Suspicious Behavior Research Articles

Deep Learning Approach for Suspicious Activity Detection from Video Surveillance

Abstract: Video surveillance is getting more important as technology improves. New systems use artificial intelligence to tell the difference between normal and suspicious behaviour in live video. Since people can be unpredictable, it's hard to know for sure if something is suspicious. This system uses deep learning to identify suspicious activity in schools, like classrooms or hallways. If the system sees something suspicious, it can alert security or the police. Here's how it works: The system breaks down the video into individual pictures, one after another. Then, it uses a special program to decide if what's happening in the pictures is suspicious or normal.

Social Grounds of Pickpocket in Metropolitan City: Attitude and Aftermath Effect among the Sufferer

The study delves into the complex phenomenon of pickpocketing, a global affliction that persists across nations like India, posing a familiar threat to individuals in crowded spaces. Despite its historical roots as an opportunistic crime, pickpocketing has evolved into an organized enterprise, often involving economically disadvantaged individuals forming groups. While prevalent in major cities worldwide, pickpocketing lacks explicit legal recognition in India's Penal Code, falling under broader theft laws. Leveraging data from automated fare collection systems and trajectory analysis, this research refines anomaly-detection applications to identify suspicious behaviors, shedding light on the environmental factors shaping pickpocketing trends. The study focuses on the social considerations of pickpocketing in Kolkata, India, exploring victim experiences and perceptions. Findings reveal the broader social impacts of pickpocketing, beyond just financial loss, with victims enduring significant social and psychological distress. Despite lacking specific legal penalties, pickpocketing remains a prevalent issue, highlighting the need for collective efforts from governments, law enforcement, and the public to combat it effectively.

Active Chat Monitoring and Suspicious Chat Detection Over Internet

In today's digital landscape, the proliferation of online communication platforms has brought about unprecedented opportunities for interaction and collaboration. However, this vast virtual space also presents challenges in ensuring safety and security, particularly concerning the monitoring of chat activities and the detection of suspicious behaviour. This research paper investigates the methods and technologies employed proactive chat monitoring and random chat across the web. The paper begins by examining the importance of monitoring chat platforms to prevent various forms of malicious activities, including cyberbullying, fraud, terrorism, and other illicit behaviours. It then delves into the techniques utilized for real-time monitoring of chat conversations, such as keyword filtering, sentiment analysis, and natural language processing algorithms. Furthermore, the paper explores the principles behind suspicious chat detection, highlighting the indicators of potentially harmful or illegal activities within chat communications. This includes analysing patterns of conversation, identifying anomalous behaviour, and integrating machine learning models to improve detection accuracy. Moreover, the research discusses the ethical considerations and privacy concerns associated with active chat monitoring, emphasizing the need for transparent policies and safeguards to protect user privacy while ensuring effective security measures. Keywords: Active chat monitoring, Suspicious chat detection, Online communication platforms, Cybersecurity, Malicious activities, Keyword filtering, Sentiment analysis, Natural language processing algorithms.

Cookie Hijacking: Privacy Risk

Users may accept more cookies than they require because of suspicious behavior in cookie claimers.
 And, like, more than they even know. So, as we were measuring real behavior, we were also assessing
 each of these components' effectiveness, correct? Learning about user opinions of those cookie
 disclaimers and their entire process of deciding which cookies to accept or, like, reject, bro, was very
 fascinating. So, guess what? We have conclusively shown that the various images associated with
 the accept/reject option significantly influence the judgments made by consumers. You won't believe
 it, but we also discovered that assigning a label to the rejected option has a really big impact. Further-
 more, we have confirmed previous studies showing that, well, biased content doesn't actually have
 a significant impact on customers' judgments. To sum up, black patterns in cookie disclaimers are
 really important. Users are forced to accept more cookies than they really need, and decision-making
 is greatly influenced by images and labels. Getting folks to accept those cookies is the main goal.
 According to our research on user attitudes about cookie disclosures, the presence of the disclaimer
 only slightly affects the way that various user types make decisions. We provide advice on how to
 improve. Conditions that apply to different user groups.

