Suspicious Behavior Research Articles

Cyber threats in the age of artificial intelligence: Exploiting advanced technologies and strengthening cybersecurity

The rapid advancement in artificial intelligence (AI) technologies has led to the emergence of new and complex cyber threats. These threats rely on AI capabilities to target digital systems and infrastructures with greater precision and efficiency, making them more difficult to detect and counter. AI is being used to develop automated cyber-attacks, such as malware that can adapt and learn from its digital environment. Conversely, AI is also employed to enhance cyber security by enabling early threat detection, analyzing suspicious behaviors, and providing rapid responses to hacking incidents. Based on the above, the research idea we intend to present at this article on how to achieve a balance between leveraging AI technologies to protect digital systems and the growing risks associated with using these technologies to develop more sophisticated cyber-attacks. Research Methodologies: The study requires employing both the descriptive-analytical and inferential methodologies.

Natural Language Processing for Sentiment Analysis in Social Media Posts to Identify Suspicious Behaviour

This article reports a study on the use of the natural language processing approach in the sentiment analysis of social media posts for detecting suspicious activities involving deep learning methods. The primary objective is to design an automated system capable of identifying unfavourable sentiments and patterns that imply suspicious activity such as cyberbullying, hate speech and disinformation. The methodology involves collecting a diverse data set of 2,000 posts from platforms such as X and Facebook, pre-processed by tokenisation and normalisation. The sentiment analysis model uses a long short-term memory (LSTM) network because it finds that the LSTM network is good for recognising long-term dependencies between texts. Other heuristics are used to give prominence to posts with malicious behavioural profiles to flag them accordingly. The model is designed and tuned to pre-labelled data sets. This work shows considerable improvement over previous techniques at categorising the posts as suspicious by showing higher accuracy, precision, recall and F1 score ratios. In doing so, this research positively impacts the enhancement of paranoid security or the totality of social media safety by offering a strong early detection mechanism for reducing risky behaviours on social networks.

8255 Munchausen Syndrome After Partial Pancreatectomy for Insulinoma: A Case Report

Abstract Disclosure: C.R. de Macedo: None. T.D. Bertolacini da Silva: None. I.P. de Magalhães: None. M.A. Pereira: None. I.C. Marques: None. J.L. Pinheiro Filho: None. G.F. Darce: None. D.M. Lourenço Junior: None. Background: Insulinoma is a rare, insulin-secreting pancreatic neuroendocrine tumor, most commonly benign, solitary, sporadic, well differentiated and with a slight female predominance. The investigation of hypoglycemia requires a fasting test to exclude, among other causes, factitious hypoglycemia. Clinical Case: A 27-year-old woman was admitted to the emergency room after seizure by hypoglycemic crisis (capillary blood glucose < 55 mg/dL). She had recurrent episodes of adrenergic/neuroglycopenic symptoms since the age of 14, occurring during fasting, mainly in the morning, with improvement after eating. After confirming Whipple's triad, a fasting test showed hyperinsulinemic hypoglycemia (glycemia 53 mg/dL, normal value >55 mg/dL; insulin 11.5 mcU/mL, nv <3mcU/mL) and increased C-peptide (3.83 ng/mL, nv <0,6 ng/mL). Abdominal magnetic resonance showed a 1.4 cm nodule in the pancreas tail. During hospitalization, she experienced Raynaud's phenomenon, leading to the diagnosis of Raynaud's disease (RD). Distal pancreatectomy was performed, confirming the diagnosis of insulinoma by immunohistochemistry. There was brief remission of the hypoglycemic episodes once they recurred two weeks later. As imaging tests did not show new pancreatic lesions, somatostatin analogue was begun, resulting in improvement for a few more weeks. Due to recurrent hypoglycemia and new seizures, she required continuous intravenous infusion of glucose during hospitalization. A new fasting test revealed hyperinsulinemic hypoglycemia (glycemia 42 mg/dL, insulin 43.3 mcU/mL), low C-peptide (0.5 ng/mL) and normal proinsulin (2.39 pmol/L, nv <5 pmol/L) values. Due RD, auto-immune hypoglycemia was excluded by negative anti-insulin antibodies. Insulinomatosis was also considered, as potential etiology, but review of surgical specimen excluded this hypothesis. After 4 months of hospitalization, due suspicious behavior, a review of her personal belongings revealed insulin vials and needles for injection. The patient confessed to applying insulin to simulate the symptoms, also revealing a history of childhood trauma and previous suicide attempts. After psychiatric evaluation, borderline personality disorder was identified, and Munchausen syndrome was considered. Since then, the patient did not require intravenous glucose with no recurrence of hypoglycemia. Conclusion: The investigation of hypoglycemia can be challenging, requiring appropriate diagnostic protocols. Even in patients who appear to have a well-defined diagnosis, as in our case with anatomopathological confirmation of insulinoma, it is always important to rule out factitious hypoglycemia. The incorrect diagnosis can lead to inappropriate invasive and surgical procedures with definitive consequences, such as exogenous pancreatic insufficiency and insulin-dependent diabetes mellitus. Presentation: 6/1/2024

