Dr. D. Kalpanadevi is currently working as an Assistant Professor in the faculty of Department of Computer Science, School of Science, Gitam University, Bengaluru Campus, Bengaluru. She has 5+ years of teaching experience and 2 years of research experience. She received her Ph. D Degree (in Regular) from PSG College of Arts and Science affiliated Bharathiar University and received First class with Distinction in MCA. More than published 30 research articles indexed in various more than 20 Scopus and others are Peer reviewed / UGC approved Journals. She attended 19 International conferences and 7 National Conferences. Published three patents and one book. Also attained International Young Researcher Award on 2020- 2021. She has reviewer in IEEE Access and 4 reputed journals. Her area of Interest is Data Mining, Machine Learning, Big Data Analytics, Artificial Intelligence
AI Beyond the Veil: Techniques for Privacy Preservation Responsible AI Principles and Practices, 2026
Machine Learning Involved in Explainable Artificial Intelligence in Cybersecurity and Legal Systems D. Kalpanadevi Attacks and Defenses in Explainable Artificial Intelligence, 2026 Machine learning (ML) has a substantial effect on the realms of cybersecurity as well as the legal field, but it often lacks transparency in its decision-making process. This is where explainable artificial intelligence (XAI) comes in to bridge the gap. Here's how ML interacts with XAI in these two fields: cybersecurity and legal system. ML algorithms excel at identifying patterns in vast amounts of data, making them ideal for tasks. In this research, it focuses on malware detection on classifying files as benign or malicious based on features. By challenging with opacity, ML models are especially used in black boxes. Security analysts often struggle to understand why the system flags something as suspicious. With this lack of explainability, some problems can lead to two ways. One of the ways is “alert fatigue,” which represents a constant barrage of unexplained alerts can overwhelm security teams and lead to ignoring important threats. The second way is “difficulty in trusting the system,” which represents if analysts do not understand the reasoning behind detections, they might hesitate to rely on the system's judgment. By incorporating XAI, legal professionals can ensure that ML-driven decisions are fair, unbiased, and understandable, fostering trust in the legal system. The aim of this paper is to show how ML provides the analytical power in cybersecurity and legal systems, while XAI acts as the interpreter, connecting the intricate algorithms to the human comprehension. This combination is crucial for building trust and ensuring responsible decision-making in these critical fields.
Amazon Alexa Sentiment Analysis using Optimized CNN-BiLSTM Deep Learning Model with Advanced Word Embeddings Kartheeban K, Sri Ram Ganesh M, Nirmal Dhanabalan, D. Kalpanadevi, C Priya Proceedings of 3rd International Conference on Intelligent Cyber Physical Systems and Internet of Things Icoici 2025, 2025 Sentiment analysis is an essential task in natural language Processing, and allows systems to parse subjective text and classify it into categories. This paper introduces a hybrid deep learning method of sentiment classification of customer reviews of Amazon Alexa that leverages Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (Bi-LSTM) networks. CNNs are used to find local dependencies and focus on individual phrases, while the Bi-LSTM layers are used to model long-range context. The model also incorporates two state-of-the-art word-embedding methods (GloVe and FastText) that attempt to convey some semantic meaning to the words. The word embeddings allow each word to be learned using dense embeddings, which have less overhead generally, while also embedding vastly greater amounts of context. The following combinations of the embeddings and neural networks were covered in this study: CNN with GloVe embedding, CNN with FastText embedding, Bi-LSTM with GloVe embedding, Bi-LSTM with FastText embedding; the use of CNN was investigated first. The results indicated that CNN with GloVe embedding provided the highest performance with a precision of 0.96, a recall of 0.98, an F1-score of 0.97, and an overall accuracy of 94% on the Amazon Alexa reviews dataset. Overall, this paper suggests that a combination of state-of-the-art word embeddings and hybrid neural networks is quite effective for dealing with short, user-generated content. Overall, the model and methodology has real-world practical applications for voice feedback systems with intelligent agents, tracking customer sentiment by product, and jazzing-up the e-commerce experience.
