Genuine User Research Articles

Enhancing Biometric Security with Bimodal Deep Learning and Feature-level Fusion of Facial and Voice Data

Recent research in biometric technologies underscores the benefits of multimodal systems that use multiple traits to enhance security by complicating the replication of samples from genuine users. To address this, we present a bimodal deep learning network (BDLN or BNet) that integrates facial and voice modalities. Voice features are extracted using the SincNet architecture, and facial image features are obtained from convolutional layers. Proposed network fuses these feature vectors using either averaging or concatenation methods. A dense connected layer then processes the combined vector to produce a dual-modal vector that encapsulates distinctive user features. This dual-modal vector, processed through a softmax activation function and another dense connected layer, is used for identification. The presented system achieved an identification accuracy of 99% and a low equal error rate (EER) of 0.13% for verification. These results, derived from the VidTimit and BIOMEX-DB datasets, highlight the effectiveness of the proposed bimodal approach in improving biometric security.

Leveraging Artificial Intelligence and Machine Learning for Detecting Fake Accounts on Social Media

Detecting fake accounts is a significant challenge for social media platforms, e-commerce sites, and online services. Malicious users create fake profiles to spread misinformation, commit fraud, or manipulate system behavior. Traditional rule based methods often fail to catch sophisticated fake accounts that resemble genuine user activities. This paper examines the application of Artificial Intelligence (AI) and Machine Learning (ML) techniques to improve the detection of these deceptive accounts. Using KNN model and Gradient Boosting algorithm offer a more robust solution to identifying and combating fake accounts in Social Media Platform.

Enhancing security through continuous biometric authentication using wearable sensors

The paper presents a novel approach for biometric continuous driver authentication (CDA) for secure and safe transportation using wearable photoplethysmography (PPG) sensors and deep learning. Conventional one-time authentication (OTA) methods, while effective for initial identity verification, fail to continuously verify the driver’s identity during vehicle operation, potentially leading to safety, security, and accountability issues. To address this, we propose a system that employs Long Short-Term Memory (LSTM) models to predict subsequent PPG values from wrist-worn devices and continuously compare them with real-time sensor data for authentication. Our system calculates a confidence level representing the probability that the current user is the authorized driver, ensuring robust availability to genuine users while detecting impersonation attacks. The raw PPG data is directly fed into the LSTM model without pre-processing, ensuring lightweight processing. We validated our system with PPG data from 15 volunteers driving for 15 min in varied conditions. The system achieves an Equal Error Rate (EER) of 4.8%. Our results demonstrate that the system is a viable solution for CDA in dynamic environments, ensuring transparency, efficiency, accuracy, robust availability, and lightweight processing. Thus, our approach addresses the main challenges of classical driver authentication systems and effectively safeguards passengers and goods with robust driver authentication.

Participatory design in telehealth research: Practical case examples.

Participatory design (PD) is a methodology that emphasizes user participation in the design of new technologies to leverage change within organizations and services. PD originated in the computer science field in the 1970s and 1980s when new programs and technologies were developed to empower workers, by involving them in decisions that affected them. PD in health research has been proven to change clinical practice. Genuine user involvement that includes all stakeholders, and robust collaborations across sciences, sectors, and disciplines are basic elements of successful research to change clinical practice and to implement novel technical and organizational approaches. This paper summarizes seven case studies involving the use of PD in telehealth research. All cases presented promoted organizational changes supported by health information and communications technology, and have been implemented at either international, national, regional, or local levels. We describe how PD can be applied in health sciences and used to facilitate organizational changes, new perspectives, and new communications methods. The relevance and suitability of PD as a research design in health science is explained, and recommendations for conducting PD studies in telehealth research are presented. In PD, mutual learning and co-creation is facilitated. Consequently, learning from users, rather than studying them, corroborates our understanding and the emergence of new knowledge.

