Raghi K R
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Raghi K R has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Arjun Paramarthalingam
University College of Engineering
2 shared publications - Abhishek Sharma
Department of Computer Science and Engineering, Graphic Era Deemed to be University, Dehradun, India
1 shared publication - Anto Melvin Mathew
The University of Western Australia
1 shared publication - Baba Fakruddin Ali B. H
Computer Science and Engineering, Nagarjuna College of Engineering and Technology
1 shared publication - Dr Balaprasad Purushottam Kurpatwar 5
Department of Mechanical Engineering Adsul Technical campus chas Ahilyanagar - 414005
1 shared publication - Harini Kanthan
Computer Science and Engineering, University College of Engineering Villupuram, Villupuram 605103, TN, India
1 shared publication
Latest Publications
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1895-1899
Generative AI-Assisted Farm Planning and Agricultural Resource Optimization
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1895-1899
Generative AI-Assisted Farm Planning and Agricultural Resource Optimization
- 17 February 2026
- Published by Springer Nature
- p. 57-68
A Generative Method for Steganography by Cover Synthesis with Secure System
- 17 February 2026
- Published by Springer Nature
- p. 57-68
A Generative Method for Steganography by Cover Synthesis with Secure System
- Open Access
A Privacy-Preserving Approach to Health Insurance Fraud Detection Using Vertical Federated Learning
In fraud detection, centralized approaches often face challenges related to data protection, security, and potential data breaches. Such methods require sensitive healthcare and insurance data to be pooled in one location, which increases vulnerability to misuse. This paper introduces FraudNetX, a privacy-preserving fraud detection framework, by utilizing Vertical Federated Learning (VFL) to address centralized system limitations. VFL enables models to be trained collaboratively while ensuring data privacy and security through quantifiable Differential Privacy (DP) guarantees (ε = 1.0, δ = 1 × 105). FraudNetX implements a noise injection based on Differential Privacy (DP) with Gaussian noise (s = 1.2) in the process of training the model to guarantee confidentiality of the personal data. This research entails two partner organizations, which are a hospital and an insurance company, in an actual VFL configuration. The model is trained on 10 communication rounds in this federated setup, and the local optimization of each client is followed by the global aggregation step. Hospitals and insurers can learn fraud patterns without sharing their data. The proposed FraudNetX is a hybrid architecture which is composed of Feedforward Neural Networks (FFNNs) and transformer encoders. An adaptive weighting strategy has been applied to handle category imbalance concern and enhance recall of a few categories which are hard to detect, especially in fraud involving minorities, balancing the recall performance. The framework also includes a decision model that uses hospital data and claim behavior to classify each claim as legitimate, under review, or fraudulent. The experimental evaluation on the real-world dataset demonstrates that FraudNetX enhances the accuracy and F1-score of fraud detection (accuracy = 99.91%, F1 = 99.94%, ROC-AUC = 0.98) but does not violate data privacy.
Open AccessA Privacy-Preserving Approach to Health Insurance Fraud Detection Using Vertical Federated Learning
In fraud detection, centralized approaches often face challenges related to data protection, security, and potential data breaches. Such methods require sensitive healthcare and insurance data to be pooled in one location, which increases vulnerability to misuse. This paper introduces FraudNetX, a privacy-preserving fraud detection framework, by utilizing Vertical Federated Learning (VFL) to address centralized system limitations. VFL enables models to be trained collaboratively while ensuring data privacy and security through quantifiable Differential Privacy (DP) guarantees (ε = 1.0, δ = 1 × 105). FraudNetX implements a noise injection based on Differential Privacy (DP) with Gaussian noise (s = 1.2) in the process of training the model to guarantee confidentiality of the personal data. This research entails two partner organizations, which are a hospital and an insurance company, in an actual VFL configuration. The model is trained on 10 communication rounds in this federated setup, and the local optimization of each client is followed by the global aggregation step. Hospitals and insurers can learn fraud patterns without sharing their data. The proposed FraudNetX is a hybrid architecture which is composed of Feedforward Neural Networks (FFNNs) and transformer encoders. An adaptive weighting strategy has been applied to handle category imbalance concern and enhance recall of a few categories which are hard to detect, especially in fraud involving minorities, balancing the recall performance. The framework also includes a decision model that uses hospital data and claim behavior to classify each claim as legitimate, under review, or fraudulent. The experimental evaluation on the real-world dataset demonstrates that FraudNetX enhances the accuracy and F1-score of fraud detection (accuracy = 99.91%, F1 = 99.94%, ROC-AUC = 0.98) but does not violate data privacy.
- 1 September 2025
- Published by Springer Nature
- p. 307-317
Transforming Lives: AIoT-Driven Deep Learning for Real-Time Speech Assistance for the Ophthalmologically Impaired
- 1 September 2025
- Published by Springer Nature
- p. 307-317
Transforming Lives: AIoT-Driven Deep Learning for Real-Time Speech Assistance for the Ophthalmologically Impaired
- Open Access
- 10 May 2025
- Published by Academic Science Publications and Distributions in International Journal of Environmental Sciences
- Vol. 11 (4s) , 1246-1253
Deep Ensemble Learning For Accurate Prediction Of Neurodegenerative Disorders Using Temporal Clinical Data
— as stated the proposed research seeks to work on designing a resilient prediction model on neurodegenerative disorders through deep ensemble learning. alzheimer’s and parkinson’s diseases and other related diseases are disorders that involve the gradual loss of neurons. early identification of disorder is highly essential since proper care and treatment can only be given if disorder is predicted accurately. regarding challenge we suggest the deep ensemble learning approach incorporating several deep learning models, namely recurrent neural networks (rnns), long short-term memory (lstm) networks to analyze temporal structure of the patient records and clinical data. the lstm model is specifically effective in registering long term temporal dependencies while rnns are especially useful in modelling temporal sequentiality. these models are integrated using ensemble learning with a weighted voting system to boost up the model’s predictive prowess and to minimize the problem of overfitting the data. because of this, the ability of the ensemble to harness strengths from each of the individual models makes the whole system very resilient, even if patient data is noisy or otherwise partially complete. substantial testing and verification on clinical datasets show that the introduced approach outperforms existing ones in identifying the risk factors of neurological disorders and their evolution, which can help improve the accuracy of clinical decision-making and improve patient care.
- 10 May 2025
- Published by Academic Science Publications and Distributions in International Journal of Environmental Sciences
- Vol. 11 (4s) , 1246-1253
Open AccessDeep Ensemble Learning For Accurate Prediction Of Neurodegenerative Disorders Using Temporal Clinical Data
— as stated the proposed research seeks to work on designing a resilient prediction model on neurodegenerative disorders through deep ensemble learning. alzheimer’s and parkinson’s diseases and other related diseases are disorders that involve the gradual loss of neurons. early identification of disorder is highly essential since proper care and treatment can only be given if disorder is predicted accurately. regarding challenge we suggest the deep ensemble learning approach incorporating several deep learning models, namely recurrent neural networks (rnns), long short-term memory (lstm) networks to analyze temporal structure of the patient records and clinical data. the lstm model is specifically effective in registering long term temporal dependencies while rnns are especially useful in modelling temporal sequentiality. these models are integrated using ensemble learning with a weighted voting system to boost up the model’s predictive prowess and to minimize the problem of overfitting the data. because of this, the ability of the ensemble to harness strengths from each of the individual models makes the whole system very resilient, even if patient data is noisy or otherwise partially complete. substantial testing and verification on clinical datasets show that the introduced approach outperforms existing ones in identifying the risk factors of neurological disorders and their evolution, which can help improve the accuracy of clinical decision-making and improve patient care.