Gajula Ramesh
Research Subjects
Gajula Ramesh has a total of 101 co-authors. Here are the top co-authors by number of shared publications:
K. Shyam Sunder Reddy
Department of CSE, Maturi Venkata Subba Rao (MVSR) Engineering College, Hyderabad, India
3 shared publications- Abhay Kumar Agarwal
Department of Computer Science and Engineering, Kamla Nehru Institute of Technology, Sultanpur, India
2 shared publications - Arjun Singh
Department of Computer and Communication Engineering, Manipal University Jaipur, Jaipur 303007, India
2 shared publications - Lipika Goel
Department of Computer Science and Engineering (AIML), Sridevi Women’s Engineering College, Hyderabad, Telangana, 500075, India
2 shared publications P. Surekha
Department of Computer Science and Engineering, Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad, India
2 shared publications- Prabhakar Kandukuri
Department of Artificial Intelligence and Machine Learning, Chaitanya Bharathi Institute of Technology, Hyderabad, India
2 shared publications
Latest Publications
- Open Access
- 31 July 2026
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 20 (1) , 151
Correction to: Optimized cortical EEG modeling for Parkinson disease diagnosis with snow Shepherd Stride tuning mechanism
- 31 July 2026
- Published by Springer Nature in Cognitive Neurodynamics
- Vol. 20 (1) , 151
Open AccessCorrection to: Optimized cortical EEG modeling for Parkinson disease diagnosis with snow Shepherd Stride tuning mechanism
- 23 February 2026
- Published by Edtech Publishers (OPC) Private Limited in International Journal of Scientific Research in Engineering and Management
- Vol. 10 (02) , 1-9
Social Engineering Simulator: Enhancing Cybersecurity Awareness Through Realistic Simulation
— Social engineering attacks continue to pose a serious threat to modern organizational security by exploiting human behavior rather than technical weaknesses. This paper presents the design and implementation of a Social Engineering Simulator developed to enhance cybersecurity awareness through realistic and controlled attack simulations. The proposed system replicates common attack vectors including phishing, vishing, smishing, impersonation, and baiting to evaluate user responses and improve decision-making skills in high-risk situations. The simulator integrates automated campaign management, behavioral analytics, real-time monitoring, and feedback mechanisms to strengthen organizational resilience against human- centric threats. A modular architecture consisting of frontend, backend, database, and detection layers supports scalability and efficient data processing. Machine learning–based analysis enables identification of risky patterns and supports adaptive training strategies tailored to user performance. Experimental evaluation demonstrates improvements in user awareness, reduced response time to suspicious activities, and measurable decline in unsafe interactions during repeated simulations. The findings indicate that immersive simulation-based training enhances preparedness compared to traditional awareness programs. The system provides a practical, scalable, proactive approach for organizations seeking to minimize human-factor vulnerabilities and build a strong security culture against evolving social engineering threats. Keywords—Social Engineering, Cybersecurity Awareness Training, Phishing Simulation, Human- Centric Security, Behavioral Security Analytics, Insider Threat Detection, Cyber Attack Simulation, Security Awareness Systems.
- 23 February 2026
- Published by Edtech Publishers (OPC) Private Limited in International Journal of Scientific Research in Engineering and Management
- Vol. 10 (02) , 1-9
Social Engineering Simulator: Enhancing Cybersecurity Awareness Through Realistic Simulation
— Social engineering attacks continue to pose a serious threat to modern organizational security by exploiting human behavior rather than technical weaknesses. This paper presents the design and implementation of a Social Engineering Simulator developed to enhance cybersecurity awareness through realistic and controlled attack simulations. The proposed system replicates common attack vectors including phishing, vishing, smishing, impersonation, and baiting to evaluate user responses and improve decision-making skills in high-risk situations. The simulator integrates automated campaign management, behavioral analytics, real-time monitoring, and feedback mechanisms to strengthen organizational resilience against human- centric threats. A modular architecture consisting of frontend, backend, database, and detection layers supports scalability and efficient data processing. Machine learning–based analysis enables identification of risky patterns and supports adaptive training strategies tailored to user performance. Experimental evaluation demonstrates improvements in user awareness, reduced response time to suspicious activities, and measurable decline in unsafe interactions during repeated simulations. The findings indicate that immersive simulation-based training enhances preparedness compared to traditional awareness programs. The system provides a practical, scalable, proactive approach for organizations seeking to minimize human-factor vulnerabilities and build a strong security culture against evolving social engineering threats. Keywords—Social Engineering, Cybersecurity Awareness Training, Phishing Simulation, Human- Centric Security, Behavioral Security Analytics, Insider Threat Detection, Cyber Attack Simulation, Security Awareness Systems.
