Normal Behavior Patterns Research Articles

Integrating Sensor Embeddings with Variant Transformer Graph Networks for Enhanced Anomaly Detection in Multi-Source Data

With the rapid development of sensor technology, the anomaly detection of multi-source time series data becomes more and more important. Traditional anomaly detection methods deal with the temporal and spatial information in the data independently, and fail to make full use of the potential of spatio-temporal information. To address this issue, this paper proposes a novel integration method that combines sensor embeddings and temporal representation networks, effectively exploiting spatio-temporal dynamics. In addition, the graph neural network is introduced to skillfully simulate the complexity of multi-source heterogeneous data. By applying a dual loss function—consisting of a reconstruction loss and a prediction loss—we further improve the accuracy of anomaly detection. This strategy not only promotes the ability to learn normal behavior patterns from historical data, but also significantly improves the predictive ability of the model, making anomaly detection more accurate. Experimental results on four multi-source sensor datasets show that our proposed method performs better than the existing models. In addition, our approach enhances the ability to interpret anomaly detection by analyzing the sensors associated with the detected anomalies.

Abnormal Behavior Recognition Based on 3D Dense Connections.

Abnormal behavior recognition is an important technology used to detect and identify activities or events that deviate from normal behavior patterns. It has wide applications in various fields such as network security, financial fraud detection, and video surveillance. In recent years, Deep Convolution Networks (ConvNets) have been widely applied in abnormal behavior recognition algorithms and have achieved significant results. However, existing abnormal behavior detection algorithms mainly focus on improving the accuracy of the algorithms and have not explored the real-time nature of abnormal behavior recognition. This is crucial to quickly identify abnormal behavior in public places and improve urban public safety. Therefore, this paper proposes an abnormal behavior recognition algorithm based on three-dimensional (3D) dense connections. The proposed algorithm uses a multi-instance learning strategy to classify various types of abnormal behaviors, and employs dense connection modules and soft-threshold attention mechanisms to reduce the model's parameter count and enhance network computational efficiency. Finally, redundant information in the sequence is reduced by attention allocation to mitigate its negative impact on recognition results. Experimental verification shows that our method achieves a recognition accuracy of 95.61% on the UCF-crime dataset. Comparative experiments demonstrate that our model has strong performance in terms of recognition accuracy and speed.

Mining Abnormal Patterns in Moving Target Trajectories Based on Multi-Attribute Classification

As a type of time series data, trajectory data objectively record the location information and corresponding time information of an object’s activities. It not only describes the spatial activity trajectory of a moving object but also contains the unique attributes, states, and behavioral characteristics of the moving object itself. It can also reflect the interaction relationship between the object’s activities and various elements in the environment to a certain extent. Therefore, mining from moving target trajectory data to discover implicit, effective, and potentially useful spatiotemporal behavior patterns of moving targets, such as anomaly detection, will have significant research significance. This paper proposes a method for mining abnormal patterns in the trajectory of moving targets based on multi-attribute classification. Firstly, to explore the activity location patterns of single moving targets, a frequent sequence discovery method for moving targets based on sequence patterns is proposed. Furthermore, for moving target trajectory data sets containing multiple attributes, numerical attributes are extracted, and the data are clustered according to attribute classification to extract a set of normal behavior patterns of moving targets. Then, combining the activity location patterns and normal behavior patterns of the moving target, the original trajectory data are compared with them to achieve the goal of detecting abnormal behavior of the moving target. Finally, an incremental anomaly detection scheme is proposed to address the characteristics of fast updates and large numbers of data in trajectory data sets. This involves synchronously updating the frequency of moving target activity patterns and the range of values for normal behavior patterns while updating the trajectory data set, in order to meet the needs of database updates and improve the accuracy and credibility of results.

