Ajit Kumar Singh Yadav
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ajit Kumar Singh Yadav has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- D. Sivaganesan
Department of Computer Science and Engineering, Dr.Mahalingam College of Engineering and Technology, Pollachi, India
3 shared publications - Selligoundanur Subramaniyam Sivaraju
Department of Electrical and Electronics Engineering, RVS College of Engineering and Technology, Kannampalayam, Coimbatore, India
3 shared publications - Balaji Venkatesalu Ramasamy
Department of ECE, Sri Krishna College of Engineering and Technology, Coimbatore, India
2 shared publications - Marpe Sora
Rajiv Gandhi Univ, Dept Comp Sci & Engn, Itanagar, Arunachal Prade, India
2 shared publications - Srithar Samidurai
Department of Computer Science and Engineering, Koneru Lakshmaiah Education Foundation, Vaddeswaram, India
2 shared publications - V. Mani
Department of Electrical and Electronics Engineering, SNS College of Engineering, Coimbatore, India
2 shared publications
Latest Publications
- 25 March 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (06)
Effective data management using heuristic predictive modeling for security applications in games
Secure and private user data are more important than ever with the explosion of online gaming platforms and the resulting deluge of user information. Intending to protect gaming ecosystems and maintain user confidence, Heuristic Predictive Modeling provides a proactive security strategy by allowing early detection and mitigation of potential risks. The ever-changing nature of the game, the wide variety of user interactions, and the always-evolving strategies of cybercriminals all contribute to the singular problems that data management and security encounter in modern gaming settings. This research proposes Heuristic Predictive Modeling for Gaming Security (HPM-GS). This system can analyze gaming data in real time and detect trends and abnormalities that could indicate security breaches. It uses advanced algorithms and machine learning approaches. With HPM-GS, gaming platforms can keep their users safe and secure by anticipating and proactively addressing security threats. Several areas of gaming security can benefit from HPM-GS, such as user authentication, detection of cheats, prevention of fraud, and incident response. Enhanced user experience and platform reliability can be achieved by incorporating HPM-GS into pre-existing security frameworks, which allows gaming platforms to strengthen their defenses and efficiently reduce risks. Extensive simulation studies assess the effectiveness of HPM-GS in gaming security. The performance metrics of HPM-GS, such as detection accuracy, false positive rates, and response time, are evaluated using real-world datasets and simulated attack scenarios. The simulation findings show that HPM-GS is a good solution for protecting gaming environments from cyber-attacks. The HPM-GS is a proactive, elastic gaming application data management and security method. The purpose of this research is to emphasize the potential of HPM-GS to improve the security posture of online gaming platforms and to ensure that players have a gaming experience that is both safer and more pleasant. This is accomplished by addressing the significance of HPM-GS, potential difficulties, proposed techniques, implementations, and simulation analysis.
- 25 March 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (06)
Effective data management using heuristic predictive modeling for security applications in games
Secure and private user data are more important than ever with the explosion of online gaming platforms and the resulting deluge of user information. Intending to protect gaming ecosystems and maintain user confidence, Heuristic Predictive Modeling provides a proactive security strategy by allowing early detection and mitigation of potential risks. The ever-changing nature of the game, the wide variety of user interactions, and the always-evolving strategies of cybercriminals all contribute to the singular problems that data management and security encounter in modern gaming settings. This research proposes Heuristic Predictive Modeling for Gaming Security (HPM-GS). This system can analyze gaming data in real time and detect trends and abnormalities that could indicate security breaches. It uses advanced algorithms and machine learning approaches. With HPM-GS, gaming platforms can keep their users safe and secure by anticipating and proactively addressing security threats. Several areas of gaming security can benefit from HPM-GS, such as user authentication, detection of cheats, prevention of fraud, and incident response. Enhanced user experience and platform reliability can be achieved by incorporating HPM-GS into pre-existing security frameworks, which allows gaming platforms to strengthen their defenses and efficiently reduce risks. Extensive simulation studies assess the effectiveness of HPM-GS in gaming security. The performance metrics of HPM-GS, such as detection accuracy, false positive rates, and response time, are evaluated using real-world datasets and simulated attack scenarios. The simulation findings show that HPM-GS is a good solution for protecting gaming environments from cyber-attacks. The HPM-GS is a proactive, elastic gaming application data management and security method. The purpose of this research is to emphasize the potential of HPM-GS to improve the security posture of online gaming platforms and to ensure that players have a gaming experience that is both safer and more pleasant. This is accomplished by addressing the significance of HPM-GS, potential difficulties, proposed techniques, implementations, and simulation analysis.