- 19 August 2025
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 222,p. e68426
Hybrid Convolutional Neural Network and Generative Adversarial Network Framework for Robust Deepfake Detection: Advancing Accuracy, Scalability, and Multimodal Integration
Deepfakes pose critical threats to digital media integrity and societal trust. This paper presents a hybrid deepfake detection framework combining Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) to address challenges in scalability, generalizability, and adversarial robustness. The framework integrates adversarial training, a temporal decay analysis model, and multimodal detection across audio, video, and text domains. Evaluated on the FaceForensics++ benchmark dataset, the hybrid CNN-GAN model achieved state-of-the-art performance with 98.5% accuracy, 99.1% precision, 97.9% recall, and a Receiver Operating Characteristic-Area Under Curve of 0.995, surpassing conventional models like ResNet50, XceptionNet, and EfficientNet. The temporal analysis component models the decay of deepfake detectability over time, guiding proactive retraining, while the multimodal module identified 97.8% of cross-modal inconsistencies. Simulation of deepfake propagation in social networks revealed significant socio-technical impacts, emphasizing the need for early detection systems and public awareness campaigns to curb misinformation spread. Optimized for real-time deployment, the framework balances high scalability with ethical considerations. This study highlights the importance of interdisciplinary approaches to deepfake detection in advancing AI-driven safeguards for digital ecosystems. Future work will focus on developing lightweight detection models, improving adversarial defenses, and expanding domain-specific applications to enhance the framework's applicability and resilience.
- 19 August 2025
- Published by MyJove Corporation in Journal of Visualized Experiments
- No. 222,p. e68426
Hybrid Convolutional Neural Network and Generative Adversarial Network Framework for Robust Deepfake Detection: Advancing Accuracy, Scalability, and Multimodal Integration
Deepfakes pose critical threats to digital media integrity and societal trust. This paper presents a hybrid deepfake detection framework combining Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) to address challenges in scalability, generalizability, and adversarial robustness. The framework integrates adversarial training, a temporal decay analysis model, and multimodal detection across audio, video, and text domains. Evaluated on the FaceForensics++ benchmark dataset, the hybrid CNN-GAN model achieved state-of-the-art performance with 98.5% accuracy, 99.1% precision, 97.9% recall, and a Receiver Operating Characteristic-Area Under Curve of 0.995, surpassing conventional models like ResNet50, XceptionNet, and EfficientNet. The temporal analysis component models the decay of deepfake detectability over time, guiding proactive retraining, while the multimodal module identified 97.8% of cross-modal inconsistencies. Simulation of deepfake propagation in social networks revealed significant socio-technical impacts, emphasizing the need for early detection systems and public awareness campaigns to curb misinformation spread. Optimized for real-time deployment, the framework balances high scalability with ethical considerations. This study highlights the importance of interdisciplinary approaches to deepfake detection in advancing AI-driven safeguards for digital ecosystems. Future work will focus on developing lightweight detection models, improving adversarial defenses, and expanding domain-specific applications to enhance the framework's applicability and resilience.
- 19 August 2025
- Published by MyJove Corporation
- No. 222,p. e68426
JoVE Video Dataset
Deepfakes pose critical threats to digital media integrity and societal trust. This paper presents a hybrid deepfake detection framework combining Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) to address challenges in scalability, generalizability, and adversarial robustness. The framework integrates adversarial training, a temporal decay analysis model, and multimodal detection across audio, video, and text domains. Evaluated on the FaceForensics++ benchmark dataset, the hybrid CNN-GAN model achieved state-of-the-art performance with 98.5% accuracy, 99.1% precision, 97.9% recall, and a Receiver Operating Characteristic-Area Under Curve of 0.995, surpassing conventional models like ResNet50, XceptionNet, and EfficientNet. The temporal analysis component models the decay of deepfake detectability over time, guiding proactive retraining, while the multimodal module identified 97.8% of cross-modal inconsistencies. Simulation of deepfake propagation in social networks revealed significant socio-technical impacts, emphasizing the need for early detection systems and public awareness campaigns to curb misinformation spread. Optimized for real-time deployment, the framework balances high scalability with ethical considerations. This study highlights the importance of interdisciplinary approaches to deepfake detection in advancing AI-driven safeguards for digital ecosystems. Future work will focus on developing lightweight detection models, improving adversarial defenses, and expanding domain-specific applications to enhance the framework's applicability and resilience.