A Critical Analysis of Machine Learning Techniques for Online Transaction Security

The widespread adoption of Unified Payments Interface (UPI) as the preferred digital transaction method has led to a sharp increase in fraudulent activities, presenting a considerable challenge to the security and reliability of online payments.Top of Form To address this pressing concern, our project focuses on developing a sophisticated fraud detection system tailored specifically for UPI transactions. We utilize Convolutional Neural Networks (CNNs), an advanced deep learning architecture, to intricately capture the sequence of operations involved in UPI transaction processing. Through extensive data collection and preprocessing, we curate a diverse dataset comprising various transaction types, amounts, and temporal patterns, ensuring comprehensive coverage of transaction behaviour. The CNN is meticulously trained on this dataset, leveraging its innate ability to discern subtle patterns and dependencies within transaction sequences. Notably, our approach emphasizes the importance of minimizing false positives to prevent genuine transactions from being erroneously flagged as fraudulent. In real-time, incoming UPI transactions are subjected to scrutiny by the trained CNN, which evaluates their conformity to learned normal behaviour patterns. Transactions exhibiting deviations indicative of potential fraud are promptly flagged for further investigation, thus enabling proactive mitigation of fraudulent activities. Most importantly, our study contains a comprehensive comparative analysis that compares the efficacy of conventional machine learning methods such as Support Vector Machines (SVM), Random Forests, and Logistic Regression with CNN-based fraud detection. Through meticulous evaluation, we demonstrate the superior efficacy of CNNs in detecting fraudulent UPI transactions, owing to their inherent capacity to capture intricate spatial and temporal dependencies. Our findings underscore the transformative potential of deep learning techniques in bolstering the security and resilience of digital payment ecosystems, thereby safeguarding users and fostering trust in online transactions.

Analysis of a Fuzzy Based Intrusion Detection System in Wireless Ad Hoc Networks

Technology and its growth is considerably enormous. This massive growth allows the opening of new fields of application in the domain of wireless networking and mobile ad-hoc networks (MANET) is one of its kinds. Mobile ad hoc network is widely used from collaborative computing to time critical applications in indoor and outdoor environment. Mobility of ad hoc network makes very attractive in all areas of mobile applications. Nodes participating in mobile ad hoc networks are autonomous, self-configurable and act as a router as well. These types of networks are dynamic in nature and have sort life time. Dynamic nature and limitations of the wireless transmission medium make MANET unsecured and vulnerable to various attacks. It is very tough to implement security for this networks and it opens up doors for further research work on this area. There is a great scope for designing a system to identify attacks and take countermeasures to minimize it and keep the performance of the network within acceptable limits. Intrusion detection system is one of its kinds and considered as security mechanism for MANET. This thesis explores different types of intrusion detection system like misuse detection and anomaly detection for mobile ad hoc networks. Anomaly detection techniques for ad hoc networks depend on the characterization of normal behavior pattern of wireless nodes. This research work focuses on wireless node behavior based detection technique. Most of anomaly intrusion detection systems are focusing on upper layer traffic to a profile normal behavior of wireless node. This research work focus on media access control (MAC) layer and network layer of wireless node. It is inefficient to use a large feature set of MAC layer and network layer due to energy limitation in ad-hoc network. A minimal feature set from MAC layer and network layer were proposed. This research work proposed an anomaly intrusion detection system for mobile ad hoc network using fuzzy logic and weighted average method. The network performance of mobile ad hoc network with intrusion detection system was analyzed using various network parameters. Further the performance of the performance of the intrusion detection system was analyzed using detection rate and false alarms. Results show good improvement in detection rate and other performance metrics

A novel credit card fraud detection using supervised machine learning model

Financial fraud, especially in credit card transactions, is a growing concern. To tackle this, data mining techniques are used to automatically analyze large and complex financial datasets. Detecting credit card fraud is tricky because the patterns of normal and fraudulent behavior keep changing, and the data about fraud is much less common compared to legitimate transactions Several techniques were tried on a dataset from European cardholders, including Decision Tree, Random Forest, SVC, XGBoost, K-Nearest Neighbors, and Logistic Regress The dataset had information from 284,786 credit card transactions. To address the challenges, six advanced data mining approaches (Logistic Regression, K-Nearest Neighbors, Support Vector Classifiers, Decision Tree, Random Forests, and XGBoost) are evaluated. A comparative analysis is conducted to identify the best-performing model.