- 11 March 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (03)
Cloud-enabled fog computing framework with wireless sensor networks for data center systems on IoT platform
The main idea of this framework is that it is capable of overcoming the drawbacks that are always linked with conventional cloud-based methods. Computation and storage resources in Internet of Things (IoT) networks are distributed closer to the network’s edge; therefore, the amount of data processed in real time is reduced. By decreasing the distance of the data transfers, less bandwidth is used. Structural problems, data safety, interoperability, and resource allocation-related matters denote challenges preventing the successful implementation of those ideas. The proposed work is the cloud-enabled fog computing framework (C-FCF) in data center systems based on the IoT platform. It brings cloud computing to a new level of scalability by compounding the following: Scalable architecture, uniform communication interfaces, the dynamic algorithms that allocate resources, and the data-centered approach, on the one hand, and strong security protocols, on the other. The wireless sensor network (WSN) approach to this technology represents a greater versatility of this system as it can be applied to perform different tasks in various industries like smart cities, healthcare, transportation, and industrial automation services. The application of the given services illustrates C-FCF’s capability of creating innovation, modeling effectiveness, and uncovering potential for integration within the IoT network. The virtual simulation analysis is necessary to validate C-FCF’s effectiveness in real-life scenarios. The simulations provide evidence of features, including low latency, efficient resource utilization, and overall system performance, which underlines the practical aspects of applying C-FCF in different IoT settings. Developing this advanced computing architecture, which can surpass the limitations of conventional technology and the ability to entail many different use cases, will potentially change the data processing and management paradigm in IoT-enabled settings.
- 11 March 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (03)
Cloud-enabled fog computing framework with wireless sensor networks for data center systems on IoT platform
The main idea of this framework is that it is capable of overcoming the drawbacks that are always linked with conventional cloud-based methods. Computation and storage resources in Internet of Things (IoT) networks are distributed closer to the network’s edge; therefore, the amount of data processed in real time is reduced. By decreasing the distance of the data transfers, less bandwidth is used. Structural problems, data safety, interoperability, and resource allocation-related matters denote challenges preventing the successful implementation of those ideas. The proposed work is the cloud-enabled fog computing framework (C-FCF) in data center systems based on the IoT platform. It brings cloud computing to a new level of scalability by compounding the following: Scalable architecture, uniform communication interfaces, the dynamic algorithms that allocate resources, and the data-centered approach, on the one hand, and strong security protocols, on the other. The wireless sensor network (WSN) approach to this technology represents a greater versatility of this system as it can be applied to perform different tasks in various industries like smart cities, healthcare, transportation, and industrial automation services. The application of the given services illustrates C-FCF’s capability of creating innovation, modeling effectiveness, and uncovering potential for integration within the IoT network. The virtual simulation analysis is necessary to validate C-FCF’s effectiveness in real-life scenarios. The simulations provide evidence of features, including low latency, efficient resource utilization, and overall system performance, which underlines the practical aspects of applying C-FCF in different IoT settings. Developing this advanced computing architecture, which can surpass the limitations of conventional technology and the ability to entail many different use cases, will potentially change the data processing and management paradigm in IoT-enabled settings.