Robust Estimation Method against Poisoning Attacks for Key-Value Data with Local Differential Privacy

Local differential privacy (LDP) protects user information from potential threats by randomizing data on individual devices before transmission to untrusted collectors. This method enables collectors to derive user statistics by analyzing randomized data, thereby presenting a promising avenue for privacy-preserving data collection. In the context of key–value data, in which discrete and continuous values coexist, PrivKV has been introduced as an LDP protocol to ensure secure collection. However, this framework is susceptible to poisoning attacks. To address this vulnerability, we propose an expectation maximization (EM)-based algorithm combined with a cryptographic protocol to facilitate secure random sampling. Our LDP protocol, known as emPrivKV, exhibits two key advantages: it improves the accuracy of statistical information estimation from randomized data, and enhances resilience against the manipulation of statistics, that is, poisoning attacks. These attacks involve malicious users manipulating the analysis results without detection. This study presents the empirical results of applying the emPrivKV protocol to both synthetic and open datasets, highlighting a notable improvement in the precision of statistical value estimation and robustness against poisoning attacks. As a result, emPrivKV improved the frequency and the mean gains by 17.1% and 25.9%, respectively, compared to PrivKV, with the number of fake users being 0.1 of the genuine users. Our findings contribute to the ongoing discourse on refining LDP protocols for key–value data in scenarios involving privacy-sensitive information.

Enhancing authenticity and trust in social media: an automated approach for detecting fake profiles

<p class="Abstract">Fake profile detection on social media is a critical task intended for detecting and alleviating the existence of deceptive or fraudulent user profiles. These fake profiles, frequently generated with malicious intent, could engage in different forms of spreading disinformation, online fraud, or spamming. A range of techniques is employed to solve these problems such as natural language processing (NLP), machine learning (ML), and behavioural analysis, to examine engagement patterns, user-generated content, and profile characteristics. This paper proposes an automated fake profile detection using the coyote optimization algorithm with deep learning (FPD-COADL) method on social media. This multifaceted approach scrutinizes user-generated content, engagement patterns, and profile attributes to differentiate genuine user accounts from deceptive ones, ultimately reinforcing the authenticity and trustworthiness of social networking platforms. The presented FPD-COADL method uses robust data pre-processing methods to enhance the uniformness and quality of data. Besides, the FPD-COADL method applies deep belief network (DBN) for the recognition and classification of fake accounts. Extensive experiments and evaluations on own collected social media datasets underscore the effectiveness of the approach, showcasing its potential to identify fake profiles with high scalability and precision.</p>

A hybrid NLP and domain validation technique for disposable email detection

Disposable email address services have gained popularity recently, offering users a way to register for online services without revealing their primary email. However, this poses challenges for organizations aiming to engage with genuine users. This paper introduces a novel approach combining natural language processing (NLP) and domain validation to identify disposable email addresses. The technique employs various machine learning methods, including Support Vector Classifier (SVC), Multinomial Naive Bayes (MNB), Gaussian Naive Bayes (GNB), Logistic Regression, XGBoost Classifier, K Neighbors Classifier, Random Forest Classifier, Linear, and Discriminant Analysis. Unlike traditional methods relying on blacklists, this technique achieves a 97% accuracy rate by effectively detecting and classifying new disposable emails.

Designing Empathetic Service Experiences Using Storytelling Approach: A Case Study

Storytelling is a well-known technique to emphasize emotions. This paper explores the integration of storytelling techniques into service design to cultivate empathetic service experiences. Emphasizing the significance of emotional connection in social service contexts, we designed services for Adolescents Living with HIV (ALHIV) in India. Due to a lack of awareness and social stigma, ALHIV are unable to make timely and responsible life decisions around intimacy and HIV status disclosure. We designed services that disseminate essential information and enable ALHIV to make informed life decisions. The solution is an interactive story media where users make decisions around in-design encounters that parallel with possible real-life encounters. As a result, they get to explore multiple possibilities in a safe space, leading to improved real-life decision-making. This paper demonstrates the replicable framework of orchestration of human and technology touchpoints through multiple dynamic and choice-based scenarios. The study examines the synergy between service design and narrative frameworks. Through a detailed case study, the research underscores the pivotal role of storytelling in fostering genuine user engagement by infusing empathy-driven narratives into service encounters. Overall, this research contributes a novel perspective on designing empathetic services, demonstrating the role of storytelling in crafting meaningful service experiences.