Detecting criminal intent in social interactions: The influence of autism and theory of mind.

Defense attorneys sometimes suggest that social-cognitive difficulties render autistic individuals vulnerable to involvement in crime, often arguing that theory of mind (ToM) difficulties that undermine inferences about others' intentions underpin this vulnerability. We examined autistic adults' ability to respond adaptively to criminal intent during interactions and whether difficulties were associated with poor ToM. Compared with nonautistic adults, autistic adults were expected to be less likely to respond adaptively to another's criminal intent and less likely to do so early in interactions, with poorer performance associated with ToM difficulties. We developed the Suspicious Activity Paradigm, in which autistic (n = 102) and nonautistic (n = 95) adults listened (as if participating in an interaction) to audio scenarios in which cues suggesting their impending involvement in a crime gradually emerged. At periodic intervals, they were required to indicate how they would react toward the other person's behavior, with response coding reflecting detection of, and adaptive responding to, suspicious activity. We observed similar patterns of suspicion and adaptive responses in autistic and nonautistic adults as the scenarios progressed. Regardless of diagnostic status, pronounced ToM difficulties and low verbal ability were independently associated with a lower likelihood of reporting suspicion and responding adaptively. Our results do not support the perspective that autistic adults are uniquely vulnerable to crime involvement due to an inability to recognize and respond adaptively to suspicious behavior. The potential for heightened criminal vulnerability was associated with significant ToM difficulties (and verbal ability) regardless of autism diagnostic status, although such difficulties were more prevalent in the autistic sample. The finding that pronounced ToM difficulties may heighten criminal vulnerability for both autistic and nonautistic individuals challenges the validity of a generalized "autism" legal defense based on assumed rather than measured ToM difficulties. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Using Hardware Events to Detect Side-Channel Attacks

The article discusses the monitoring system being developed by authors for the detection of Side-Channel Attacks, SCA. The theoretical justification of the chosen method of detecting SCA attacks is given: analysis of hardware events of the target system. The architecture of the monitoring system is described, including low-level data collection processes and an expert system that analyzes the collected data. The results of the proof-of-concept on various classes of SCA-attacks are presented. The paper considers a class of attacks on side-channels that use the processor cache to obtain secret information. A method of countering attacks of this class based on the use of a hardware event mechanism is proposed. As a result of the analysis of existing hardware events, predicates have been built for the Intel architecture, allowing one to identify the suspicious behavior of programs in the Linux OS environment. To test the proposed method, a prototype monitoring system was implemented that successfully coped with the detection of simulated SCA-attacks. The advantages and disadvantages of this method are considered, and the direction of further research is indicated.

Dentists' knowledge about domestic violence against women: A questionnaire-based study.