- 28 August 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
Noncooperative game-theoretic model for security enhancement in cloud gaming environments
The cloud-based gaming services have given rise to an abundance of fresh worries around data protection and safety. The present cloud gaming security solutions suffer from a key issue in that they do not include a comprehensive framework. This framework would take into consideration the dynamic and interactive nature of the data center environment. When trying to mitigate risks in today’s continuously moving cyber threat scenario, it is imperative to consider all relevant factors, including the health of the system, individual costs, and potential advantages. The existing works address about how game theory might be used to improve the safety of cloud-based gaming. The purpose of this research is to overcome the research gap by developing a novel model that captures the multiple linkages between security measures and potential attacks, supplying a more holistic perspective on cloud gaming security in the process. In this line of investigation, a Noncooperative Game-Theoretic Model is used to investigate the dynamics of predators and prey. In this research, a Noncooperative Game-Theoretic Model is presented with the purpose of analysing and perfecting security measures inside the dynamic data-center environment of cloud gaming. Precautionary safety measures take on the role of a predator, while potential threats take on the role of prey. The results prove that the Noncooperative Game-Theoretic Model is an effective tool for enhancing safeguards in cloud-based online gaming environments.
- 28 August 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
Noncooperative game-theoretic model for security enhancement in cloud gaming environments
The cloud-based gaming services have given rise to an abundance of fresh worries around data protection and safety. The present cloud gaming security solutions suffer from a key issue in that they do not include a comprehensive framework. This framework would take into consideration the dynamic and interactive nature of the data center environment. When trying to mitigate risks in today’s continuously moving cyber threat scenario, it is imperative to consider all relevant factors, including the health of the system, individual costs, and potential advantages. The existing works address about how game theory might be used to improve the safety of cloud-based gaming. The purpose of this research is to overcome the research gap by developing a novel model that captures the multiple linkages between security measures and potential attacks, supplying a more holistic perspective on cloud gaming security in the process. In this line of investigation, a Noncooperative Game-Theoretic Model is used to investigate the dynamics of predators and prey. In this research, a Noncooperative Game-Theoretic Model is presented with the purpose of analysing and perfecting security measures inside the dynamic data-center environment of cloud gaming. Precautionary safety measures take on the role of a predator, while potential threats take on the role of prey. The results prove that the Noncooperative Game-Theoretic Model is an effective tool for enhancing safeguards in cloud-based online gaming environments.
- 21 August 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (06)
Distributed denial of service (DDOS) attacks and mitigation method using logistic regression-based GoogLeNet for real time in security games
Distributed Denial of Service (DDoS) attacks remain a persistent and formidable threat in the ever-changing world of cyber security. These attacks have the potential to disrupt internet services and cause substantial financial and reputational concerns. The major challenge is developing an adaptable and real-time Intrusion Detection System (IDS) that can detect and neutralize DDoS attacks effectively and quickly, even when attackers use increasingly advanced ways to avoid detection. The problem concerns the development of a dynamic and real-time intrusion detection technique that combines the benefits of logistic regression for anomaly detection with GoogLeNet for deep learning-based network traffic analysis. This paper proposes a unique framework for intrusion detection that blends logistic regression-based anomaly detection with GoogLeNet deep learning capabilities. The combination of these technologies makes it easier to identify and mitigate DDoS attacks, hence improving the security of internet-based systems. The proposed IDS framework utility is proved through experimental evaluations, which highlight its capacity to effectively identify DDoS attacks while minimizing false positives. The use of this technology in real-time during security games demonstrates its potential to improve online service security infrastructure and reduce the impact of DDoS attacks on critical networks and data resources.
- 21 August 2024
- Published by World Scientific Pub Co Pte Ltd in International Journal of Modeling, Simulation, and Scientific Computing
- Vol. 17 (06)
Distributed denial of service (DDOS) attacks and mitigation method using logistic regression-based GoogLeNet for real time in security games
Distributed Denial of Service (DDoS) attacks remain a persistent and formidable threat in the ever-changing world of cyber security. These attacks have the potential to disrupt internet services and cause substantial financial and reputational concerns. The major challenge is developing an adaptable and real-time Intrusion Detection System (IDS) that can detect and neutralize DDoS attacks effectively and quickly, even when attackers use increasingly advanced ways to avoid detection. The problem concerns the development of a dynamic and real-time intrusion detection technique that combines the benefits of logistic regression for anomaly detection with GoogLeNet for deep learning-based network traffic analysis. This paper proposes a unique framework for intrusion detection that blends logistic regression-based anomaly detection with GoogLeNet deep learning capabilities. The combination of these technologies makes it easier to identify and mitigate DDoS attacks, hence improving the security of internet-based systems. The proposed IDS framework utility is proved through experimental evaluations, which highlight its capacity to effectively identify DDoS attacks while minimizing false positives. The use of this technology in real-time during security games demonstrates its potential to improve online service security infrastructure and reduce the impact of DDoS attacks on critical networks and data resources.