Anomaly detection system for network transport with machine learning approach

The rapid growth of network infrastructures and the increasing volume of data transmitted through them have led to a critical need for efficient and accurate anomaly detection systems in network transport. This paper proposes a novel anomaly detection system that utilizes machine learning techniques to identify abnormal patterns and deviations in network traffic. The proposed system follows a multi-layered approach, starting with the collection of network traffic data from various sources, including routers, switches, and gateways. The data is then preprocessed to extract relevant features and eliminate noise. Feature extraction is carried out using statistical, time-series, and flow-based analysis to capture the inherent characteristics of network communication. Machine learning algorithms, such as neural networks, or in this case, auto-encoders, will be trained to learn the patterns of normal network behavior and subsequently detect deviations from these patterns as anomalies. The system provides alerts and notifications to network administrators, allowing prompt investigation and response to potential security threats or network performance issues. It effectively differentiates between benign and malicious network activities, enabling network administrators to take proactive measures to secure their infrastructure and ensure uninterrupted communication.

Detection of cyber attacks on IoT based cyber physical systems

The integration of Internet of Things (IoT) devices in Cyber- Physical Systems (CPS) continues to proliferate, ensuring the security of these interconnected systems becomes paramount. In existing research work focuses on the development and implementation of a cyber attack detection system for IoT-based CPS, leveraging Support Vector Machine (SVM) models. The SVM model, known for its effectiveness in binary classification tasks, is trained on historical data to distinguish between normal and malicious behavior patterns exhibited by IoT devices within the CPS. The SVM model is trained to learn the normal behavior of the system, enabling it to identify deviations indicative of cyber attacks. Realworld experiments and simulations demonstrate the efficacy of the SVMbased detection system in identifying various types of cyber threats. However, this research also acknowledges certain limitations. The SVM model's performance may be impacted by the dynamic and evolving nature of cyber threats, as it relies heavily on historical data for training and detection accuracy issues. To address the limitations of present cyber threat detection model , in this research work proposed a novel deep learning based CNN Model. The proposed model improve cyber attacks detection and performance metrics. The proposed model outperforms with the comparison of previous model. The performance measured in terms of accuracy, precision, recall and f1-score.

Real-Time Intelligent Anomaly Detection and Prevention System

Real-time anomaly detection in network traffic is a method that detects unexpected and anomalous behaviour by identifying normal behaviour and statistical patterns in network traffic data. This method is used to detect potential attacks or other anomalous conditions in network traffic. Real-time anomaly detection uses different algorithms to detect abnormal activities in network traffic. These include statistical methods, machine learning and deep learning techniques. By learning the normal behaviour of network traffic, these methods can detect unexpected and anomalous situations. Attackers use various techniques to mimic normal patterns in network traffic, making it difficult to detect. Real-time anomaly detection allows network administrators to detect attacks faster and respond more effectively. Real-time anomaly detection can improve network performance by detecting abnormal conditions in network traffic. Abnormal traffic can overuse the network's resources and cause the network to slow down. Real-time anomaly detection detects abnormal traffic conditions, allowing network resources to be used more effectively. In this study, blockchain technology and machine learning algorithms are combined to propose a real-time prevention model that can detect anomalies in network traffic.

A Predictive Model Using Deep Learning Neural Network for Efficient Intrusion Detection

Abstract: Network intrusion detection system helps to detect exploitations and mitigate damages. A network intrusion detection system detects the network traffic that deviates from the normal behavioral pattern. Developing an efficient intrusion detection system has many challenges and the patterns associated with one type of intrusion differ from other intrusions. In such situations, understanding different patterns and differentiating intrusions becomes essential to detect anomalies and attacks in the network. Deep learning models offer more power and intelligence to the detection system and extend the ability to differentiate & understand the network feature characteristics, also machine learning models with feature selection showed high performance in intrusion detections. This paper evaluates the proposed deep learning neural network model and machine learning models using feature selection for efficient intrusion detection using real world dataset.

