AI-Automated Proctoring and Evaluation Model for Digital Education Ecosystems Vanshika Rastogi, Tushar Singh, Satyam Srivastav, Sreesh Gaur 2025 2nd International Conference on Advanced Computing and Emerging Technologies Acet 2025, 2025 With the growing adoption of digital learning platforms, there is an increasing need for online examination systems that are both secure and widely accessible. This study outlines and proposes a system that uses natural language processing (NLP) to automatically evaluate both objective questions and descriptive answers. Furthermore, it leverages computer vision models to facilitate real-time monitoring. Experiments conducted on 200 exam sessions across 100 participants achieved a detection accuracy of 93.2%, demonstrating around a 20% decrease in false positives compared to prior systems. The platform is designed to scale effectively for large deployments, providing cloud-based support with reliable latency even during peak usage times. The proposed framework uniquely combines multi-modal AI monitoring with automated assessment and embedded privacy safeguards into one unified platform. The results demonstrate significant improvements in both efficiency and reliability, positioning this system as a viable option for trustworthy and ethical online assessments.
Skin Diseases Classification with Deep Learning Techniques: A Review R Thanuja Reddy, Lavanya U, Anusha G, Ummehani S, Vanshika Rastogi, Anand R Proceedings of 2025 International Conference on Emerging Technologies in Computing and Communication Etcc 2025, 2025 One major health problem is the frequency of skin disorders. For treatment to be effective, a precise diagnosis is essential. Traditional diagnostic techniques, which frequently rely on dermoscopy and visual inspection, can be laborious and subjective. One possible method for automated skin disease diagnostics is a subset of artificial intelligence. This review explores the use of deep learning methods in this sector by looking at the many kinds of skin conditions. datasets using augmentation and preprocessing methods performance benchmarks and criteria for evaluating model designs We also go into the drawbacks and difficulties, such as interpretability of data quality models and ethical issues. By utilizing deep learning, we can increase the precision and effectiveness of diagnosing skin diseases, which will enhance patient outcomes.
Real-Time Driver Intention Prediction for Lane Merging using Deep Neural Networks Kirankumar G. Sutar, M. Prabha, Vanshika Rastogi, Nirmala G, S Murugan, Sathishkumar V E Proceedings of 5th International Conference on Soft Computing for Security Applications Icscsa 2025, 2025 This paper introduces a real-time framework for predicting driver intentions during lane merging using deep neural networks (DNNs). The main goal is to improve road safety and optimize traffic flow by accurately predicting a driver's intention to merge. This approach examines essential data inputs, including vehicle velocity, acceleration, surrounding traffic dynamics, and driver behavioral indicators. Trained the DNN model on a comprehensive dataset that includes diverse driving scenarios and conditions. Through extensive testing, the model achieved notable enhancements in predictive accuracy, attaining a 98.96% accuracy rate in reliable intention detection compared to traditional methods. The results indicate that utilizing advanced deep learning (DL) methodologies may significantly enhance the advancement of intelligent driver-assistance systems (ADAS). It highlights the significance of real-time processing capabilities in automotive applications, enabling improved safety measures and informed decision-making while driving. Future work will focus on verifying the concept in real-world scenarios.
Facial Landmark In Emotion Analysis: A Review Bhumika S, Manu Shree N, Suma K, Varsha M, Vanshika Rastogi, Anand R Proceedings of 2025 International Conference on Emerging Technologies in Computing and Communication Etcc 2025, 2025 Facial landmarks are essential for emotion analysis, offering a framework for interpreting facial expressions in facial emotion recognition systems. This review examines the historical progression of emotion detection methods, including traditional techniques like Active Shape Models and modern approaches using Convolutional Neural Networks. We explore feature extraction methods, emotion classification techniques, and the challenges of accuracy, data imbalance, and real-time processing. Future research directions encompass advanced model architectures, multimodal integration, and ethical considerations. This review emphasizes the critical role of facial landmarks in enhancing emotion recognition applications.
Detection and classification of COVID-19 disease using SWHO-based deep neural network classifier Vanshika Rastogi, Ajit Kumar Jain Computer Methods in Biomechanics and Biomedical Engineering Imaging and Visualization, 2023 Corona is an unanticipated disease that invaded the lives of millions of people and caused a global pandemic. Along with that, the disease affected the normal lifestyle and initiated a massive economic crisis. In this research, COVID-19 disease detection and severity identification are performed using the proposed SWHO-based deep Neural Network (SWHO-based deep NN) classifier. In this optimised deep NN classifier, the network parameters of the deep NN classifier are optimised using the Spadger Wolf Hawk Optimization (SWHO), which tunes the weight and bias of the classifier. The importance of the SWHO algorithm relies on faster convergence and less time is taken for the computation. Moreover, the severity of corona is classified based on mild, moderate, and severe classes using the SWHO-based deep NN, which helps medical professionals to equip the patients based on their necessity. The severity analysis is performed in this research, and the proficiency of the research is analysed based on the performance measures, accuracy, sensitivity, and specificity. The proposed method acquired the accuracy, sensitivity, and specificity of 92.809%, 95.082%, and 96.296% in terms of k-fold and 95.870%, 96.875%, and 98.800% in terms of training percentage, respectively. The proposed method effectively analysed, predicted, and classified the disease efficiently.