- 19 August 2025
- Published by MyJove Corporation
- No. 222,p. e68426
JoVE Video Dataset
Deepfakes pose critical threats to digital media integrity and societal trust. This paper presents a hybrid deepfake detection framework combining Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) to address challenges in scalability, generalizability, and adversarial robustness. The framework integrates adversarial training, a temporal decay analysis model, and multimodal detection across audio, video, and text domains. Evaluated on the FaceForensics++ benchmark dataset, the hybrid CNN-GAN model achieved state-of-the-art performance with 98.5% accuracy, 99.1% precision, 97.9% recall, and a Receiver Operating Characteristic-Area Under Curve of 0.995, surpassing conventional models like ResNet50, XceptionNet, and EfficientNet. The temporal analysis component models the decay of deepfake detectability over time, guiding proactive retraining, while the multimodal module identified 97.8% of cross-modal inconsistencies. Simulation of deepfake propagation in social networks revealed significant socio-technical impacts, emphasizing the need for early detection systems and public awareness campaigns to curb misinformation spread. Optimized for real-time deployment, the framework balances high scalability with ethical considerations. This study highlights the importance of interdisciplinary approaches to deepfake detection in advancing AI-driven safeguards for digital ecosystems. Future work will focus on developing lightweight detection models, improving adversarial defenses, and expanding domain-specific applications to enhance the framework's applicability and resilience.
- Open Access
- 8 November 2024
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 588, 03013
Synergistic Effect of Graphene and Boron Nitride on the Electrical Conductivity and Mechanical Properties of Hybrid Polymer Composites
This work examines the combined influence of graphene and hexagonal boron nitride (h-BN) on the electrical conductivity, mechanical characteristics, and thermal stability of polyvinylidene fluoride (PVDF)-based hybrid polymer composites. Graphene and h-BN were exfoliated and amalgamated in different ratios (1:1, 2:1, and 3:1), thereafter integrated into the PVDF matrix at nanofiller concentrations of 0.5 wt%, 1.0 wt%, 2.0 wt%, and 5.0 wt%. Mechanical tests demonstrated substantial improvements in tensile strength and Young’s modulus, especially for composites with a 2:1 graphene-to-h-BN ratio. Electrical conductivity significantly enhanced with increasing nanofiller content, reaching a peak of 5.0 × 10⁻² S/m at a 5.0 wt% nanofiller concentration with a 3:1 graphene-to-h-BN ratio. Thermal stability has also improved with degradation temperatures increased by up to 70°C more as compared to PVDF. The findings demonstrate that the ideal nanofiller structure (2:Adding one graphene ratio at 5.0 wt% strikes a balance of these properties enabling the composites to be applied in electronics, sensors, and advanced structures
- 8 November 2024
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 588, 03013
Open AccessSynergistic Effect of Graphene and Boron Nitride on the Electrical Conductivity and Mechanical Properties of Hybrid Polymer Composites
This work examines the combined influence of graphene and hexagonal boron nitride (h-BN) on the electrical conductivity, mechanical characteristics, and thermal stability of polyvinylidene fluoride (PVDF)-based hybrid polymer composites. Graphene and h-BN were exfoliated and amalgamated in different ratios (1:1, 2:1, and 3:1), thereafter integrated into the PVDF matrix at nanofiller concentrations of 0.5 wt%, 1.0 wt%, 2.0 wt%, and 5.0 wt%. Mechanical tests demonstrated substantial improvements in tensile strength and Young’s modulus, especially for composites with a 2:1 graphene-to-h-BN ratio. Electrical conductivity significantly enhanced with increasing nanofiller content, reaching a peak of 5.0 × 10⁻² S/m at a 5.0 wt% nanofiller concentration with a 3:1 graphene-to-h-BN ratio. Thermal stability has also improved with degradation temperatures increased by up to 70°C more as compared to PVDF. The findings demonstrate that the ideal nanofiller structure (2:Adding one graphene ratio at 5.0 wt% strikes a balance of these properties enabling the composites to be applied in electronics, sensors, and advanced structures
- 9 August 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (20) , 22853-22891
Deep learning based RAGAE-SVM for Chronic kidney disease diagnosis on internet of health things platform