Wind Turbine Gearbox Condition Monitoring Using Hybrid Attentions and Spatio-Temporal BiConvLSTM Network

Gearbox fault deterioration can significantly impact the safety, reliability, and efficiency of wind turbines, resulting in substantial economic losses for wind farms. However, current condition monitoring methods face challenges in effectively mining the hidden spatio-temporal features within SCADA data and establishing reasonable weight allocations for model input variables. To tackle these issues, we proposed a novel condition monitoring method for wind turbine gearboxes called HBCE, which integrated a feature-time hybrid attention mechanism (HA), the bidirectional convolutional long short-term memory networks (BiConvLSTM), and an improved exponentially weighted moving-average (iEWMA). Specifically, utilizing historical health SCADA data acquired through the modified Thompson tau data-cleaning algorithm, a normal behavior model (HA-BiConvLSTM) of gearbox was constructed to effectively extract the spatio-temporal features and learn normal behavior patterns. An iEWMA-based outlier detection approach was employed to set dynamic adaptive thresholds, and real-time monitor the prediction residuals of HA-BiConvLSTM to identify the early faults of gearbox. The proposed HBCE method was validated through actual gearbox faults and compared with conventional spatio-temporal models (i.e., CNN-LSTM and CNN&LSTM). The results illustrated that the constructed HA-BiConvLSTM model achieved superior prediction precision in terms of RMSE, MAE, MAPE, and R2, and the proposed method HBCE can effectively and reliably identify early anomalies of a wind turbine gearbox in advance.

Fingolimod Modulates the Gene Expression of Proteins Engaged in Inflammation and Amyloid-Beta Metabolism and Improves Exploratory and Anxiety-Like Behavior in Obese Mice.

Obesity is considered a risk factor for type 2 diabetes mellitus, which has become one of the most important health problems, and is also linked with memory and executive function decline. Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid that regulates cell death/survival and the inflammatory response via its specific receptors (S1PRs). Since the role of S1P and S1PRs in obesity is rather obscure, we examined the effect of fingolimod (an S1PR modulator) on the expression profile of genes encoding S1PRs, sphingosine kinase 1 (Sphk1), proteins engaged in amyloid-beta (Aβ) generation (ADAM10, BACE1, PSEN2), GSK3β, proapoptotic Bax, and proinflammatory cytokines in the cortex and hippocampus of obese/prediabetic mouse brains. In addition, we observed behavioral changes. Our results revealed significantly elevated mRNA levels of Bace1, Psen2, Gsk3b, Sphk1, Bax, and proinflammatory cytokines, which were accompanied by downregulation of S1pr1 and sirtuin 1 in obese mice. Moreover, locomotor activity, spatially guided exploratory behavior, and object recognition were impaired. Simultaneously, fingolimod reversed alterations in the expressions of the cytokines, Bace1, Psen2, and Gsk3b that occurred in the brain, elevated S1pr3 mRNA levels, restored normal cognition-related behavior patterns, and exerted anxiolytic effects. The improvement in episodic and recognition memory observed in this animal model of obesity may suggest a beneficial effect of fingolimod on central nervous system function.

Review on Malware Classification with a Hybrid Deep Learning

This study introduces advanced methodology for classifying malware by leveraging hybrid deep learning algorithms. The research presents a pioneering framework that seamlessly integrates Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) models to deliver a robust malware classification approach. The primary objective is to effectively differentiate between normal behavioral patterns and malicious network data. The efficacy of this innovative approach is evaluated by comparing it with conventional machine learning techniques like Support Vector Machines (SVM). Through this comparative analysis, the investigation aims to uncover the unique strengths and potential limitations of the proposed method, with the intention of establishing it as a superior alternative to current malware classification methods. By harnessing the individual capabilities of CNN and LSTM models, the proposed framework achieves a higher level of accuracy in identifying and categorizing malware compared to existing approaches. While CNN models excel in feature extraction from raw data, LSTM models exhibit proficiency in understanding sequential patterns. By synergizing these models, the resulting framework demonstrates significantly improved performance in classifying malware. The empirical assessment strongly suggests that the newly proposed framework is poised to outperform established techniques. These research findings hold great promise in advancing the development of more efficient systems for detecting and preventing malware.