Dentists are the first healthcare professionals to identify cases of domestic violence and abuse (DVA) with head and neck injuries. The aim of this study was to assess dentists' knowledge regarding behavioral and physical findings in female victims of DVA. The study included 558 volunteer dentists who completed a two-part questionnaire designed to assess their knowledge and awareness of DVA against women. The first part of the questionnaire inquired about the participants' demographic data, including age, sex, specialty, and the duration of professional practice. The second part assessed 15 statements on a 5-point Likert scale, concerning both behavioral (5/15) and clinical aspects (10/15) of DVA. For each statement, if a minimum of 70% of respondents selected the same option, it was interpreted as being agreed upon by the participants. The level of statistical significance was set at p < 0.05. Most of the participants were between the ages of 31 and 40 (29.1%). The study sample was predominantly female (70.4%), with 57.2% of the participants being married. The statement "Abused women tend to avoid eye contact" had the highest agreement rate (70.6%) for the behavioral assessments. However, the participants were mostly "undecided" on the remaining 4 statements in this section. In contrast, the agreement rate for 5 statements related to the clinical assessment of head, neck and intraoral injuries exceeded 70%. One-fifth of the behavioral assessment statements and half of the clinical assessment statements were negatively correlated with the age of the participants (p < 0.05). The findings of this study indicate that dentists can recognize and diagnose DVA symptoms to a certain extent. However, they may encounter difficulties in identifying the suspicious behavior that is indicative of DVA.

Machine Learning Approaches for Network Intrusion Detection: An Evaluation of their Efficacy in Bolstering Security

Abstract: In today's digital environment, securing networked systems is critical due to the increasing sophistication and frequency of cyberattacks. Network Intrusion Detection Systems (NIDS) are essential for detecting and addressing unauthorized and harmful activities within a network. NIDS function by continuously monitoring network traffic and analyzing data packets for signs of suspicious behavior or known attack patterns. This paper provides an in-depth examination of NIDS, focusing on the application of three machine learning algorithms—K Neighbors Classifier, Logistic Regression, and Random Forest Classifier—to create a robust model for network intrusion detection using the NSL-KDD dataset. Our findings highlight the superior performance of the Random Forest Classifier, which achieved an accuracy of 99.31%, proving its effectiveness in differentiating between normal and malicious traffic. The analysis of ICMP, TCP, and UDP protocols reveals unique attack patterns, underscoring the need for protocol-specific security measures. Additionally, the study emphasizes the importance of integrating NIDS with other security systems for a multi-layered defense strategy and the crucial role of skilled personnel in managing and interpreting NIDS alerts. The results advocate for ongoing innovation and adaptation in NIDS technologies and strategies to effectively counter evolving cyber threats

ML-DRIVEN ONLINE SAFETY: AN AUTOMATED DETECTION OF CHILD PREDATORS AND CYBER HARASSERS

Social media has become a vital component of our existence, facilitating global connections between individuals. While these platforms provide numerous advantages, they have also exposed vulnerable individuals, particularly children, to online risks. Individuals who prey on children and engage in online harassment utilize the anonymity and wide reach of social media platforms to inflict harm upon others. Previously, these dangers were addressed through the use of manual reporting and human moderators. Users reported suspicious behavior, prompting human moderators to review the content for compliance with platform requirements. Nevertheless, this responsive approach frequently resulted in a delay in taking action, so enabling the spread of harmful content. Researchers have employed machine learning, a type of artificial intelligence that enables computers to learn from data and make predictions, to develop more proactive and effective solutions.

Advanced deep learning for masked individual surveillance

During Covid-19 pandemic, face masks have become a ubiquitous protective measure. This poses new challenges for surveillance systems that heavily rely on facial recognition. To address this critical issue, we present a novel enhanced surveillance system that leverages deep learning techniques to tackle two crucial tasks simultaneously: anomaly detection of masked individuals' activities and masked face completion for accurate recognition. For anomaly detection, we employ a custom-designed deep neural network capable of processing real-time video streams. Finding a dataset of anomaly events of masked individuals is a big challenge for us. We handle this challenge using efficient techniques such as Dlib library and other image processing techniques. The network is trained on a diverse dataset encompassing normal and abnormal activities of masked individuals, enabling it to identify suspicious behaviors effectively. The surveillance cameras will exchange information, using a suitable network protocol, about detected anomalies and share relevant image data to aid in decision-making and choose the best images for further processing. In the context of masked face completion, we present a novel architecture called CCGAN network that is a combination of convolutional neural network (CNN) and conditioned generative adversarial network (CGAN) to generate the hidden parts of the face in a form that is accurate and close to the original face shape, as shown in our results. We conduct extensive experiments on publicly available datasets, demonstrating superior performance in both anomaly detection and masked face completion tasks. We have achieved 90% accuracy for anomaly detection of masked people.