REVOLUTIONIZING BANKING SECURITY: INTEGRATING ARTIFICIAL INTELLIGENCE, BLOCKCHAIN, AND BUSINESS INTELLIGENCE FOR ENHANCED CYBERSECURITY

This paper outlines the methodology and implementation strategies necessary to revolutionize banking security and ensure a resilient financial ecosystem. In the dynamic landscape of banking, security stands as a cornerstone for financial institutions. The rise of digital banking and the proliferation of online transactions have heightened the need for robust cybersecurity measures to protect sensitive financial data and ensure the integrity of transactions. Traditional security approaches are often reactive and struggle to keep pace with the evolving tactics of cybercriminals. Consequently, there is a pressing need for innovative solutions that can adapt to emerging threats in real-time. The integration of Artificial Intelligence (AI), Blockchain, and Business Intelligence (BI) offers a paradigm shift in banking security. AI, with its ability to analyze vast amounts of data and identify patterns indicative of suspicious behavior, serves as a proactive defense mechanism against cyber threats. By continuously monitoring transactions and network activities, AI-powered systems can swiftly detect anomalies and potential security breaches, enabling banks to respond effectively and mitigate risks before they escalate. Blockchain technology introduces a decentralized and immutable ledger that enhances the security and transparency of transactions. By utilizing cryptographic principles, Blockchain ensures that transactional data remains tamper-proof and verifiable, reducing the risk of fraud and unauthorized access. This technology not only secures financial transactions but also enables the secure sharing of data among stakeholders, facilitating seamless collaboration while maintaining data integrity. Business Intelligence (BI) complements AI and Blockchain by providing actionable insights derived from data analytics. BI tools enable banks to gain a deeper understanding of their security posture, identify vulnerabilities, and prioritize remediation efforts.
 Keywords: Banking, Security, AI, Integration, Blockchain.

Enhancing secure financial transactions through the synergy of blockchain and artificial intelligence

Blockchain and artificial intelligence are innovative technologies that can securely process and share data across unreliable networks. Due to data leakage from user information, it is critical to keep the data confidential and completely protected because criminals are looking for this information to attack the system or steal information in the banking sector. To address this issue, this paper suggests an Integrated Blockchain and Artificial intelligence (IBAI) Framework for secure financial transactions. A blockchain can store every customer's data in one place, while AI-driven algorithms can speedily examine that data and make an unbiased decision. The rising blockchain technology provides a decentralized architecture that enables the secure sharing of data and resources to the different networks and is promoted for removing centralized control and resolving the problems of AI. When suspicious behaviour happens, alerts can be triggered to avert theft. Furthermore, registration protocols based on AI are utilized to keep this data in comprehensive security. The numerical results show that the suggested IBAI model enhances the suspicious behaviour detection ratio and increases accuracy up to 98% compared to other models.

Deep-Learning-Based Action and Trajectory Analysis for Museum Security Videos

Recent advancements in deep learning and video analysis, combined with the efficiency of contemporary computational resources, have catalyzed the development of advanced real-time computational systems, significantly impacting various fields. This paper introduces a cutting-edge video analysis framework that was specifically designed to bolster security in museum environments. We elaborate on the proposed framework, which was evaluated and integrated into a real-time video analysis pipeline. Our research primarily focused on two innovative approaches: action recognition for identifying potential threats at the individual level and trajectory extraction for monitoring museum visitor movements, serving the dual purposes of security and visitor flow analysis. These approaches leverage a synergistic blend of deep learning models, particularly CNNs, and traditional computer vision techniques. Our experimental findings affirmed the high efficacy of our action recognition model in accurately distinguishing between normal and suspicious behaviors within video feeds. Moreover, our trajectory extraction method demonstrated commendable precision in tracking and analyzing visitor movements. The integration of deep learning techniques not only enhances the capability for automatic detection of malevolent actions but also establishes the trajectory extraction process as a robust and adaptable tool for various analytical endeavors beyond mere security applications.

Identification of suspicious behavior through anomalies in the tracking data of fishing vessels

Automated positioning devices can generate large datasets with information on the movement of humans, animals and objects, revealing patterns of movement, hot spots and overlaps among others. However, in the case of Automated Information Systems (AIS), attached to vessels, observed strange behaviors in the tracking datasets may come from intentional manipulation of the electronic devices. Thus, the analysis of anomalies can provide valuable information on suspicious behavior. Here, we analyze anomalies of fishing vessel trajectories obtained with the Automatic Identification System. The map of silent anomalies, those that occur when positioning data are absent for more than 24 hours, shows that they are most likely to occur closer to land, with 87.1% of anomalies observed within 100 km of the coast. This behavior suggests the potential of identifying silence anomalies as a proxy for illegal activities. With the increasing availability of high-resolution positioning of vessels and the development of powerful statistical analytical tools, we provide hints on the automatic detection of illegal activities that may help optimize the management of fishing resources.

