Negative Selection Algorithm Research Articles

Self-Supervised EEG Representation Learning for Robust Emotion Recognition

Emotion recognition based on electroencephalography (EEG) is becoming a growing concern of researchers due to its various applications and portable devices. Existing methods are mainly dedicated to EEG feature representation and have made impressive progress. However, the problem of scarce labels restricts their further promotion. In light of this, we propose a self-supervised framework with contrastive learning for robust EEG-based emotion recognition, which can effectively leverage both readily available unlabeled EEG signals and labeled ones to learn highly discriminative EEG features. Firstly, we construct a specific pretext task according to the sequential non-stationarity of emotional EEG signals for contrastive learning, which aims at extracting pseudo-label information from all EEG data. Meanwhile, we propose a novel negative segment selection algorithm to reduce the noise of unlabeled data during the contrastive learning process. Secondly, to mitigate the overfitting issue induced by a small number of labeled samples during learning, we originate a loss function with label smoothing regularization that can guide the model to learn generalizable features. Extensive experiments over three benchmark datasets demonstrate the effectiveness and superiority of our model on EEG-based emotion recognition task. Besides, the generalization and robustness of the model have also been proved through sufficient experiments.

Ontology-based Negative Selection Algorithm For Pairwise Testing

Ontology-based Negative Selection Algorithm For Pairwise Testing

A negative selection algorithm with human-in-the-loop for anomaly detection

The existing negative selection algorithms can not improve their detection performance by human intervention during the testing process. This paper proposes a negative selection algorithm with human-in-the-loop for anomaly detection. It uses self-sample clusters to train detectors with a nonrandom strategy. Its detectors and self-sample clusters fully cover state space without overlapping each other. It locally adjusts detectors and self-sample clusters with human intervention to improve its detection performance during the testing process. Experiments were performed on two synthetic datasets and the Iris dataset from the UCI repository to assess its performance. The results show that it outperforms the other anomaly detection methods in most cases.

A Network Intrusion Detection Model Based on GA-Improved NSA

With the popularization of the Internet, network security issues have also emerged. In response to network security issues, there are certain shortcomings in current network intrusion detection technologies. To improve and optimize this technology, a network intrusion detection model based on genetic algorithm and improved negative selection algorithm is designed. The generation of detectors in the selection algorithm is replaced by genetic algorithm, and the non-self spatial distribution of detectors is optimized. This paper proposes a network intrusion detection model using a genetic algorithm-improved negative selection algorithm (GA-INSA) and an improved LeNet-5 CNN. The GA enhances detector generation and distribution in NSA while SMOTE handles class imbalance for CNN. Experiments show GA-INSA has over 9% higher accuracy than NSA, SVM and GA-BP across different data sizes. The improved LeNet-5 demonstrates superior accuracy and recall rates over by over 20% baseline LeNet-5. However, more comprehensive evaluation on public datasets, design details on architectures, and discussion around limitations are warranted. The data shows that the designed network intrusion detection model has better performance. The model can provide technical support for network intrusion detection in reality and can enrich the content of network intrusion detection technology.

A negative selection algorithm with hypercube interface detectors for anomaly detection

Negative selection algorithms play an important role in anomaly detection. Interface detectors are a special negative selection algorithm that completely eliminates outer holes, but there are detection blind areas. This paper proposes a negative selection algorithm with hypercube interface detectors for anomaly detection. It uses self-sample clusters to construct self-space, and boundary self-sample clusters to describe the interface detectors. It eliminates the detection blind area and improves the detection rate. To validate the performance of the proposed method, experiments were conducted using the iris dataset, the skin segmentation dataset, the Breast Cancer of Wisconsin (BCW) dataset, and the Waveform Database Generator (Version 2) dataset. Experimental results show that the proposed method in this paper has a higher detection rate, lower false alarm rate, and fewer detectors than other anomaly detection methods for the same parameters.