Game Theoretic Approaches to Modelling Cooperative and Competitive Dynamics for Enhanced Routing and Resource Allocation

The goal of the field of game theory is to simulate scenarios in which decision-makers must choose between courses of action that may or may not conflict and in which the choices made by one element may affect the choices made by other elements. Techniques from game theory have been widely used to solve a variety of engineering design issues. Game theory may be applied to wireless networks to create cooperative strategies between nodes, terminals and network providers, among other organizations. The majority of the time, game theory has been used in networking to address routing and resource allocation issues in competitive settings. Because radio communication channels are typically shared in wireless networks, the actions of one wireless device may have an impact on the communication capabilities of a neighboring device. Game theory may be used to simulate situations such as these. A multi-layered viewpoint is presented in Applications of Game Theory in Wireless Networking, highlighting the domains in which game theory might find practical application. This explains how different wireless network interactions may be represented as games and how game-theoretical methods can accurately mimic or forecast genuine user behavior in cooperative or competitive settings. Because of these similarities, it is possible to grasp the intricate relationships between nodes in this highly dynamic and dispersed environment by using a strong mapping between classical game theory components and network parts

Machine Learning Based Spam Comments Detection on Youtube: A Comparison of Regression Models

The rise of spam comments on platforms like YouTube has become a significant concern, as they not only hinder genuine user engagement but also pose serious risks to users' safety and privacy. Machine Learning (ML) offers a powerful solution to combat spam comments by automating the process of detecting and preventing them. With the ability to analyze vast amounts of data and patterns, ML algorithms can effectively distinguish between legitimate comments and those that are spam. One of the commonly employed approaches in ML for spam comment detection is the Naive Bayes classification algorithm. Naive Bayes is a probabilistic algorithm that calculates the likelihood of a comment being spam based on its characteristics and the occurrence of specific keywords or phrases that are typical of spam content. By training the algorithm on a labeled dataset of spam and non-spam comments, it can learn to recognize patterns and generalize its understanding to new, unseen comments. Achieving a detection accuracy of 92.78% is indeed promising, but researchers and developers continue to explore other ML techniques and combinations to further improve the accuracy and robustness of spam comment detection systems. Ensemble methods, deep learning, and Natural Language Processing (NLP) techniques are among the advanced approaches gaining attention in this domain. One crucial aspect of an effective spam detection system is its adaptability and responsiveness to emerging spam tactics.

MotionID: Towards practical behavioral biometrics-based implicit user authentication on smartphones

Traditional one-time authentication mechanisms cannot authenticate smartphone users’ identities throughout the session — the concept of using behavioral-based biometrics captured by the built-in motion sensors and touch data is a candidate to solve this issue. Many studies proposed solutions for behavioral-based continuous authentication; however, they are still far from practicality and generality for real-world usage. To date, no commercially deployed implicit user authentication scheme exists because most of those solutions were designed to improve detection accuracy without addressing real-world deployment requirements. To bridge this gap, we tackle the limitations of existing schemes and reach towards developing a more practical implicit authentication scheme, dubbed MotionID, based on a one-class detector using behavioral data from motion sensors when users touch their smartphones. Compared with previous studies, our work addresses the following challenges: ① Global mobile average to dynamically adjust the sampling rate for sensors on any device and mitigate the impact of using sensors’ fixed sampling rate; ② Over-all-apps to authenticate a user across all the mobile applications, not only on-specific application; ③ Single-device-evaluation to measure the performance with multiple users’ (i.e., genuine users and imposters) data collected from the same device; ④ Rapid authentication to quickly identify users’ identities using a few samples collected within short durations of touching (1–5 s) the device; ⑤ Unconditional settings to collect sensor data from real-world smartphone usage rather than a laboratory study. To show the feasibility of MotionID for those challenges, we evaluated the performance of MotionID with ten users’ motion sensor data on five different smartphones under various settings. Our results show the impracticality of using a fixed sampling rate across devices that most previous studies have adopted. MotionID is able to authenticate users with an F1-score up to 98.5% for some devices under practical requirements and an F1-score up to roughly 90% when considering the drift concept and rapid authentication settings. Finally, we investigate time efficiency, power consumption, and memory usage considerations to examine the practicality of MotionID.