The New Neighborhood Watch: An Exploratory Study of the Nextdoor App and Crime Narratives

Community members use the Nextdoor App to prevent crime and circulate information when suspicious activity or criminal misconduct is afoot. The Nextdoor App operates like other popular social media platforms, but unlike Facebook and Twitter, it connects citizens based on geography instead of areas of interest. One unique aspect of the app is posting events and perceptions of suspicious behavior and criminal acts. User posts can provide narratives on feelings, incidents, and perceptions of crime in designated neighborhoods. This exploratory study focuses on how community members in an urban Colorado area use the app as a high-tech Neighborhood Watch. A qualitative research approach with a thematic analysis is implemented to examine neighbors’ perceptions of crime events and community safety. The findings depict that citizens engaged in Nextdoor communications are frustrated with petty community incivilities, property crimes, and law enforcement actions. In addition, the results show that lessons in being a capable guardian, possible increases in fear levels, and surveillance activities are important aspects of understanding social media and crime.

Towards enhanced monitoring framework with smart predictions

Abstract Context: Predicting security and trust vulnerabilities and issues is crucial for IoT interconnected systems and ecosystems, especially when integrating new, third-party or open-source components. Objective: One way to ensure timely predictions is by using a smart monitoring framework to continuously verify functional and non-functional property violations during the executions of the systems and their components. Method: This paper presents a set of guidelines for the Smart Monitoring Framework definition and its application process. Results and Conclusion: The paper provides the reference architecture of the Smart Monitoring Framework and its possible implementation to promptly detect suspicious behavior or property violations. The paper also illustrates how the provided implementation satisfies the defined guidelines by design.

Biomimetic Deep Learning Networks With Applications to Epileptic Spasms and Seizure Prediction.

In this study, we present a novel biomimetic deep learning network for epileptic spasms and seizure prediction and compare its performance with state-of-the-art conventional machine learning models. Our proposed model incorporates modular Volterra kernel convolutional networks and bidirectional recurrent networks in combination with the phase amplitude cross-frequency coupling features derived from scalp EEG. They are applied to the standard CHB-MIT dataset containing focal epilepsy episodes as well as two other datasets from the Montefiore Medical Center and the University of California Los Angeles that provide data of patients experiencing infantile spasm (IS) syndrome. Overall, in this study, the networks can produce accurate predictions (100%) and significant detection latencies (10 min). Furthermore, the biomimetic network outperforms conventional ones by producing no false positives. Biomimetic neural networks utilize extensive knowledge about processing and learning in the electrical networks of the brain. Predicting seizures in adults can improve their quality of life. Epileptic spasms in infants are part of a particular seizure type that needs identifying when suspicious behaviors are noticed in babies. Predicting epileptic spasms within a given time frame (the prediction horizon) suggests their existence and allows an epileptologist to flag an EEG trace for future review.

Virtual Overseer for Examination Using Artificial Intelligence

The Virtual Overseer for Examination using AI is a proposed system that utilizes artificial intelligence (AI) to monitor and supervise online exams. The system aims to enhance the credibility and integrity of online exams by detecting and preventing cheating behaviors such as copying answers or using unauthorized materials. The proposed system utilizes computer vision and machine learning algorithms to analyze exam sessions in real time and flag suspicious behaviors for review by human proctors. The system also employs facial recognition technology to authenticate test-takers' identity and ensure that the right person is taking the exam. Overall, the Virtual Overseer for Examination using AI has the potential to revolutionize online education and assessment by providing a secure and reliable platform for conducting remote exams.

AI-Based Network Intrusion Detection System

Abstract: This report discusses the research done on the chosen topic, which is Developing an AI-based Network Intrusion Detection System using ML and DL algorithms. Recently we have seen so much progress in Internet and communication technologies it is not just connecting computer networks and people but it is also connecting devices involving Big Data. It has so many benefits in each field which are crucial in today's world like education, health, digital transactions, traveling, and anything we can think of. With so many benefits it comes with its negative effects like cyber-attacks which can happen to anybody who is connected to the Internet. So, Networks and Security become very desirable areas of research and work. To be saved from the attacks we have to detect the cyber-attack and stop the intruder from causing harm to our system. We use IDS (Intrusion Detection System), so we can identify incoming attacks. A Network Intrusion detection system provides security by constantly monitoring the network traffic for suspicious behavior. For building an Intrusion Detection System we need a good dataset with a huge amount of data which is of good quality and can be used for training the System so it can predict the output more accurately. In this paper we used NSL- KDD [6] data set a refined version of the KDD’99 dataset. We developed many models using machine learning and deep learning and compared them for detecting intrusion in networks. We used supervised machine learning and deep learning techniques to train and build many classification models that can differentiate between attacking traffic and normal traffic. We compared the accuracy of every model of different datasets so we can find the model that is performing best for network intrusion detection. After performing all the research and comparison we found that a fully connected Deep Learning model is giving better performance than a machine learning model. We used Autoencoder for feature selection for the best-performing Deep Learning Classification Model.