Artificial Immune Systems in Local and Network Cybersecurity: An Overview of Intrusion Detection Strategies

In this paper, an overview of artificial immune systems (AIS) used in intrusion detection systems (IDS) is provided, along with a review of recent efforts in this field of cybersecurity. In particular, the focus is on the negative selection algorithm (NSA), a popular, prominent algorithm of the AIS domain based on the human immune system. IDS offer intrusion detection capabilities, both locally and in a network environment. The paper offers a review of recent solutions employing AIS in IDS, capable of detecting anomalous network traffic/breaches and operating system file infections caused by malware. A discussion regarding the reviewed research is presented with an analysis and suggestions for further research, and then the work is concluded.

Improving damage detection by combining multiple classifiers in different feature spaces

Vibration-based damage identification methods have been explored in several engineering fields. However, the effective damage-sensitivity of vibration-based features and the balance between false alarms and readiness of the anomaly detection are still open issues. Here, a cost-effective damage detection methodology based on a Deterministically Generated Negative Selection Algorithm is implemented and tested over a wide dataset generated through the numerical simulation of a three-storey concrete building in different damage scenarios. The approach allows training the classifiers only based on undamaged samples and ensures a prompt response to damage outbreak by fusing the classification conducted in parallel on different feature spaces, exploiting their distinct sensitivity to damage. A strategy to improve the reliability of the alarm is implemented by introducing counters and thresholds on consecutive classification outputs. A satisfactory trade-off between false and true alarms is achieved by comparing their probabilities, obtained through Markov chains, for different values of such thresholds.

Damage detection in composite beams using negative selection algorithm method

Most of the structural members act under periodic loading conditions which lead to cracks or damages. The initiation of cracks may change the physical properties of a structure mostly the flexibility and stiffness changes. The dynamic changes will directly change the natural frequencies and mode shapes. This study was an adaptive evolutionary algorithm to identify the damage location based on the immune system of the vertebrates. The first three natural frequencies from FEA and experimental techniques act as an input to the negative selection algorithm. The proposed method results are compared with the results of experimental analysis and FEA. The percentage error in both cases was found to be 3 % which is measurable.

Damage analysis in composite materials using a hybrid artificial immune system

Structural health diagnosis and condition monitoring of structures and components meant for aerospace industries is a challenging task for researchers and aerospace scientists. Presence of a minute crack in a structural element causes variation in local stiffness. The variations of local stiffness depend on crack location and crack depth. Change in stiffness in consequence affects the vibration behaviour of whole structural member. These abnormities in the structures if sustain over a long period cause catastrophic failure. Unexpected failure of any components of space vehicles results huge loss of life, more casualty and affect the economy of a country. To stay away from such situations, regular monitoring on safe operation of structures is extremely essential. Development of efficient damage detection methods is vital for early stage crack detection and its severity. In this work a novel online crack detection method based on artificial immune system has proposed. This article presents analytical and numerical (Finite element method) analysis for crack detection in a hybrid carbon fiber/ kevlar reinforced epoxy composite cantilever beam under free vibration. From the analysis, by varying crack sites and crack depth, alteration in natural frequencies and corresponding mode shapes curvature are recorded. Data (first three natural frequencies) from FEA has trained in negative selection algorithm (NSA). The result of FEA and NSA has compared. Then total average percentage of error is calculated. One statistical method has been used to make NSA more adaptive for the online damage detection. For this purpose a data mining technique named as “Regression Analysis” (RA) is used to reduce residual errors during the data acquisition method. The results are compared with standalone NSA and flexible NSA. Total average percentage of error has calculated.

DGA-PSO: An improved detector generation algorithm based on particle swarm optimization in negative selection

The negative selection algorithm (NSA) is an essential algorithm in the artificial immune system used to achieve anomaly detection by generating detectors. The traditional NSA algorithm generates candidate detectors randomly, which leads to a partially dense and redundant distribution of detectors in the nonself areas, resulting in the presence of holes that are not covered by detectors. A detector generation algorithm based on particle swarm optimization (DGA-PSO) is proposed to overcome these defects. DGA-PSO converts the self-tolerance process into an adaptation function to guide particles to move in a specific direction by artificial settings and variants, generates efficient detectors covering the nonself space, reduces the redundancy among detectors and fills holes not covered. Thus, we successfully reduce the number of detectors while improving the detection rate of the algorithm. Through experimental validation analysis, DGA-PSO ranks first in detector training time and the detection rate on four UCI datasets compared to the classical algorithms RNSA and V-Detector and the improved algorithms BIORV-NSA, ADC-NSA and IFB-NSA.