NOTRY: Deniable messaging with retroactive avowal

Modern secure messaging protocols typically aim to provide deniability. Achieving this requires that convincing cryptographic transcripts can be forged without the involvement of genuine users. In this work, we observe that parties may wish to revoke deniability and avow a conversation after it has taken place. We propose a new protocol called Not-on-the-Record-Yet (NOTRY) which enables users to prove a prior conversation transcript is genuine. As a key building block we propose avowable designated verifier proofs which may be of independent interest. Our implementation in- curs roughly 8× communication and computation overhead over the standard Signal protocol during regular operation. We find it is nonetheless deployable in a realistic setting as key exchanges (the source of the overhead) still complete in just over 1ms on a modern computer. The avowal protocol induces only constant computation and communication performance for the communicating parties and scales linearly in the number of messages avowed for the verifier—in the tens of milliseconds per avowal.

An integrated SDN framework for early detection of DDoS attacks in cloud computing

Cloud computing is a rapidly advancing technology with numerous benefits, such as increased availability, scalability, and flexibility. Relocating computing infrastructure to a network simplifies hardware and software resource monitoring in the cloud. Software-Defined Networking (SDN)-based cloud networking improves cloud infrastructure efficiency by dynamically allocating and utilizing network resources. While SDN cloud networks offer numerous advantages, they are vulnerable to Distributed Denial-of-Service (DDoS) attacks. DDoS attacks try to stop genuine users from using services and drain network resources to reduce performance or shut down services. However, early-stage detection of DDoS attack patterns in cloud environments remains challenging. Current methods detect DDoS at the SDN controller level, which is often time-consuming. We recommend focusing on SDN switches for early detection. Due to the large volume of data from diverse sources, we recommend traffic clustering and traffic anomalies prediction which is of DDoS attacks at each switch. Furthermore, to consolidate the data from multiple clusters, event correlation is performed to understand network behavior and detect coordinated attack activities. Many existing techniques stay behind for early detection and integration of multiple techniques to detect DDoS attack patterns. In this paper, we introduce a more efficient and effectively integrated SDN framework that addresses a gap in previous DDoS solutions. Our framework enables early and accurate detection of DDoS traffic patterns within SDN-based cloud environments. In this framework, we use Recursive Feature Elimination (RFE), Density Based Spatial Clustering (DBSCAN), time series techniques like Auto Regressive Integrated Moving Average (ARIMA), Lyapunov exponent, exponential smoothing filter, dynamic threshold, and lastly, Rule-based classifier. We have evaluated the proposed RDAER model on the CICDDoS 2019 dataset, that achieved an accuracy level of 99.92% and a fast detection time of 20 s, outperforming existing methods.

BotGSL: Twitter Bot Detection with Graph Structure Learning

Abstract Twitter bot detection is an important and meaningful task. Existing methods can be bypassed by the latest bots that disguise themselves as genuine users and evade detection by mimicking them. These methods also fail to leverage the clustering tendencies of users, which is the most important feature for detecting bots at the community level. Moreover, they neglect the implicit relations between users that contain crucial clues for detection. Furthermore, the user relation graphs, which are essential for graph-based methods, may be unreliable due to noise and incompleteness in datasets. To address these issues, a bot detection framework with graph structure learning is proposed. The framework constructs a heterogeneous graph with users and their relations, extracts multiple features to characterise user intent and establishes a feature similarity graph using metric learning. Implicit relations are discovered to derive an implicit relation graph. Additionally, a semantic relation graph is generated by aggregating relation semantics among users. The graphs are then fused and embedded into a Graph Transformer for training with partially known user labels. The framework demonstrated a 91.92% average detection accuracy on three real-world benchmark, outperforming state-of-the-art methods, while also showcasing the effectiveness and necessity of each module.