Deep Learning-based Binary Classification of Bone Fractures using a Hybrid MobileNetV3-CNN Architecture and Clinical X-ray Dataset Jona. J. B, Kartheeban K, D. Kalpanadevi, Jeevitha S, Sri Ram Ganesh M, Somasundaram R S Proceedings of 6th International Conference on Iot Based Control Networks and Intelligent Systems Icicnis 2025, 2025 Identifying the presence of bone fractures from radiographic images is a clinically important and time-consuming task that is limited by inter-observer variability or availability of radiologists. This research proposes an end-to-end deep learning framework for the automated classification of fractures using a hybrid architecture of MobileNetV3-Small and specifically designed new convolutional layers. The framework was trained and validated on the Fracture Multi-Region X-ray dataset, a publicly available dataset containing 10,580 grayscale X-ray images labelled from different anatomically connected regions (bones of limbs, spine, hips, knees). To improve diagnostic accuracy and freedoms, we implemented transfer learning, data preprocessing, a comprehensive and internal augmentation pipeline (including flipping, rotation, contrast, and zoom), etc. MobileNetV3-Small was used as a lightweight feature extractor and further optimized with our convolutional blocks, batch normalization, dropout, and global average pooling. Our training strategy included two separate phases, where the first phase aimed to solely extract the features, with the second phase featuring feature extraction and then fine-tuning, while optimization was executed using AdamW with a fully adaptive learning schedule. Evaluation performance on the test set resulted in isolated classification accuracy of 97%, and was candidates set against DenseNet121, VGG-16, and ResNet50. The findings validate the effectiveness of using lightweight architectures enhanced with domain-specific features for precise medical image analysis. Due to its computational efficiency, the model is suitable for real-time and resource-constrained applications, such as point-of-care and mobile health systems. This work establishes a foundation for future improvement through transformer-based architectures, multimodal data types, and explainable AI to facilitate diagnostic confidence and opportunities for clinical decision-making.
Hybrid Machine Learning Integration with Task Scheduling Algorithms in Cloud Environments D. Kalpanadevi, Eswararao Boddepalli, Igor' Geleta, More Swami Das, Anil Kumar Lamba, M S. Mohamed Mallick 2025 Global Conference in Emerging Technology Ginotech 2025, 2025 Effective task scheduling and resource allocation are significant difficulties in cloud computing, frequently resulting in delayed execution times, elevated energy usage, and cost inefficiencies. Conventional scheduling techniques, like Round Robin and Genetic Algorithms, fail to dynamically adjust to variations in workload, leading to resource inefficiency and extended processing delays. This study introduces RATS-HM (Resource Allocation and job Scheduling using Hybrid Machine Learning), an innovative framework that combines Improved Cat Swarm Optimisation (CSO) with Group Optimization-based Deep Neural Networks (DNNs) for enhanced job scheduling and resource allocation. The suggested approach guarantees lowered execution time (2.5s), optimised makespan (120s), enhanced energy efficiency (85%), and increased cost savings (30%) relative to traditional models. The system constantly adjusts to changes in workload, guaranteeing efficient task execution while improving cloud infrastructure performance. Experimental findings demonstrate that RATS-HM surpasses current methodologies regarding execution speed, scalability, and efficiency, establishing it as a competitive alternative for future cloud computing environments.
High Performance Multimodal MRI and CT Scan-based Deep Learning Model for Brain Tumor Identification Jona. J. B, D. Kalpanadevi, Kartheeban K, Jeevitha S, Sri Ram Ganesh M, Somasundaram R S Proceedings of 5th International Conference on Ubiquitous Computing and Intelligent Information Systems Icuis 2025, 2025 Accurate and fast diagnosis of tumors in the brain continues to be difficult in medical images, considering the difficult anatomies and classic diagnosis shortcomings. We, in this paper, propose a robust multimodal deep learning framework, based on trusted data (both MRI and CT), and, moreover, automatic classification of brain tumors. We managed to create this framework, taking advantage of an openly available data set of 9,618 labeled MRI and CT images and brain tumors. The four classes of consideration were: Healthy (MRI/CT) and Tumor (MRI/CT). Constructed upon the respective strengths of MRIs and of CTs, enabling trusted diagnosis, a hybrid structure was developed, founded on the basis of MobileNet V3, and enriched by the inclusion of additional custom convolution layers so as so extract high-level features and learn spatial information. Various state-of-art training schemes were incorporated, including real-time data augmentation and use of the AdamW optimizer, aiding generalizability of the model and overfitting prevention. Performance of each method was evaluated on the basis of baseline architectures: ResNet50, DenseNet121, CNN-BiLSTM, and we deduced that optimized MobileNetV3+CNN is superior over all tested approaches. It achieves the highest and best level of accuracy of 99%. Moreover, the model incorporates a light-howitz structure, enabling it, thus, to be efficient and deployable as an edge model in real-time diagnosis approaches. Our work here has verified the hypothesis that, provided we use light but informative neural architectures, and optimized training approaches, we will extract superior levels of classification accuracy and strongly scalable methods in work of analysis of medical images. Future work will involve cross domain validation and explainable AI modules so aid clinical decision support and trust. The research shows the possibility of deep learning application in neuro-oncology or any medical imaging facing similar challenges with the given research question, and the possibility of similar applications in industrial images.