A novel negative selection algorithm with hyper-rectangle detectors based on full coverage of state space for anomaly detection

The traditional negative selection algorithm with a randomly generated hypersphere detector is unable to satisfy the development needs of continuous learning due to the inherent defects of the detector. This paper proposes a novel negative selection algorithm for hyper-rectangle detectors that overcomes the shortcomings of randomly generated hyper-sphere detectors and lays the foundation for a negative selection algorithm with continuous learning capability. It uses self-sample clusters of equal-sized hypercubes instead of self-samples for training. The hyper-rectangle detectors are generated by cutting the nonself-space along the boundary of the self-sample clusters. The state space is covered without overlapping each other by self-sample clusters and detectors. The anomaly detection performance of the proposed method was demonstrated using Iris data, vowel recognition data (Vowel), and Wisconsin Breast Cancer (BCW) data. The experimental results show that the proposed method outperforms other artificial immune algorithms and clustering algorithms under the same parameter conditions.

Research on prediction of coal and gas outburst risk based on TIWTD-CNSA

When processing the current coal and gas outburst risk index data, it is observed that the data sequence does not have intuitive regularity, and the acquired data signals are generally mixed with different degrees of noise, which hinders data processing and prediction. To reduce the influence of noise signal on prediction, it is proposed to use Translation-invariant Wavelet Threshold Denoising (TIWTD) to process signal data, and the result shows that the noise signal in the original data is significantly reduced. On this basis, a Compact Negative Selection Algorithm (CNSA) is introduced to detect abnormal signals exceeding the critical value in the data and make predictions. After training the model with the gas hazard index feature subset collected during the excavation process of No. 11192 mining face in a coal mine from Guizhou province in China, input the processed indicator data into the model for prediction. The results show that the model can effectively identify the anomalies in the data and accurately predict the risk of coal and gas outburst, the TIWTD-CNSA model can be used to guide engineering practice.

A Modified Gray Wolf Optimizer-Based Negative Selection Algorithm for Network Anomaly Detection

Intrusion detection systems are crucial in fighting against various network attacks. By monitoring the network behavior in real time, possible attack attempts can be detected and acted upon. However, with the development of openness and flexibility of networks, artificial immunity-based network anomaly detection methods lack continuous adaptability and hence have poor detection performance. Thus, a novel framework for network anomaly detection with adaptive regulation is built in this paper. First, a heuristic dimensionality reduction algorithm based on unsupervised clustering is proposed. This algorithm uses the correlation between features to select the best subset. Then, a hybrid partitioning strategy is introduced in the negative selection algorithm (NSA), which divides the feature space into a grid based on the sample distribution density and generates specific candidate detectors in the boundary grid to effectively mitigate the holes caused by boundary diversity. Finally, the NSA is improved by self-set clustering and a novel gray wolf optimizer to achieve adaptive adjustment of the detector radius and position. The results show that the proposed NSA algorithm based on mixed hierarchical division and gray wolf optimization (MDGWO-NSA) achieves a higher detection rate, lower false alarm rate, and better generation quality than other network anomaly detection algorithms.

Screening potential lncRNA biomarkers for breast cancer and colorectal cancer combining random walk and logistic matrix factorization.