A comprehensive analysis of gesture recognition systems: Advancements, challenges, and future direct

Gesture recognition emerges as a potent avenue for human-computer interaction, harnessing mathematical algorithms to interpret gestures. It promises to surpass text-based or graphical interfaces, enabling touchless device control through simple gestures. Our review of 7 papers encompassing various fields and methods underscores its diverse applications. Challenges persist, such as distinguishing genuine user intent from accidental actions amid environmental interference. Creating a universal EMG pattern recognition model demands intricate individual pre-training. Sensor-based gesture recognition grapples with real-world dynamics, necessitating adaptable models that discern user intent from non-intent actions. Addressing these gaps holds the key. Adaptable models and personalized approaches can enhance robustness and accuracy across applications, surmounting challenges in the gesture interaction technology realm.

An Anonymous and Supervisory Cross-chain Privacy Protection Protocol for Zero-trust IoT Application

Internet of things (IoT) development tends to reduce the reliance on centralized servers. The zero-trust distributed system combined with blockchain technology has become a hot topic in IoT research. However, distribution data storage services and different blockchain protocols make network interoperability and cross-platform more complex. Relay chain is a promising cross-chain technology that solves the complexity and compatibility issues associated with blockchain cross-chain transactions by utilizing relay blockchains as cross-chain connectors. Yet relay chain cross-chain transactions need to collect asset information and implement asset transactions via two-way peg. Due to the release of user transaction information, there is the issue of privacy leakage. In this article, we propose a cross-chain privacy protection protocol based on the Groth16 zero-knowledge proof algorithm and coin-mixing technology, which changes the authentication mechanism and uses a combination of generating functions to map virtual external addresses in transactions. It allows fast cross-chain anonymous transactions while hiding the genuine user’s address. The experiment shows that, in a zero-trust IoT context, our scheme can effectively protect user privacy information, accomplish controlled transaction traceability operations, and guarantee cross-chain transaction security.

Behaviour Based Malware Detection using Machine Learning

Abstract: Considering all the researches done, it appears that over last decade, malware has been growing exponentially and also has been causing significant financial losses to different organizations. Thus, it becomes important to detect if a file contains any malware or not. The malwares can cause a lot of damage to the system such as slowing downthe system and also stealing sensitive information from the system. malware is a executable program specifically to destroy a genuine user's computer by spreading harmful virusin different ways .In the current times, one of the most important assets of the people is their data and information which needs to be protected. Hence, in order to protect the data and information, there is a need for software which couldperform this task and help in ensuring the integrity of our system. Our method for malware detection uses different machine learning algorithms such as decision tree, random forest etc. The algorithm which has the maximum accuracy gets selected which provides a great detection ratio for the system. Furthermore, the performance of the system is detected by calculating the false positive and false negative rates using the confusion matrix.

A Novel Approach for Detection of DoS / DDoS Attack in Network Environment using Ensemble Machine Learning Model