PIGNUS: A Deep Learning model for IDS in industrial internet-of-things

The heterogeneous nature of the Industrial Internet of Thing (IIoT) has a considerable impact on the development of an effective Intrusion Detection System (IDS). The proliferation of linked devices results in multiple inputs from industrial sensors. IDS faces challenges in analyzing the features of the traffic and identifying anonymous behavior. Due to the unavailability of a comprehensive feature mapping method, the present IDS solutions are non-usable to identify zero-day vulnerabilities.In this paper, we introduce the first comprehensive IDS framework that combines an efficient feature-mapping technique and cascading model to solve the above-mentioned problems. We call our proposed solution deeP learnIG model intrusioN detection in indUStrial internet-of things (PIGNUS). PIGNUS integrates Auto Encoders (AE) to select optimal features and Cascade Forward Back Propagation Neural Network (CFBPNN) for classification and attack detection. The cascading model uses interconnected links from the initial layer to the output layer and determines the normal and abnormal behavior patterns and produces a perfect classification. We execute a set of experiments on five popular IIoT datasets: gas pipeline, water storage tank, NSLKDD+, UNSW-NB15, and X-IIoTID. We compare PIGNUS to the state-of-the-art models in terms of accuracy, False Positive Ratio (FPR), precision, and recall. The results show that PIGNUS provides more than 95% accuracy, which is 25% better on average than the existing models. In the other parameters, PIGNUS shows 20% improved FPR, 10% better recall, and 10% better in precision. Overall, PIGNUS proves its efficiency as an IDS solution for IIoTs. Thus, PIGNUS is an efficient solution for IIoTs.

A Novel Intrusion Detection System Using Multiple Linear Regression

The internet is no doubt the biggest and the most important tool of modern civilisation. But along with its numerous benefits, it also comes with its own set of risks, the most important of them being breaches in security and privacy. An anomaly-based Intrusion Detection System (IDS) is a type of security system that is used to detect and alert on unusual or abnormal behaviour that may indicate an attack or intrusion. Unlike signature-based IDS, which rely on known patterns of attack, anomaly-based IDS is designed to detect previously unseen or unknown attacks by identifying deviations from normal patterns of behaviour. Multiple linear regression is a statistical technique used to analyse the relationship between a dependent variable and multiple independent variables. In this technique, a linear equation is established between the dependent variable and multiple independent variables, with the aim of predicting the value of the dependent attribute for a given set of values of the independent attribute. In this paper, we collected a data set of 125974 entries and 42 attributes from Kaggle, pre-processed the data and used logistic regression to predict the dependent variable (called xAttack) using 25 independent variables, as we found a high correlation between the aforementioned variables The results are simulated using 10-fold cross validation, using various train test splits of the data set. The data has been split into 80-20,50-50, and 66-34. After testing the given data set in different train test splits, an accuracy of 92.73 was achieved.

Assessing Dominant-Submissive Behavior in Adult Rats Following Traumatic Brain Injury

Competition over resources such as food, territory, and mates significantly influences relationships within animal species and is mediated through social hierarchies that are often based on dominant-submissive relationships. The dominant-submissive relationship is a normal behavioral pattern among the individuals of a species. Traumatic brain injury is a frequent cause of social interaction impairment and the reorganization of dominant-submissive relationships in animal pairs. This protocol describes submissive behavior in adult male Sprague-Dawley rats after the induction of traumatic brain injury using a fluid-percussion model compared to naive rats through a series of dominant-submissive tests performed between 29 days and 33 days after induction. The dominant-submissive behavior test shows how brain injury can induce submissive behavior in animals competing for food. After traumatic brain injury, the rodents were more submissive, as indicated by them spending less time at the feeder and being less likely to arrive first at the trough compared to the control animals. According to this protocol, submissive behavior develops after traumatic brain injury in adult male rats.

Using Visual Observations to Compare the Behavior of Previously Immobilized and Non-Immobilized Wild Polar Bears

During 17 field seasons between 1973 and 1999, we conducted a long-term study of the behavior of undisturbed wild polar bears in Radstock Bay, southwest Devon Island, Nunavut. In a subset of 11 seasons (6 spring and 5 summer) between 1975 and 1997, we used three different drug combinations to chemically immobilize a small number of adult and subadult polar bears on an opportunistic basis and applied a temporary dye mark so that individual bears could be visually reidentified. We then used multinomial logistic regression to compare the behavior of 35 previously immobilized bears of five different demographic classes (sex, age, and reproductive status) to the behavior of non-immobilized bears of the same demographic classes in the same years and seasons. During the first two days after immobilization, bears slept significantly more and spent less time hunting than did bears that had not been immobilized. However, previously immobilized bears returned to the same behavioral patterns and proportion of total time spent hunting as non-immobilized bears within two days and no further negative behavioral effects were detected in the following 21 d. We visually confirmed successful hunting by three adult bears within 0.4 to 2.1 d of being immobilized, all of which went on to make additional kills within the following 24 h. The return to normal behavior patterns, including the ability to hunt successfully, within 48 h of immobilization appears consistent with the hypothesis that polar bears do not experience longer-term behavioral effects following brief chemical immobilization for conservation and management purposes.