Breast cancer and colorectal cancer are two of the most common malignant tumors worldwide. They cause the leading causes of cancer mortality. Many researches have demonstrated that long noncoding RNAs (lncRNAs) have close linkages with the occurrence and development of the two cancers. Therefore, it is essential to design an effective way to identify potential lncRNA biomarkers for them. In this study, we developed a computational method (LDA-RWLMF) by integrating random walk with restart and Logistic Matrix Factorization to investigate the roles of lncRNA biomarkers in the prognosis and diagnosis of the two cancers. We first fuse disease semantic and Gaussian association profile similarities and lncRNA functional and Gaussian association profile similarities. Second, we design a negative selection algorithm to extract negative LncRNA-Disease Associations (LDA) based on random walk. Third, we develop a logistic matrix factorization model to predict possible LDAs. We compare our proposed LDA-RWLMF method with four classical LDA prediction methods, that is, LNCSIM1, LNCSIM2, ILNCSIM, and IDSSIM. The results from 5-fold cross validation on the MNDR dataset show that LDA-RWLMF computes the best AUC value of 0.9312, outperforming the above four LDA prediction methods. Finally, we rank all lncRNA biomarkers for the two cancers after determining the performance of LDA-RWLMF, respectively. We find that 48 and 50 lncRNAs have the highest association scores with breast cancer and colorectal cancer among all lncRNAs known to associate with them on the MNDR dataset, respectively. We predict that lncRNAs HULC and HAR1A could be separately potential biomarkers for breast cancer and colorectal cancer and need to biomedical experimental validation.

Ensemble Voting based Intrusion Detection Technique using Negative Selection Algorithm

This paper proposes an Intrusion Detection Technique (IDT) using an Artificial Immune System (AIS) based on Negative Selection Algorithm (NSA) to distinguish the self and non-self (intrusion) in computer networks. The novelties of the work are 1) use of Stacked Autoencoders (SAEs) and random forest for dimensionality reduction of data, 2) use of AIS to exploit its feature like self-learning, distributed, self-adaption, self-regulation with self and non-self-distinguishing capability, 3) implementation of two algorithms i.e., NSA based on Cosine distance (NSA_CD) and NSA based on Pearson Distance (NSA_PD) to explore their intrusion detection capabilities, and iv) development of a new ensemble voting based Intrusion Detection Technique (IDT-NSAEV) to detect and test the anomalies in the system. The proposed IDT-NSAEV technique combines the power of NSA_CD, NSA_PD and NSA based on Euclidean distance (NSA_ED) algorithms to enhance the detection rate by reducing the false alarm rate. The performance of the proposed technique is tested on standard benchmark NSL-KDD dataset and the results are compared with the state-of-the-art techniques. The results are in the favour of the proposed technique.

Sampled Value Attack Detection for Busbar Differential Protection Based on a Negative Selection Immune System

Considering a variety of sampled value (SV) attacks on busbar differential protection (BDP) which poses challenges to conventional learning algorithms, an algorithm to detect SV attacks based on the immune system of negative selection is developed in this paper. The healthy SV data of BDP are defined as self-data composed of spheres of the same size, whereas the SV attack data, i.e., the nonself data, are preserved in the nonself space covered by spherical detectors of different sizes. To avoid the confusion between busbar faults and SV attacks, a self-shape optimization algorithm is introduced, and the improved self-data are verified through a power-frequency fault-component-based differential protection criterion to avoid false negatives. Based on the difficulty of boundary coverage in traditional negative selection algorithms, a self-data-driven detector generation algorithm is proposed to enhance the detector coverage. A testbed of differential protection for a 110 kV double busbar system is then established. Typical SV attacks of BDP such as amplitude and current phase tampering, fault replays, and the disconnection of the secondary circuits of current transformers are considered, and the delays of differential relay operation caused by detection algorithms are investigated.

Adaptive Dendritic Cell-Negative Selection Method for Earthquake Prediction

Earthquake prediction (EQP) is an extremely difficult task, which has been overcome by adopting various technologies, with no further transformation so far. The negative selection algorithm (NSA) is an artificial intelligence method based on the biological immune system. It is widely used in anomaly detection due to its advantages of requiring little normal data to detect anomalies, including historical seismic-events-based EQP. However, NSA can suffer from the undesirable effect of data drift, resulting in outdated normal patterns learned from data. To tackle this problem, the data changes must be detected and processed, stimulating fast algorithmic adaptation strategies. This study proposes a dendritic cell algorithm (DCA)-based adaptive learning method for drift detection and negative selection algorithm (DC-NSA) that dynamically adapts to new input data. First, this study adopts the Gutenberg–Richter (GR) law and other earthquake distribution laws to preprocess input data. Then, the NSA is employed for EQP, and then, the dendritic cell algorithm (DCA) is employed to detect changes to trigger gradient descent strategies and update the self-set in NSA. Finally, the proposed approach is implemented to predict the earthquakes of MW > 5 in Sichuan and the surroundings during the next month. The experimental results demonstrate that our proposed DC-NSA is superior to the existing state-of-the-art EQP approaches.