One of the most serious threat to network security is Denial of service (DOS) attacks. Internet and computer networks are now important parts of our businesses and daily lives. Malicious actions have become more common as our reliance on computers and communication networks has grown. Network threats are a big problem in the way people communicate today. To make sure that the networks work well and that users' information is safe, the network data must be watched and analysed to find malicious activities and attacks. Flooding may be the simplest DDoS assault. Computer networks and services are vulnerable to DoS and DDoS attacks. These assaults flood target systems with malicious traffic, making them unreachable to genuine users. The work aims to enhance the resilience of network infrastructures against these attacks and ensure uninterrupted service delivery. This research develops and evaluates enhanced DoS/DDoS detection methods. DoS attacks usually stop or slow down legal computer or network use. Denial-of-service (DoS) attacks prevent genuine users from accessing and using information systems and resources. The OSI model's layers make up the computer network. Different types of DDoS strikes target different layers. The Network Layer can be broken by using ICMP Floods or Smurf Attacks. The Transport layer can be attacked using UDP Floods, TCP Connection Exhaustion, and SYN Floods. HTTP-encrypted attacks can be used to get through to the application layer. DoS/DDoS attacks are malicious attacks. Protect network data from harm. Computer network services are increasingly threatened by DoS/DDoS attacks. Machine learning may detect prior DoS/DDoS attacks. DoS/DDoS attacks proliferate online and via social media. Network security is IT's top priority. DoS and DDoS assaults include ICMP, UDP, and the more prevalent TCP flood attacks. These strikes must be identified and stopped immediately. In this work, a stacking ensemble method is suggested for detecting DoS/DDoS attacks so that our networked data doesn't get any worse. This paper used a method called "Ensemble of classifiers," in which each class uses a different way to learn. In proposed methodology Experiment#1 , I used the Home Wifi Network Traffic Collected and generated own Dataset named it as MywifiNetwork.csv, whereas in proposed methodology Experiment#2, I used the kaggle repository “NSL-KDD benchmark dataset” to perform experiments in order to find detection accuracy of dos attack detection using python language in jupyter notebook. The system detects attack-type or legitimate-type of network traffic during detection ML classification methods are used to compare how well the suggested system works. The results show that when the ensembled stacking learning model is used, 99% of the time it is able to find the problem. In proposed methodology two Experiments are implemented for comparing detection accuracy with the existing techniques. Compared to other measuring methods, we get a big step forward in finding attacks. So, our model gives a lot of faith in securing these networks. This paper will analyse the behaviour of network traffics.

VibPath

Technical advances in the smart device market have fixated smartphones at the heart of our lives, warranting an ever more secure means of authentication. Although most smartphones have adopted biometrics-based authentication, after a couple of failed attempts, most users are given the option to quickly bypass the system with passcodes. To add a layer of security, two-factor authentication (2FA) has been implemented but has proven to be vulnerable to various attacks. In this paper, we introduce VibPath, a simultaneous 2FA scheme that can understand the user's hand neuromuscular system through touch behavior. VibPath captures the individual's vibration path responses between the hand and the wrist with the attention-based encoder-decoder network, authenticating the genuine users from the imposters unobtrusively. In a user study with 30 participants, VibPath achieved an average performance of 0.98 accuracy, 0.99 precision, 0.98 recall, 0.98 f1-score for user verification, and 94.3% accuracy for user identification across five passcodes. Furthermore, we also conducted several extensive studies, including in-the-wile, permanence, vulnerability, usability, and system overhead studies, to assess the practicability and viability of the VibPath from multiple aspects.

Digital Technology and Copyright Protection the Challenges of Indian Performers Rights Society

Indian music is ever-green. Every restaurant, pub. Airports, stadium, television etc plays them making them world famous. India has 465 FM Radio stations which constantly play Indian Bollywood music. Though the song becomes well known the singer who sang the song is not recognized nor does he or she receive sufficient compensation for the song. The artist in India is the one who works so hard but ends up getting nothing. The ‘Indian Copyright law was amended in 2012’ to acknowledge the rights of performers. The singers were recognized as performers. The law provided for royalty to be paid to the singers. In reality this royalty never reached the singers. The problem for the singer was identifying the user playing his song and the problem for a genuine user was identifying the location of the singer to pay the royalty directly to him. Therefore, there was a need for an independent body to administer the copyright and collect royalty. ISRA is the bridge between the singer and the user which helps in collecting royalty. This article discusses the issues faced by the singers and the need to develop a smother collective administration of copyright.