications using feature selection and classification techniques

Classification, using the decision tree algorithm, is a widely studied problem in data streams. The challenge is when to split a decision node into multiple leaves. Concentration inequalities, that exploit variance information such as Bernstein’s and Bennett’s inequalities, are often substantially strict as compared with Hoeffding’s bound which disregards variance. Many machine learning algorithms for stream classification such as very fast decision tree (VFDT) learner, AdaBoost and support vector machines (SVMs), use the Hoeffding’s bound as a performance guarantee. In this paper, we propose a new algorithm based on the recently proposed empirical Bernstein’s bound to achieve a better probabilistic bound on the accuracy of the decision tree. Experimental results on four synthetic and two real world data sets demonstrate the performance gain of our proposed technique.

With the rapid development of information construction, the network has been everywhere; this technology has brought a lot of convenience to people, but there are also some security problems. In order to solve these problems, many methods have been proposed, among which intrusion detection technology is an important part of solving security problems. Therefore, the traditional intrusion detection system technology is difficult to adapt to the current security system requirements, and it is necessary to put forward a more effective method. At present, most of the clustering algorithm in a large number of network data set to set the cluster number and the lack of the ability of handling character properties in network transaction this kind of situation, is because these algorithms for clustering number shows the dependence of the determination of K value is very important, with the result of clustering and clustering front is fixed, and the lack of ability to deal with focus on the characters of similar properties, k-means or are using K-center is difficult to solve. Therefore, by combining decision tree classification and ant colony clustering, this paper proposes a multi-level hybrid algorithm based on decision tree and ant colony algorithms. Experiments show that this new method is very effective in intrusion detection, with a very low false alarm rate, while maintaining a relatively acceptable false alarm rate, and can also appropriately detect unknown intrusion detection so as to improve the intrusion detection rate.

Traditional Intrusion Detection Systems often face difficulties in detecting the diverse classes of intrusions in the Industrial Internet of Things (IIoT) environment due to its heterogeneity and complexity. Such environments encounter a mixture of deterministic and stochastic cyberattack patterns, corresponding to discrete and continuous features. This challenge results in inefficient feature extraction in current Asynchronous Advantage Actor-Critic (A3C) algorithms. To address these issues, this article proposes a Hybrid A3C with Decision Tree (HADT) algorithm for intrusion detection. In this proposed algorithm, the simple A3C is used to detect stochastic intrusions related to continuous features, while the Decision Tree (DT) algorithm is employed for detecting fixed-pattern (deterministic) intrusions related to discrete features. The outputs of these algorithms are combined in a Feedforward Neural Network (FFNN) layer to accurately classify the intrusion type. Additionally, the X-IIoTID dataset, obtained from a real IIoT environment, was used to compare the performance of HADT with existing A3C algorithms in related studies. Experimental results demonstrate that the HADT algorithm outperforms baseline A3C algorithms in terms of accuracy, precision, recall, and F1 score, maintaining a detection rate above 99.8%, especially for intrusions that are underrepresented in the dataset, while also consuming less power.

In traditional decision (classification) tree algorithms, the label is assumed to be a categorical (class) variable. When the label is a continuous variable in the data, two possible approaches based on existing decision tree algorithms can be used to handle the situations. The first uses a data discretization method in the preprocessing stage to convert the continuous label into a class label defined by a finite set of nonoverlapping intervals and then applies a decision tree algorithm. The second simply applies a regression tree algorithm, using the continuous label directly. These approaches have their own drawbacks. We propose an algorithm that dynamically discretizes the continuous label at each node during the tree induction process. Extensive experiments show that the proposed method outperforms the preprocessing approach, the regression tree approach, and several nontree-based algorithms.

Some security protocols or mechanisms have been designed for wireless sensor networks (WSNs). However, an intrusion detection system (IDS) should always be deployed on security critical applications to defense in depth. Due to the resource constraints, the intrusion detection system for traditional network cannot be used directly in WSNs. Several schemes have been proposed to detect intrusions in wireless sensor networks. But most of them aim on some specific attacks (e.g. selective forwarding) or attacks on particular layers, such as media access layer or routing layer. In this paper, we present a framework of machine learning based intrusion detection system for wireless sensor networks. Our system will not be limited on particular attacks, while machine learning algorithm helps to build detection model from training data automatically, which will save human labor from writing signature of attacks or specifying the normal behavior of a sensor node.

This study proposes an innovative methodological framework for assessing flood susceptibility in deltaic areas, by integrating fuzzy logic with three decision tree algorithms: CART, Random Forest and Alternating Decision Tree. The novelty of the research lies in the explicit combination of the ability of fuzzy logic to manage the uncertainty and ambiguity of geographical factors with the predictive power and interpretability of decision tree algorithms, thus responding to a gap in the specialized literature, where these methods are often applied separately or without a clear comparative justification. The methodology is justified by the specific complexity of deltaic areas, characterized by low altitudes, multiple hydrological interactions and variable anthropogenic influences, which classical or ‘black-box’ models incompletely capture. The study uses eight determinants, weighted by fuzzy coefficients validated by historical data, and evaluates the performance of the models using ROC, AUC and statistical indicators. The results show excellent performance for all models (AUC > 0.90), highlighting relevant functional differences: Fuzzy-CART offers a conservative approach, Fuzzy-RF ensures the highest accuracy and stability, and Fuzzy-ADT provides a balanced and nuanced assessment. The study provides a robust and transparent tool for flood risk management in the Danube Delta.