Enhancing Monitoring Performance: A Microservices Approach to Monitoring with Spyware Techniques and Prediction Models.

In today's digital landscape, organizations face significant challenges, including sensitive data leaks and the proliferation of hate speech, both of which can lead to severe consequences such as financial losses, reputational damage, and psychological impacts on employees. This work considers a comprehensive solution using a microservices architecture to monitor computer usage within organizations effectively. The approach incorporates spyware techniques to capture data from employee computers and a web application for alert management. The system detects data leaks, suspicious behaviors, and hate speech through efficient data capture and predictive modeling. Therefore, this paper presents a comparative performance analysis between Spring Boot and Quarkus, focusing on objective metrics and quantitative statistics. By utilizing recognized tools and benchmarks in the computer science community, the study provides an in-depth understanding of the performance differences between these two platforms. The implementation of Quarkus over Spring Boot demonstrated substantial improvements: memory usage was reduced by up to 80% and CPU usage by 95%, and system uptime decreased by 119%. This solution offers a robust framework for enhancing organizational security and mitigating potential threats through proactive monitoring and predictive analysis while also guiding developers and software architects in making informed technological choices.

MCAD: A Machine Learning Based Cyber Attack Detector using SDN for Healthcare Systems

The healthcare industry, increasingly reliant on digital technology, has become a prime target for cyberattacks. While traditional security measures address external threats, they often fail to effectively counter a particularly dangerous foe: insider threats. This project proposes MCAD (Machine Learning-based CyberAttack Detector), a groundbreaking approach that leverages the power of machine learning within Software-Defined Networks (SDNs) to bolster healthcare network security. Healthcare networks are inherently complex ecosystems, housing a diverse range of medical devices alongside traditional IT infrastructure. This intricate web creates a larger attack surface for malicious insiders with access to critical systems. These insiders can be disgruntled employees or attackers exploiting vulnerabilities in poorly designed systems. The COVID-19 pandemic further exacerbated this vulnerability as the surge in telehealth services and remote access points opened new avenues for exploitation. The statistics are alarming, with a staggering 92% of healthcare organizations reporting insider-caused security breaches. These breaches not only compromise sensitive patient data but can also disrupt critical healthcare services, potentially jeopardizing patient safety. MCAD, a Machine Learning-based Cyber Attack Detector, tackles the growing threat of insider attacks in healthcare networks. It employs a multi-pronged approach: collecting both normal and abnormal network traffic to train a real-time machine learning model. This model continuously analyzes network activity, identifying suspicious behavior indicative of insider threats. MCAD seamlessly integrates with SDN controllers for efficient deployment within existing infrastructure, and undergoes rigorous testing with various machine learning algorithms and simulated attacks to ensure optimal protection against evolving cyber threats.

FLOLSTM: Fuzzy logic‐driven optimized LSTM for improved malicious traffic detection in hypervisor environments

SummaryIn the ever‐evolving realm of cloud computing, the challenge of intrusion detection has grown increasingly intricate and vital. With the proliferation of cyber‐attacks and the widespread use of virtualized environments, there is a pressing need for network security solutions that are not only innovative and robust but also easily comprehensible. These solutions must possess the ability to effectively detect malicious activities, provide visibility into network operations, adapt to changing requirements, and promptly alert stakeholders to any suspicious behavior. In this study, we introduce a groundbreaking approach known as fuzzy logic‐driven optimized long short‐term memory (LSTM)—FLOLSTM, specifically designed for hypervisor‐based environments. By integrating fuzzy logic with an optimized LSTM neural network, FLOLSTM aims to significantly enhance the detection of attack traffic within hypervisor networks by combining the interpretability and uncertainty management capabilities of fuzzy logic with the temporal pattern recognition prowess of LSTM. The research methodology involves meticulous data collection from hypervisor monitors, followed by rigorous cleaning and preprocessing to enhance data reliability. Subsequently, the preprocessed data is input into the FLOLSTM classifier to identify malicious activities. Furthermore, the performance of the LSTM is fine‐tuned using the waterwheel plant optimization (WPO) algorithm. Experimental evaluations compare the efficiency of the proposed FLOLSTM with existing techniques across various metrics including accuracy, recall, precision, F‐measure, specificity, false‐positive rate (FPR), and false‐negative rate (FNR). Overall, the FLOLSTM model represents a significant advancement in intrusion detection for cloud environments, offering a potent blend of interpretability, accuracy, and efficiency. Its superior performance underscores its capacity to enhance network security and effectively mitigate cyber threats in dynamic and virtualized settings, thereby making a substantial impact on the field of cyber security.