An Adaptive AI Model for Intelligent Fraud Detection and Customer Engagement in Digital Banking Rohini Chittakula, D. Kalpanadevi, Jayalakshmi V, RVS Praveen, S. Pragadeeswaran, M. Amsa 2025 IEEE 4th International Conference for Advancement in Technology Iconat 2025, 2025 Digital banking has transformed financial accessibility while concurrently heightening clients' exposure to fraud threats and reducing personal user experiences. Conventional fraud detection techniques frequently struggle to adjust to changing attack vectors, while consumer interaction strategies lack in personalisation and real-time reactivity. This research introduces an adaptable AI system that integrates a Long Short-Term Memory (LSTM) network with Gradient Boosted Decision Trees (GBDT) for fraud detection, with a reinforcement learning module for enhanced customer engagement. This dual-module system utilises real-world transaction datasets from Kaggle, employing advanced feature engineering, concept drift adaptability, and sentiment-aware personalisation. The suggested approach markedly outperformed baseline models, including Logistic Regression, Random Forest, and XGBoost, in fraud detection, attaining a precision of 0.98, recall of 0.96, F1-score of 0.97, and AUC-ROC of 0.99. These measures confirm the strength and dependability of the system in complex financial contexts. The engagement module enhanced response and conversion rates by dynamically customising interactions according to user behaviour and sentiment analysis. This integrated and flexible model strengthens security and improves the digital banking experience compared to previous alternatives. The findings indicate significant improvements over traditional methods, offering a thorough answer to current digital banking issues.
Earthquake Early Warning System Utilizing an CNN-LSTM-TL Based Method for Detection and Parameters Classification D. Kalpanadevi, M. Siva, Chetan Shashikant Chavan, S. Kaliappan, S. Jothilakshmi, K Venkata Ramana 1st International Conference on Electronics Computing Communication and Control Technology Iceccc 2024, 2024 The obvious first line of protection against powerful earthquake motion is to reinforce houses and other structures. The goal of real-time earthquake catastrophe prevention, in contrast to real-time seismology, is to mitigate damage while an earthquake is still underway. A disaster preventive measure that can be put into action in real-time in the event of an earthquake requires an early warning system (EEW). In order to avoid disasters, should not rely solely on EEW. The order of preprocessing, feature selection, and training the model must be meticulously followed. The preparation phase includes data encoding and normalization. Feature selection incorporates principal component analysis and linear discriminant analysis. It utilized CNN-LSTM-TL for the model's training. The results demonstrate a remarkable 96.49% accuracy.
Temperature Variation Modelling in Mushroom Growing Hall with IAN-Bidirectional GRU Model D. Kalpanadevi, S. Nandhini Devi, Silas Stephen D, Sunita Jadhav, P M D Ali Khan, Muralidharan J 4th International Conference on Sustainable Expert Systems Icses 2024 Proceedings, 2024 Modern developments in computing and electronics have made intelligent systems suitable for use in the mechanization of agricultural processes. This proposed used multilayered perceptron and radial basis function networks to forecast the temperature shifts in the mushroom cultivation chamber based on independent variables including water temperature, ambient temperature, water tap, fresh air and circulation air dampers, and water tap. Preprocessing, feature extraction, and training the model are the three phases that comprise this proposed method. Reflectance calibration, normalization, and spectral smoothing are the steps involved in data preprocessing. Feature extraction made quick work of extracting information from mushroom photos by employing the SIFT algorithm. The proposed model was trained using an IAN-Bidirectional GRU. When compared to more conventional approaches, the proposed methodology is superior. The method's application resulted in a 93.75% improvement in accuracy.