- 9 August 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (20) , 22853-22891
Deep learning based RAGAE-SVM for Chronic kidney disease diagnosis on internet of health things platform
- 1 August 2024
- Published by Scalable Computing: Practice and Experience in Scalable Computing: Practice and Experience
- Vol. 25 (5) , 3701-3714
Optimizing Task Scheduling: Exploring Advanced Machine Learning in Dew-Powered Cloud Environments
Research into Dew computing environments has recently emerged as a result of the increasing prevalence and processing power of mobile and IoT devices. In these settings, even low-powered devices can share some of their computational resources with their neighbors. This paper proposes a novel approach to workflow scheduling in dew enabled cloud computing environment, called Deep Q-learning (DQN) + Chronological Geese Migration Optimization (CGMO). DQN is a deep learningbased method for scheduling workflows, while CGMO is a hybrid optimization algorithm that combines the chronological idea and the Wild Geese Migration Optimization (GMO) algorithm. The proposed approach aims to optimize multiple objectives, including predicted energy, Quality of Service (QoS), and resource usage, by scheduling workflows in the cloud. The approach also takes into account the current state of the Virtual Machine (VM) and the job. The assessment measures employed for DCGM include maximum QoS, minimum energy usage, and maximum resource utilization. The results show that DCGM achieved the highest QoS (0.865), lowest energy usage (0.0322), and highest resource utilization (1.000) compared to other approaches.
- 1 August 2024
- Published by Scalable Computing: Practice and Experience in Scalable Computing: Practice and Experience
- Vol. 25 (5) , 3701-3714
Optimizing Task Scheduling: Exploring Advanced Machine Learning in Dew-Powered Cloud Environments
Research into Dew computing environments has recently emerged as a result of the increasing prevalence and processing power of mobile and IoT devices. In these settings, even low-powered devices can share some of their computational resources with their neighbors. This paper proposes a novel approach to workflow scheduling in dew enabled cloud computing environment, called Deep Q-learning (DQN) + Chronological Geese Migration Optimization (CGMO). DQN is a deep learningbased method for scheduling workflows, while CGMO is a hybrid optimization algorithm that combines the chronological idea and the Wild Geese Migration Optimization (GMO) algorithm. The proposed approach aims to optimize multiple objectives, including predicted energy, Quality of Service (QoS), and resource usage, by scheduling workflows in the cloud. The approach also takes into account the current state of the Virtual Machine (VM) and the job. The assessment measures employed for DCGM include maximum QoS, minimum energy usage, and maximum resource utilization. The results show that DCGM achieved the highest QoS (0.865), lowest energy usage (0.0322), and highest resource utilization (1.000) compared to other approaches.
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3101 (1) , 050004
Clustering algorithms in data science: Evaluating the time and space complexities of K-means, DBSCAN, and hierarchical methods
In the expansive domain of data science, clustering algorithms play a pivotal role in segmenting datasets into meaningful groups without prior knowledge of thei In the expansive domain of data science, clustering algorithms play a pivotal role in segmenting datasets into meaningful groups without prior knowledge of thei
- 1 January 2024
- Published by AIP Publishing in AIP Conference Proceedings
- Vol. 3101 (1) , 050004
Clustering algorithms in data science: Evaluating the time and space complexities of K-means, DBSCAN, and hierarchical methods
In the expansive domain of data science, clustering algorithms play a pivotal role in segmenting datasets into meaningful groups without prior knowledge of thei In the expansive domain of data science, clustering algorithms play a pivotal role in segmenting datasets into meaningful groups without prior knowledge of thei
- 20 December 2023
- Published by Springer Nature in Earth Science Informatics
- Vol. 17 (1) , 711-723
Enhanced Elliptic Curve- Diffie Hellman Technique with Bigdata Analytics for Satellite Image Security Enhancement in Internet of Things Systems
- 20 December 2023
- Published by Springer Nature in Earth Science Informatics
- Vol. 17 (1) , 711-723
Enhanced Elliptic Curve- Diffie Hellman Technique with Bigdata Analytics for Satellite Image Security Enhancement in Internet of Things Systems