FiFrauD: Unsupervised Financial Fraud Detection in Dynamic Graph Streams

Given a stream of financial transactions between traders in an e-market, how can we accurately detect fraudulent traders and suspicious behaviors in real time? Despite the efforts made in detecting these fraudsters, this field still faces serious challenges, including the ineffectiveness of existing methods for the complex and streaming environment of e-markets. As a result, it is still difficult to quickly and accurately detect suspected traders and behavior patterns in real-time transactions, and it is still considered an open problem. To solve this problem and alleviate the existing challenges, in this article, we propose FiFrauD, which is an unsupervised, scalable approach that depicts the behavior of manipulators in a transaction stream. In this approach, real-time transactions between traders are converted into a stream of graphs and, instead of using supervised and semi-supervised learning methods, fraudulent traders are detected precisely by exploiting density signals in graphs. Specifically, we reveal the traits of fraudulent traders in the market and propose a novel metric from this perspective, i.e., graph topology, time, and behavior. Then, we search for suspicious blocks by greedily optimizing the proposed metric. Theoretical analysis demonstrates upper bounds for FiFrauD's effectiveness in catching suspicious trades. Extensive experiments on five real-world datasets with both actual and synthetic labels demonstrate that FiFrauD achieves significant accuracy improvements compared with state-of-the-art fraud detection methods. Also, it can find various suspicious behavior patterns in a linear runtime and provide interpretable results. Furthermore, FiFrauD is resistant to the camouflage tactics used by fraudulent traders.

A Two-Mode, Adaptive Security Framework for Smart Home Security Applications

With the growth of the Internet of Things (IoT), the number of cyber attacks on the Internet is on the rise. However, the resource-constrained nature of IoT devices and their networks makes many classical security systems ineffective or inapplicable. We introduce TWINKLE, a two-mode, adaptive security framework that allows an IoT network to be in regular mode for most of the time, which incurs a low resource consumption rate, and to switch to vigilant mode only when suspicious behavior is detected, which potentially incurs a higher overhead. Compared to the early version of this work, this article presents a more comprehensive design and architecture of TWINKLE, describes challenges and details in implementing TWINKLE, and reports evaluations of TWINKLE based on real-world IoT testbeds with more metrics. We show the efficacy of TWINKLE in two case studies where we examine two existing intrusion detection and prevention systems and transform both into new, improved systems using TWINKLE. Our evaluations show that TWINKLE is not only effective at securing resource-constrained IoT networks, but can also successfully detect and prevent attacks with a significantly lower overhead and detection latency than existing solutions.

Violence Detection using Deep Learning

In an era where technological advancements are reshaping various aspects of society, crime prevention and public safety have emerged as critical areas of concern.This project proposes a novel approach to crime analytics by harnessing the power of live footage and deep learning technologies.The invention of Crime Analytics Using Live Footage represents a groundbreaking advancement in the field of crime prevention and law enforcement. This technology harnesses the power of real-time video analysis, predictive analytics, and data-driven decision-making to proactively detect and prevent criminal activities, ensuring public safety and enhancing the efficiency of law enforcement efforts.The system's core functionality lies in its ability to analyze live video footage from a network of surveillance cameras, leveraging advanced computer vision algorithms and machine learning techniques. This real-time analysis enables the immediate identification of potential threats, suspicious behavior, and emerging criminal activities within monitored areas.

DoWNet—classification of Denial-of-Wallet attacks on serverless application traffic

Abstract Serverless computing is an ever-growing programming paradigm being adopted by developers all over the world. Its highly scalable, automatic load balancing, and pay for what you use design is a powerful tool that can also greatly reduce operational costs. However, these advantages also leave serverless computing open to a unique threat, Denial-of-Wallet (DoW). It is the intentional targeting of serverless function endpoints with request traffic in order to artificially raise the usage bills for the application owner. A subset of these attacks are leeches. They perform DoW at a rate that could go undetected as it is not a sudden violent influx of requests. We devise a means of detecting such attacks by utilizing a novel approach of representing request traffic as heat maps and training an image classification algorithm to distinguish between normal and malicious traffic behaviour. Our classifier utilizes convolutional neural networks and achieves 97.98% accuracy. We then design a system for the implementation of this model that would allow application owners to monitor their traffic in real time for suspicious behaviour.