An Optimized PV-Based Multi-Port Plug-In EV Charger Using Improved Particle Swarm Optimization Algorithm Naresh Kumar M., Venkatesh Kumar C., D. Kamalakkannan, Kaushalya Thopate, Sathish Kumar Shanmugam, T. Vignesh, Anand Goswami, D. Kalpanadevi Electric Power Components and Systems, 2024 In order to significantly improve the factor power and State of Charge, we proposed a design of Particle Swarm Optimization (PSO) with Maximum Power Point Tracking (MPPT) and control the Pulse Width Modulation (PWM) by PI controller that improves performance and reduces losses than other two existing works such as PV grid system designed with PI controller and P&O with MPPT and Fuzzy control-based PV grid system for EV charging. Here, the proposed PSO technique is improved by adding frequency scaling. The photovoltaic (PV) controlling, maximum power tracking and pulse generation are improving the result by adopting Improved Particle Swarm Optimization (IPSO). Multi-port charging station utilizes battery. After, getting the voltage link DC, the grid utilization is fed with the inverter logic. Here, the multi-port charger with PV and convertor boost DC-DC is directly interfaced with the electric vehicles (EV) charging port; additionally, the battery stores the energy. Depends on the utility grid, the energy is utilized. Thus, the proposed method attains the result of efficient multi-port utility, improved percentage of State of Charge (SoC), and better switching circuit with PFC and larger load utility. Overall, the proposed work is done in the 2018a version that adaptation with MATLAB/Simulink.
Deep Learning-based Binary Classification of Bone Fractures using a Hybrid MobileNetV3-CNN Architecture and Clinical X-ray Dataset K Kartheeban, D Kalpanadevi, S Jeevitha, RS Somasundaram 2025 6th International Conference on IoT Based Control Networks and … , 2025 2025 Citations: 1
High Performance Multimodal MRI and CT Scan-based Deep Learning Model for Brain Tumor Identification D Kalpanadevi, K Kartheeban, S Jeevitha, RS Somasundaram 2025 5th International Conference on Ubiquitous Computing and Intelligent … , 2025 2025
Amazon Alexa Sentiment Analysis using Optimized CNN-BiLSTM Deep Learning Model with Advanced Word Embeddings K Kartheeban, N Dhanabalan, D Kalpanadevi, C Priya 2025 3rd International Conference on Intelligent Cyber Physical Systems and … , 2025 2025 Citations: 1
Earthquake Early Warning System Utilizing an CNN-LSTM-TL Based Method for Detection and Parameters Classification D Kalpanadevi, M Siva, CS Chavan, S Kaliappan, S Jothilakshmi, ... 2024 International Conference on Electronics, Computing, Communication and … , 2024 2024 Citations: 4
Detection of Disease in Apple Plant using its Images of Leaf Through KNN and Support Vector Machine A Kaur, D Kalpanadevi, GS Gayathri, K Shantanu, R Rastogi, ... 2023 2nd International Conference on Automation, Computing and Renewable … , 2023 2023 Citations: 1
Optimal Allocation of Resources in Data Center using Artificial Intelligence D Kalpanadevi, P Babysudha, K Kartheeban, M Mayilvaganan Recent Trends in Computational Intelligence and Its Application, 41-50 , 2023 2023 Citations: 1
Optimal allocation of Resources in Data Center using Artificial Intelligence DMM Dr. D. Kalpanadevi, P.Babysudha, Dr. K. Kartheeban Intelligent Systems, Data Engineering and Optimization , 2023 2023
“Image Denoising for Smart Laser Osteotomy Using Deep Learning-based Fast Optical Coherence Tomography (OCT) Shubhangi N. Ghate, D. Kalpanadevi, K Amudha, Priya Velayutham, Balu S, Mohd ... 2023 Second International Conference on Electronics and Renewable Systems … , 2023 2023 Citations: 1
An Effective Evaluation of SONARS using Arduino and Display on Processing IDE RG Vidhya, BK Rani, K Singh, D Kalpanadevi, JP Patra, TAS Srinivas 2022 International Conference on Computer, Power and Communications , 2023 2023 Citations: 31