- 14 March 2024
- Published by Springer Nature
- p. 131-141
Financial Statement Fraud Detection Using Optimized Deep Neural Network
- 14 March 2024
- Published by Springer Nature
- p. 131-141
Financial Statement Fraud Detection Using Optimized Deep Neural Network
- 8 February 2023
- Published by Elsevier in Materials Today: Proceedings
Experimental study of a rapid rate biofuel production in a cyclone type reactor
- 8 February 2023
- Published by Elsevier in Materials Today: Proceedings
Experimental study of a rapid rate biofuel production in a cyclone type reactor
- 26 September 2022
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 34 (27)
Unsupervised learning for financial statement fraud detection using manta ray foraging based convolutional neural network
Summary: Financial statement fraud detection is a significant topic and a challenging task related to countless applications such as financial sectors, insurance, government agencies, law enforcement and so on. This article presents an unsupervised learning model to identify the financial fraud statements using the hybrid convolutional neural network. The fraud detection scheme can be performed using a few stages: data preprocessing, sampling, feature extraction, feature selection, clustering, and classification methods. The preprocessing and sampling processes are used to clean and recognize duplicate data. Then, the proposed model analyzes both text features and financial variables to provide the best classification of financial statement fraud. In this article, a new fuzzy red deer's algorithm is proposed for selecting the optimal feature set, which can improve the classification accuracy of the fraud prediction model. Subsequently, an adaptive density based clustering (ADBC) approach is introduced for labeling the selected features through the clustering process. Finally, the fraud statements are predicted by proposing a hybrid CNN‐MRFO model wherein the hyperparameters of the CNN model are optimized through the MRFO learning algorithm. The simulation results prove the effectiveness of the proposed model in terms of accuracy (98.85%), precision (98.62%) and F‐measure (99.32%).
- 26 September 2022
- Published by Wiley in Concurrency and Computation: Practice and Experience
- Vol. 34 (27)
Unsupervised learning for financial statement fraud detection using manta ray foraging based convolutional neural network
Summary: Financial statement fraud detection is a significant topic and a challenging task related to countless applications such as financial sectors, insurance, government agencies, law enforcement and so on. This article presents an unsupervised learning model to identify the financial fraud statements using the hybrid convolutional neural network. The fraud detection scheme can be performed using a few stages: data preprocessing, sampling, feature extraction, feature selection, clustering, and classification methods. The preprocessing and sampling processes are used to clean and recognize duplicate data. Then, the proposed model analyzes both text features and financial variables to provide the best classification of financial statement fraud. In this article, a new fuzzy red deer's algorithm is proposed for selecting the optimal feature set, which can improve the classification accuracy of the fraud prediction model. Subsequently, an adaptive density based clustering (ADBC) approach is introduced for labeling the selected features through the clustering process. Finally, the fraud statements are predicted by proposing a hybrid CNN‐MRFO model wherein the hyperparameters of the CNN model are optimized through the MRFO learning algorithm. The simulation results prove the effectiveness of the proposed model in terms of accuracy (98.85%), precision (98.62%) and F‐measure (99.32%).
- 24 November 2021
- Published by Area de Innovacion y Desarrollo, S.L. 3 Ciencias in 3C Empresa: investigación y pensamiento crítico
- Vol. 10 (4) , 77-105
An optimized deep neural network-based financial statement fraud detection in text mining
- 24 November 2021
- Published by Area de Innovacion y Desarrollo, S.L. 3 Ciencias in 3C Empresa: investigación y pensamiento crítico
- Vol. 10 (4) , 77-105
An optimized deep neural network-based financial statement fraud detection in text mining