Network Intrusion Detection and Prevention System Using Hybrid Machine Learning with Supervised Ensemble Stacking Model

Network intrusion detection systems play a critical role in protecting a variety of services ranging from economic through social to commerce. However, the growing level and sophistication of malicious attacks launched on networks in the current technological landscape have necessitated the need for advanced and robust detection mechanisms to mitigate against security breaches of confidentiality, integrity, and denial‐of‐service. In this paper, we present a hybrid intrusion detection system that combines supervised and unsupervised learning models through an ensemble stacking model to increase the detection accuracy rates of attacks in networks while minimising false alarms. Three machine learning algorithms comprising a multilayer perceptron neural network, a modified self‐organizing map, and a decision tree were used for the detection framework. The intrusion detection system was trained and evaluated on benchmark datasets: NSL‐KDD and CIC‐DDoS2019. The intrusion detection system was implemented as a Java solution and the detection performance was evaluated. A 10‐fold cross‐validation performance was also performed to validate how well the detection system predicts unknown attacks for prevention. The results of the tests revealed a detection accuracy of 99.84% of the instances in the NLS‐KDD dataset with a true positive rate of 99.8% and a false positive rate of 0.10% while a detection accuracy of 99.90% was achieved with the CIC‐DDoS2019 dataset. Furthermore, the detection system was effective in distinguishing attack traffic from normal traffic in the NSL‐KDD dataset and was able to adequately detect DOS, Probe, and R2L attacks with F1 scores of 100%, 99.6%, and 95.1%, respectively, which are significantly impressive. However, the detection of less frequency attack types such as U2R attacks was quite low with an F1 score of 62.5%. The detection performance of the proposed hybrid intrusion detection system suggests that it can be deployed in network security applications to detect packets that exhibit suspicious behaviour or indicate potential threats and respond appropriately to attacks. Implementing the detection framework as a Java solution makes it possible to deploy it across various operating system platforms without any impact on the detection performance.

Credit Card Fraud Detection Model Based on Correlation Feature Selection

Credit card fraud is a widespread cybercrime that threatens financial security. Effective cybersecurity measures are essential to mitigate these risks. Machine learning has shown promising results in detecting credit card fraud by analyzing transaction data and identifying patterns of suspicious behavior. Feature selection is crucial in machine learning because it simplifies the model, improves its performance, and prevents overfitting. This research introduces a machine learning model designed for credit card fraud detection. The model makes use of three types of correlations. Pearson, Spearman, and Kendall, to identify features and enhance the fraud detection process. Testing on datasets yielded impressive results achieving category accuracies of 99.95% and 99.58% surpassing alternative approaches. Also, the results showed that Kendall correlation is the best among the three types of correlation in selecting attributes in all approved datasets.

OVERVIEW OF FRAUD DETECTION SYSTEMS AND PERFORMANCE KPI DEVELOPMENT

In this article overview was provided on several fraud detection systems, analysis result of common scheme and development of KPIs to detect performance degradation or improvement from business logic point of view. Four different systems were reviewed. Following FMS were developed by Gigamon and Argyle Data cooperation, AWS, Subex, Cvidya Amdocs. Solution developed by Gigamon and Argyle Data consists of Gigamon fabric for information collection/filtering/enrichment and Argyle Data Fraud detection system, which is based on Hadoop technology to store collected data and analysis results by application. AWS Fraud Detection collects NRTRDE flow and process it by using ML technics provided by AWS. Subex fraud management system provides flexible ETL for data collection from different sources with adjustable detection rules and ML for suspicious behavior learning. FraudView by Cvidya Amdocs collects information from varying points like OSS/BSS, CRM customer details, Prepaid platforms, HLR, Switch CDRs, Probe (SS7, VoIP, IP) and process it by different detection engines. Simplified processing FMS processing scheme and KPIs based on different timestamps were made. Following conclusions were made: In reviewed FMS was noticed that instead of using traditional NRTRDE and TAP3 file formats, data can be collected directly from network by using network tap or port mirroring with next data enrichment, cleaning, formatting for fraud detection system to consume. Following real time method can be realized by using probes to perform data preparation or some complex solution described by Gigamon; Detection is performed by rules, provided by vendor or by ML modules, which learns behavior of subscriber in order to create detection rules. Most of systems allow to modify threshold of following rules in order to meet system user demands to check data within specific time (for example fraudster night calls to subscriber) or detect specific number of suspicious sessions, etc; In order speedup fraud detection hotlists, whitelists can be used for enrichment to filter out fraudsters, emergency or business numbers. Geographical location can be used to identify fraudster’s location within network and make correlation with other possible fraud sessions; During analysis of each FMS architecture, 3 processing stages were highlighted, which allowed to create simple KPIs for business logic and data arrival check; Developed methodology allows to check data arrival and fraud recognition with used data type to define which information provides better detection or view on rules for detection in order to show, which of them should be adjusted.