Computer Vision- Hybrid Learning Based On Multi Scale Dilated Convolution Module Mechanism Implemented For Object Detection MPB Dr. D. Kalpanadevi, Dr. K. Kartheeban, Dr. M. Mayilvaganan International Conference on Automation, Computing and Renewable Systems , 2023 2023
A Novel Machine Learning Algorithm for ProstateCancer Image Segmentation using mpMR KKK Tushar Dhar Shukla, Kalpana. K, Richa Gupta, Kalpanadevi D, Md. Abul Ala ... International Conference on Sustainable Computing and Smart Systems, IEEE Xplore , 2023 2023
Computer Vision-Hybrid Learning based on Multi Scale Dilated Convolution Module Mechanism Implemented for Object Detection D Kalpanadevi, K Kartheeban, M Mayilvaganan, P Bamaruckmani 2022 International Conference on Automation, Computing and Renewable Systems … , 2022 2022
“Diagnosis Kidney Function Test Using Machine Learning Algorithm Based on Runge Kutta Method DDK Dr. K. Kartheeban Mathematical Statistician and Engineering Applications 71 , 2022 2022
Enhancement of RKBlowfish Algorithm for Data Encryption through Block Chain in Healthcare System DMK Dr. D. Kalpanadevi, Dr. M. Jansi Rani Mathematical Statistician and Engineering Applications 71 , 2022 2022 Citations: 2
Building an optimal model of cognitive using klm and complexity theory in human computer interface D Kalpanadevi 2021 5th International Conference on Electronics, Communication and … , 2021 2021 Citations: 3
Robustness of Adaptive Neuro- Fuzzy Inference System for Optimal Prediction using Roulette Wheel Method DD Kalpanadevi International Journal of Recent Technology and Engineering 8 (5 … , 2020 2020
Design and Implementation of Human-computer interface based Cognitive Model for Examine the Skill Factor of Students D Kalpanadevi 2019 3rd International Conference on Computing Methodologies and … , 2019 2019 Citations: 1
Frequent Pattern Mining of Crop Cultivation in different location at Virunagar District Based on Association Rule Mining PB Dr. D.Kalpanadevi, S. Ponmalar First National Conference on Smart Innovative Technologies on Data Analytics , 2019 2019
Analysis of Smart Bank Coaching Mobile Application Based on System Usability Scale MS [4] Dr. D.Kalpanadevi International Journal of Research in Advent Technology 7 (5S), 233-236 , 2019 2019
Efficient Analysis of Measuring Usability Metric based on Cognitive Model through Examine the Students Task Performance DD Kalpanadevi International Journal of Research in Advent Technology 7 (5S), 155-159 , 2019 2019
MOST CITED SCHOLAR PUBLICATIONS
Comparison of classification techniques for predicting the performance of students academic environment M Mayilvaganan, D Kalpanadevi 2014 International Conference on Communication and Network Technologies, 113-118 , 2014 2014 Citations: 180
An Effective Evaluation of SONARS using Arduino and Display on Processing IDE RG Vidhya, BK Rani, K Singh, D Kalpanadevi, JP Patra, TAS Srinivas 2022 International Conference on Computer, Power and Communications , 2023 2023 Citations: 31
Comparison of classification techniques for predicting the cognitive skill of students in education environment M Mayilvaganan, D Kalpanadevi 2014 IEEE International Conference on Computational Intelligence and … , 2014 2014 Citations: 23
Cognitive Skill Analysis for Students through Problem Solving Based on Data Mining Techniques DK M. Mayilvaganan Procedia Computer Science -Science Direct 47 (2015), 62-75 , 2015 2015 Citations: 21
Effective searching shortest path in graph using Prim’s Algorithm D Kalpanadevi Int J Comput Organ Trends 3, 310-3 , 2013 2013 Citations: 11
Recent trends in human computer interface to analysis the cognitive skill of students based on user interface M Mayilvaganan, D Kalpanadevi 2017 4th International Conference on Advanced Computing and Communication … , 2017 2017 Citations: 6
Designing a human computer interface system based on cognitive model M Mayilvaganan, D Kalpanadevi 2014 IEEE International Conference on Computational Intelligence and … , 2014 2014 Citations: 5
Earthquake Early Warning System Utilizing an CNN-LSTM-TL Based Method for Detection and Parameters Classification D Kalpanadevi, M Siva, CS Chavan, S Kaliappan, S Jothilakshmi, ... 2024 International Conference on Electronics, Computing, Communication and … , 2024 2024 Citations: 4