Extracellular Matrix Interactions Provide Tumor Cells With an Escape Mechanism for Commitment to Differentiation

Extracellular Matrix Interactions Provide Tumor Cells With an Escape Mechanism for Commitment to Differentiation

The Brilliance of the Zebrafish Model: Perception on Behavior and Alzheimer's Disease.

Alzheimer’s disease (AD) has become increasingly prevalent in the elderly population across the world. It’s pathophysiological markers such as overproduction along with the accumulation of amyloid beta (Aβ) plaques and neurofibrillary tangles (NFT) are posing a serious challenge to novel drug development processes. A model which simulates the human neurodegenerative mechanism will be beneficial for rapid screening of potential drug candidates. Due to the comparable neurological network with humans, zebrafish has emerged as a promising AD model. This model has been thoroughly validated through research in aspects of neuronal pathways analogous to the human brain. The cholinergic, glutamatergic, and GABAergic pathways, which play a role in the manifested behavior of the zebrafish, are well defined. There are several behavioral models in both adult zebrafish and larvae to establish various aspects of cognitive impairment including spatial memory, associative memory, anxiety, and other such features that are manifested in AD. The zebrafish model eliminates the shortcomings of previously recognized mammalian models, in terms of expense, extensive assessment durations, and the complexity of imaging the brain to test the efficacy of therapeutic interventions. This review highlights the various models that analyze the changes in the normal behavioral patterns of the zebrafish when exposed to AD inducing agents. The mechanistic pathway adopted by drugs and novel therapeutic strategies can be explored via these behavioral models and their efficacy to slow the progression of AD can be evaluated.

A REVIEW ON SESBANIA GRANDIFLORA ITS EFFECT ON THE TREATMENT OF DEMENTIA AND NEURO DISEASE AND OTHER DISEASES

One of the most often used folk medicines is Sesbania grandiflora, better known as the Agati or Hummingbird tree. Anemia, bacterial infections, TB, and other illnesses can be treated with the plant's pharmacological properties. Agati's anticancer and hepatoprotective properties are attributable to the presence of phytochemical ingredients. Sesbania grandiflora's therapeutic characteristics are the topic of this article. Normal psychological functioning and behavioural patterns are disrupted by dementia. One to four percent of the world's population over the age of 65 is affected by dementia, which is a neurological disease. Dementia is caused by a number of different factors. Because dementia is more common in those who have cardiovascular and neurodegenerative conditions (such hypertension, hypercholesterolemia, diabetes and cerebral ischemia), these people are more likely to develop it. In addition to alcohol and tobacco use, these conditions are exacerbated. Alzheimer's dementia, vascular dementia, frontotemporal dementia, semantic dementia, and dementia with Lewy bodies are the most frequent kinds of dementia. As we get older, Alzheimer's and Lewy body dementias become more common. It is the second most frequent kind of dementia after Alzheimer's disease (AD), and it can be caused by long-term cardiovascular disease (CVD). Deterioration in semantic memory is a hallmark of semantic dementia (knowledge of objects, people, concepts and words). synuclein aggregation in Lewy body dementia (LBD) impairs neuronal growth and has a significant impact on the disease aetiology. Alzheimer's disease, Parkinson's disease, schizophrenia, and other neurodegenerative diseases all have a strong genetic component, as do frontotemporal dementia and Alzheimer's disease. An important part of this study is the discussion of the role of glutathione (GSH), SOD, GPx, and Catalase (Cat) during oxidative stress and the link between mitochondrial dysfunction and many disorders, including dementia. The purpose of this review is to examine the types, risk factors, and underlying mechanisms of dementia, as well as new therapy techniques, in order to maximise the chances of finding new and effective therapeutic strategies