Covid-19 Classification Model Based on Age and Gender Analysis Using SWHO-Based Deep CNN Vanshika Rastogi, Ajit Kumar Jain Proceedings 2023 3rd International Conference on Pervasive Computing and Social Networking Icpcsn 2023, 2023 Nowadays, Corona Virus Disease-19 (COVID-19) disease is considered one of the main health issues, which is spreading in the worldwide environment, and due to this, infected patients are isolated to avoid spreading the disease. So, instant recognition is required for classifying the disease in a precise manner. Hence, the SWHO deep convolutional neural network (CNN) classification is developed to analyze the disease and the process of rising the COVID-19 classification scheme commences with collecting the blood samples and proceeds through preprocessing, feature extraction, as well as classification. Thus, the accuracy, sensitivity, as well as specificity of the developed method attained the value of 89.797%, 92.351%, and 88.252%, at 80% of training.
Analysis of the Impacts of COVID-19 using Deep Convolutional Neural Network Vanshika Rastogi, Ajit Kumar Jain 2nd International Conference on Sustainable Computing and Data Communication Systems Icscds 2023 Proceedings, 2023 COVID-19 is the transmittable disease that emerged as a recent epidemic and threatened the lives of various people. The emerged pandemic initiated a change in the people’s routine and impacted a serious financial crisis. This initiated a necessity for developing a deeper insight of the COVID-19 disease and multiple researches are performed based on the COVID-19 epidemic, which possess the challenges of basic analysis of information about the disease, lack of data, lack of knowledge about the parameters that cause disease and to overcome this a deep COVID-19 analysis epidemic via the deep CNN classifier is accomplished in the research. The impact of the disease is examined based on the gender, age group, symptoms and outbreak of the disease. This analysis provides comprehensive information about the disease and helps in making the preventive measures, which will greatly reduce the impacts of the disease. The accomplishment of deep CNN instinctively analyzes the essential features needed for the classification that helps in reducing the effort and time of the individuals. The performance is analyzed with the metrics specificity, accuracy and sensitivity, which obtained values of 0.48 %, 0.27 %, 2.82 % corresponding to and 2.88 %, 1.5 %, 0.36% considering training percentage, which is more efficient.
Android Malware Detection N Gagan, S Sai Kumar, M Keerthana, Samarth S Vaidya, Vanshika Rastogi 2023 World Conference on Communication and Computing Wconf 2023, 2023 Since android smartphones are so popular, malware writers have found them to be lucrative targets, which poses a serious risk to user security and privacy. To address this problem, several malware detection strategies have been put out recently. In this review paper, we examine an efficient method for detecting malware on Android that makes use of machine learning techniques. Our method divides Android applications into benign and harmful categories using a combination of feature extraction methods and machine learning algorithms. Overall, our suggested method offers a dependable and efficient alternative for detecting Android malware, giving consumers a more dependable and safe environment for mobile computing.
Video To Braille Transcription For Visually Impaired People Ishika Verma, Astha Rai, Himani Yadav, Vanshika Rastogi, Sugandha Satija 2021 International Conference on Simulation Automation and Smart Manufacturing Sasm 2021, 2021 Visual disability is a global issue. The number of visually impaired people is estimated to be around 285 million in the world. Out of which 246 million people have poor vision and 39 million are completely blind. The statistics of visually impaired people. Ninety percent of this population lives in developing countries. The difficulties faced by them in using the latest technologies and how they dealt with these are largely unknown and under-explored, especially in the developing world. Availability of video to Braille transcription anytime, anywhere, provides a potentially life-changing opportunity for the visually challenged to improve their communication capability. We are introducing software for hearing and visually impaired people to transcribe video to printed Braille scripts. It is evident that implementation of such a system for visually handicapped people with a productive method of Real-time speech to Braille transcription, text to speech, text to Braille, video to Braille conversion, work on commands with flexibility will be of great use
RECENT SCHOLAR PUBLICATIONS
Skin Diseases Classification with Deep Learning Techniques: A Review RT Reddy, L U, A G, U S, V Rastogi, A R 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025.0