Building an optimal model of cognitive using klm and complexity theory in human computer interface D Kalpanadevi 2021 5th International Conference on Electronics, Communication and … , 2021 2021 Citations: 3
Enhancement of RKBlowfish Algorithm for Data Encryption through Block Chain in Healthcare System DMK Dr. D. Kalpanadevi, Dr. M. Jansi Rani Mathematical Statistician and Engineering Applications 71 , 2022 2022 Citations: 2
Designing a Human Computer Interface System Based on Cognitive Model for Examining the Experts M Mayilvaganan, D Kalpanadevi 9th International Conference on Science, Engineering and Technology (SET … , 2015 2015 Citations: 2
Deep Learning-based Binary Classification of Bone Fractures using a Hybrid MobileNetV3-CNN Architecture and Clinical X-ray Dataset K Kartheeban, D Kalpanadevi, S Jeevitha, RS Somasundaram 2025 6th International Conference on IoT Based Control Networks and … , 2025 2025 Citations: 1
Amazon Alexa Sentiment Analysis using Optimized CNN-BiLSTM Deep Learning Model with Advanced Word Embeddings K Kartheeban, N Dhanabalan, D Kalpanadevi, C Priya 2025 3rd International Conference on Intelligent Cyber Physical Systems and … , 2025 2025 Citations: 1
Detection of Disease in Apple Plant using its Images of Leaf Through KNN and Support Vector Machine A Kaur, D Kalpanadevi, GS Gayathri, K Shantanu, R Rastogi, ... 2023 2nd International Conference on Automation, Computing and Renewable … , 2023 2023 Citations: 1
Optimal Allocation of Resources in Data Center using Artificial Intelligence D Kalpanadevi, P Babysudha, K Kartheeban, M Mayilvaganan Recent Trends in Computational Intelligence and Its Application, 41-50 , 2023 2023 Citations: 1
“Image Denoising for Smart Laser Osteotomy Using Deep Learning-based Fast Optical Coherence Tomography (OCT) Shubhangi N. Ghate, D. Kalpanadevi, K Amudha, Priya Velayutham, Balu S, Mohd ... 2023 Second International Conference on Electronics and Renewable Systems … , 2023 2023 Citations: 1
Design and Implementation of Human-computer interface based Cognitive Model for Examine the Skill Factor of Students D Kalpanadevi 2019 3rd International Conference on Computing Methodologies and … , 2019 2019 Citations: 1
COMPARISON OF APRIORI, FP-TREE GROWTH AND FUZZY FP-TREE GROWTH ALGORITHM FOR GENERATING ASSOCATION RULE MINING OF COGNITIVE SKILL M Mayilvaganan, D Kalpanadevi International Journal of Engineering Research and General Science 6 (2), 48-63 , 2018 2018 Citations: 1
Computational Results of Hybrid Learning in Adaptive Neuro Fuzzy Inference System for Optimal Prediction D Kalpanadevi, M Mayilvaganan International Journal of Applied Engineering Research 12 (16), 5810-5818 , 2017 2017 Citations: 1
High Performance Multimodal MRI and CT Scan-based Deep Learning Model for Brain Tumor Identification D Kalpanadevi, K Kartheeban, S Jeevitha, RS Somasundaram 2025 5th International Conference on Ubiquitous Computing and Intelligent … , 2025 2025
GRANT DETAILS
Proposal submitted for CSIR-ASPIRE: A Special Call for Research Grants for Women Scientists in the proposed title “Robustness of Adaptive Neuro-Fuzzy Inference
System for Optimal Prediction of Student Ability Level with Brain Dominant Hemisphere Using Human Computer Interface.
RESEARCH OUTPUTS (PATENTS, SOFTWARE, PUBLICATIONS, PRODUCTS)
Application number: 202241034677 A
Title: Web Based Integrating Multimedia Tool Lecture System Design for Teaching
and Learning Language.
Published Date: 24/06/2022
• Application number : 202241045393 A
Title: Designing an Artificial Intelligence module for analyzing Six Sigma based
Approaches to Business Practices in Manufacturing, Services, and Production
Publication Date: 19/08/2022
• Application number : 202241073283 A
Title: Design of an Automated Traffic Management System with Sensors to Predict
the Occurrence of Disasters and Alert the Drivers on Highway
Publication Date: 13/01/2023