Suspicious behaviour detection in multilayer social networks using PF-KMA and SS-GAE techniques

Multilayer Social networks are an important part of human life to interact on different networks at the same time. Due to the openness of such networks, they become a platform for spammers to spread malicious behaviors. Hence, there is an urgent need for effective detection of malicious behaviors; thereby, enabling the networks to take mitigation actions to decrease the possibility to reward such activities. Detection of suspicious behaviors in previous works is challenging due to the problems of community detection, the large amount of feature corruption, and memory requirements. Thus, to deal with such problems, in this paper, an efficient clustering-based detection of malicious users in multilayer social networks is proposed. Initially, the input dataset is pre-processed and used for Exponential Distribution based Erdős–Rényi based graph construction. From the graph structure, two types of data, such as user representations and graph features are extracted for graph encoding using the Soft Sign activated Graph Auto Encoder model. Then, the decoding is done to predict the information diffusion level, thereby, ranking the users using the Laplace Regularization technique. Then, the ranked users are clustered into different groups using the Pareto Front based K-Means Clustering Algorithm technique. Finally, the experimental results were analyzed to demonstrate the efficacy of the proposed model to detect malicious users in multilayer social networks.

Evaluating the Problem of Fraudulent Participants in Health Care Research: Multimethod Pilot Study.

The shift toward online recruitment methods, accelerated by the COVID-19 pandemic, has brought to the forefront the growing concern of encountering fraudulent participants in health care research. The increasing prevalence of this issue poses a serious threat to the reliability and integrity of research data and subsequent findings. This study aims to explore the experiences of health care researchers (HCRs) who have encountered fraudulent participants while using online recruitment methods and platforms. The primary objective was to gain insights into how researchers detect and mitigate fraudulent behavior in their work and provide prevention recommendations. A multimethod sequential design was used for this pilot study, comprising a quantitative arm involving a web-based survey followed by a qualitative arm featuring semistructured interviews. The qualitative description approach framed the qualitative arm of the study. Sample sizes for the quantitative and qualitative arms were based on pragmatic considerations that in part stemmed from encountering fraudulent participants in a concurrent study. Content analysis was used to analyze open-ended survey questions and interview data. A total of 37 HCRs participated, with 35% (13/37) of them engaging in qualitative interviews. Online platforms such as Facebook, email, Twitter (subsequently rebranded X), and newsletters were the most used methods for recruitment. A total of 84% (31/37) of participants indicated that fraudulent participation occurred in studies that mentioned incentives in their recruitment communications, with 71% (26/37) of HCRs offering physical or electronic gift cards as incentives. Researchers identified several indicators of suspicious behavior, including email surges, discrepancies in contact or personal information, geographical inconsistencies, and suspicious responses to survey questions. HCRs emphasized the need for a comprehensive screening protocol that extends beyond eligibility checks and is seamlessly integrated into the study protocol, grant applications, and research ethics board submissions. This study sheds light on the intricate and pervasive problem of fraudulent participation in health care research using online recruitment methods. The findings underscore the importance of vigilance and proactivity among HCRs in identifying, preventing, and addressing fraudulent behavior. To effectively tackle this challenge, researchers are encouraged to develop a comprehensive prevention strategy and establish a community of practice, facilitating real-time access to solutions and support and the promotion of ethical research practices. This collaborative approach will enable researchers to effectively address the issue of fraudulent participation, ensuring the conduct of high-quality and ethically sound research in the digital age.