Facial Landmark In Emotion Analysis: A Review BS Vanshika Rastogi, M Shree N, S K, V M, A R 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025.0
Real-Time Driver Intention Prediction for Lane Merging using Deep Neural Networks KG Sutar, M Prabha, V Rastogi, S Murugan, S VE 2025 5th International Conference on Soft Computing for Security … , 2025 2025.0
Detection and classification of COVID-19 disease using SWHO-based deep neural network classifier V Rastogi, AK Jain Computer Methods in Biomechanics and Biomedical Engineering: Imaging … , 2023 2023.0
Covid-19 Classification Model Based on Age and Gender Analysis Using SWHO-Based Deep CNN V Rastogi, AK Jain 2023 3rd International Conference on Pervasive Computing and Social … , 2023 2023.0
Android Malware Detection N Gagan, SS Kumar, M Keerthana, SS Vaidya, V Rastogi 2023 World Conference on Communication & Computing (WCONF), 1-14 , 2023 2023.0 Citations: 2
Analysis of the Impacts of COVID-19 using Deep Convolutional Neural Network V Rastogi, AK Jain 2023 International Conference on Sustainable Computing and Data … , 2023 2023.0
Video To Braille Transcription For Visually Impaired People VRSS I. Verma, A. Rai, H. Yadav 2021 International Conference on Simulation, Automation & Smart … , 2021 2021.0 Citations: 2
Machine learning algorithms: Overview V Rastogi, S Satija, DPK Sharma, S Singh International Journal of Advanced Research in Engineering and Technology 11 (9) , 2020 2020.0 Citations: 15
Car’s selling price prediction using random forest machine learning algorithm A Pandey, V Rastogi, S Singh 5th international conference on next generation computing technologies (NGCT … , 2020 2020.0 Citations: 30
COVID-19: Nature, Outbreak and Role of Machine Learning AKJ V. Rastogi International Journal of Advanced Research in Engineering and Technology 11 … , 2020 2020.0 Citations: 2
Software development life cycle models-comparison, consequences V Rastogi International Journal of Computer Science and Information Technologies 6 (1 … , 2015 2015.0 Citations: 127
Pathlight AI: An Integrated LLM-Powered Platform for Personalized Career Advancement and Job Market Analysis N Gagan, S Sai Kumar, M Keerthana, SV Samarth, R Anand
AN ENHANCED CLIENT CENTRIC SOFTWARE DEVELOPMENT LIFE CYCLE MODEL WITH COST AND EFFORT ESTIMATION V Rastogi, G Swetha, EA Lakshmi, M Pauline
MOST CITED SCHOLAR PUBLICATIONS
Software development life cycle models-comparison, consequences V Rastogi International Journal of Computer Science and Information Technologies 6 (1 … , 2015 2015.0 Citations: 127
Car’s selling price prediction using random forest machine learning algorithm A Pandey, V Rastogi, S Singh 5th international conference on next generation computing technologies (NGCT … , 2020 2020.0 Citations: 30
Machine learning algorithms: Overview V Rastogi, S Satija, DPK Sharma, S Singh International Journal of Advanced Research in Engineering and Technology 11 (9) , 2020 2020.0 Citations: 15
Android Malware Detection N Gagan, SS Kumar, M Keerthana, SS Vaidya, V Rastogi 2023 World Conference on Communication & Computing (WCONF), 1-14 , 2023 2023.0 Citations: 2
Video To Braille Transcription For Visually Impaired People VRSS I. Verma, A. Rai, H. Yadav 2021 International Conference on Simulation, Automation & Smart … , 2021 2021.0 Citations: 2
COVID-19: Nature, Outbreak and Role of Machine Learning AKJ V. Rastogi International Journal of Advanced Research in Engineering and Technology 11 … , 2020 2020.0 Citations: 2
Skin Diseases Classification with Deep Learning Techniques: A Review RT Reddy, L U, A G, U S, V Rastogi, A R 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025.0
Facial Landmark In Emotion Analysis: A Review BS Vanshika Rastogi, M Shree N, S K, V M, A R 2025 International Conference on Emerging Technologies in Computing and … , 2025 2025.0
Real-Time Driver Intention Prediction for Lane Merging using Deep Neural Networks KG Sutar, M Prabha, V Rastogi, S Murugan, S VE 2025 5th International Conference on Soft Computing for Security … , 2025 2025.0
Detection and classification of COVID-19 disease using SWHO-based deep neural network classifier V Rastogi, AK Jain Computer Methods in Biomechanics and Biomedical Engineering: Imaging … , 2023 2023.0
Covid-19 Classification Model Based on Age and Gender Analysis Using SWHO-Based Deep CNN V Rastogi, AK Jain 2023 3rd International Conference on Pervasive Computing and Social … , 2023 2023.0
Analysis of the Impacts of COVID-19 using Deep Convolutional Neural Network V Rastogi, AK Jain 2023 International Conference on Sustainable Computing and Data … , 2023 2023.0
Pathlight AI: An Integrated LLM-Powered Platform for Personalized Career Advancement and Job Market Analysis N Gagan, S Sai Kumar, M Keerthana, SV Samarth, R Anand
AN ENHANCED CLIENT CENTRIC SOFTWARE DEVELOPMENT LIFE CYCLE MODEL WITH COST AND EFFORT ESTIMATION V Rastogi, G Swetha, EA Lakshmi, M Pauline