Hybrid Approach for Automated Test Data Generation

Software testing has long been thought to be a good technique to improve the software quality and reliability. Path testing is the most reliable software testing technique and the key method for improving software quality among all testing approaches. On the other hand, test data quality has a big impact on the software testing activity’s ability to detect errors or defects. To solving testing problem, one must locate the entire search space for the relevant input data to encompass the different paths in the testable program. To satisfy path coverage, it is vital test to look at the accumulated test data across the thorough search area. A new approach based on ant colony optimization and negative selection algorithm (HACO-NSA) is presented in this research which overcome the flaws associated with search-based test data by generated automated test data. The optimum path testing objective is to generate appropriate test data to maximise coverage and to enhance the test data’s efficacy, as a result, the test data’s adequacy is validated using a path-based fitness function. In the NSA generation stage, the suggested method alters the new detectors creation using ACO. The proposed approach is evaluated for metrics such as average coverage, average generation, average time, and success rate and comparison has been done with random testing, ant colony optimization and negative selection algorithm Different benchmark programs have been used for object-oriented system. The findings show that the hybrid methodology escalates the coverage percentage and curtail test data size, reduces the redundancy in data and enhances the efficiency. The proposed approach is follows IEEE 829-2008 test documentation in entire testing process.

Automatic Test Data Generation for Basis Path Testing

Objectives: This paper presents a new hybrid ACO-NSA algorithm for the automatic test data generation problem with path coverage as an objective function. Method: In it, at the first instance, test data (detectors) are generated with the ant colony optimization algorithm (ACO), and then the generated data set (detector set) has been refined by a negative selection algorithm (NSA) with Hamming distance. Findings: The algorithm’s performance is tested on several benchmark problems with different data types and variables for metrics average coverage, average generations, average time and success rate, Iteration value 1000 is set for average coverage, average generations, average time and 200 for success rate. The obtained results from the proposed approach are compared with some existing approaches. The results are very efficient with high efficacy, higher path coverage, minimal data redundancy, and less execution time. Applications: This approach can be applied in any type of software development process in software engineering to reduce the testing efforts. Novelty: The approach is based on two distinct methodologies: metaheuristic search and artificial immune search, and its fitness is measured using path coverage as the fitness function. The approach provides 99.5% average path coverage, 2.72% average number of generations in 0.07 ns, and 99.9% success rate, which is significantly better than comparable approaches. Keywords: Test data generation; Metaheuristic search; Artificial immune search; Ant colony optimization; Negative selection algorithm; Path coverage

A generic cyber immune framework for anomaly detection using artificial immune systems

Intrusion detection systems play a significant role in computer security. Artificial immune systems are the prime contender in developing an anomaly-based intrusion detection system due to their simplicity. The fundamental goal of this paper is to create a generic framework for an artificial immune system which is fast and accurate in detecting anomalies using artificial immune system concepts. Natural killer cells in the immune system and their quick response to foreign pathogens inspired the adaptation of those cells into an artificial immune system based framework. A natural killer cell-based framework is proposed to improve the accuracy and speed of anomaly detection. The structure of the proposed framework includes major histocompatibility complex class 1 representation, affinity calculation, cell generation, and cell proliferation. This framework addresses the overlapping and hole problem while creating natural killer cells to increase the system’s performance. The negative selection algorithm and the positive selection algorithm generate the cells that enhance the anomaly detection technique and give high precision. The parameter response time introduced in this paper is crucial for an intrusion system to be used in real-time.