Human Action Recognition System Using LRCN

Abstract: The importance of human action recognition is significant across various domains, driving advancements in safety, healthcare, sports analytics, and interactive technologies. Leveraging machine learning techniques like Long Recurrent Convolutional Neural Networks (LRCN) trained on datasets such as UCF50, our project focuses on action prediction from YouTube videos. This technology plays a pivotal role in enhancingsafety and security by enabling the detection of anomalies and suspicious behaviours in surveillance systems. In healthcare, it supports remote patient monitoring, rehabilitation assessment, and personalized care plans based on observed actions. Addi- tionally, in sports analytics and entertainment, human action recognition informs performance evaluation, content creation, and immersive experiences. Our integration of a pre-trained LRCN model into a Flask web application signifies the tangible impact of machine learning on human action recognition, with ongoing efforts to optimize functionality through input validation, error handling, user feedback, security measures, and performance enhancements, illustrating the practical application and societal benefits of this innovative technology

Leveraging Artificial Intelligence for Crime Detection and Prevention

This paper sheds light on how Artificial Intelligence (AI) is reshaping the landscape of crime detection and prevention, bringing about a significant change in traditional law enforcement methods. With society increasingly embracing AI-driven solutions, it's crucial to understand how they contribute to public safety. We'll dive into various aspects of AI, such as predictive policing, video surveillance analysis, forensic science, and criminal profiling, to see how they enhance law enforcement capabilities and streamline crime resolution processes.Starting with predictive policing, AI algorithms sift through massive datasets to identify patterns and predict future criminal activity, empowering law enforcement to take proactive measures. Similarly, AI-powered video surveillance systems enhance real-time monitoring and anomaly detection, allowing for quick identification of suspicious behavior in urban environments.Moving on, we'll explore AI's impact on forensic analysis, where it expedites processes like DNA sequencing, fingerprint recognition, and ballistics analysis, speeding up the resolution of cold cases. Additionally, we'll delve into AI's role in criminal profiling, highlighting its ability to decipher behavioral patterns and predict criminal motivations, though ethical considerations remain paramount.We'll also address ethical concerns surrounding AI deployment in law enforcement, including privacy issues and algorithmic biases. Emphasizing the need for transparency and accountability, we advocate for responsible use of AI-powered crime detection systems to uphold ethical standards and protect individual rights.

CRIME PREVENTION AND WORK MONITORING IN CROWD USING CCTV NETWORK

In an era where surveillance technology is becoming increasingly sophisticated, leveraging Closed-Circuit Television (CCTV) networks for crime prevention and crowd monitoring has gained significant attention. This research aims to explore the potential of CCTV systems in deterring criminal activities and ensuring public safety in crowded environments. By integrating advanced video analytics and machine learning algorithms, the CCTV network can detect suspicious behavior patterns, identify potential threats, and alert law enforcement in real-time. Moreover, by employing work monitoring features, such as crowd density analysis and traffic flow optimization, the system can enhance operational efficiency and mitigate safety risks in various public spaces, including transportation hubs, shopping malls, and urban centers. Ultimately, this study seeks to contribute to the development of proactive security measures and intelligent surveillance solutions for safer communities. Keywords— Artificial Intelligence, CCTV Network, Crime Prevention, Crowd Monitoring, Surveillance.

Active Chat Monitoring and Suspicious Chat Detection Over Internet

In the digital age, the proliferation of online communication platforms has facilitated unprecedented levels of connectivity and interaction. However, this interconnectedness has also introduced new challenges, particularly in ensuring the safety and security of users in online environments. One critical area of concern is the monitoring and detection of suspicious activities within chat platforms, where malicious actors may engage in harmful behaviors such as cyberbullying, harassment, or illicit activities. This research paper focuses on the development and implementation of active chat monitoring techniques for the detection of suspicious behavior over the internet. By leveraging advancements in natural language processing (NLP), machine learning, and data analytics, this study aims to explore effective methodologies for real-time monitoring and analysis of chat conversations to identify potentially harmful content. The paper will delve into various approaches for detecting suspicious patterns, including keyword analysis, sentiment analysis, and anomaly detection algorithms. Additionally, considerations for privacy, ethics, and legal implications surrounding chat monitoring will be discussed. Ultimately, this research endeavors to contribute to the enhancement of online safety measures by providing insights into the design and implementation of proactive monitoring systems capable of identifying and mitigating risks in virtual communication spaces. Keywords— Active Chat Monitoring, cybersecurity, user experience, digital era, security, user-controlled, effectiveness, evolving threat, Chat content analysis, Threat intelligence.

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.