Prasath Nithiyanandam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Prasath Nithiyanandam has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Balasubramanian Muthu Kumar
School of Computing and Information Technology, REVA University, Bengaluru 562157, India
2 shared publications J. Sreemathy
Sri Eshwar College of Engineering, India
2 shared publications- Pandi Vigneshwaran
Acharya University
2 shared publications - Sindhuja Manickavasagam
Rajalakshmi Engineering College
2 shared publications A. Arun
SRM Institute of Science and Technology, India
1 shared publicationAbirami K. Rama
Cybersecurity, School of Engineering and Technology, PSB Academy, Singapore
1 shared publication
Latest Publications
- 19 June 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Information Technology & Decision Making
- p. 1-32
Autonomous Cyber-Defense for IoT-Edge Networks Using a Starfish Algorithm Optimized Bayesian Constitutive Neural Networks with Cross-Domain Zero Trust
This paper presents an advanced autonomous cyber-defense framework for IoT-edge networks, leveraging a novel integration of Starfish Algorithm-Optimized Bayesian Constitutive Neural Networks (BCNN) within Cross-Domain Zero Trust architecture. The system begins with input data sourced from two comprehensive and diverse datasets, IoT-24 and TON_IoT-v2, which reflect a wide range of IoT traffic patterns and attack scenarios. To enhance data quality, a Reverse Lognormal Kalman Filter (RLKF) is employed for pre-processing, effectively normalizing and denoising the input data. The cleaned data is then analyzed by BCNNs, which provide probabilistic modeling to account for uncertainties in cyber threat detection. To overcome the inherent limitations of BCNNs in parameter tuning, the Starfish Optimization Algorithm (SFOA) is used to optimize the model, significantly improving detection accuracy while minimizing computational demands—an essential feature for deployment in resource-constrained edge environments. Security is further reinforced through a Cross-Domain Zero Trust model, enabling continuous authentication and verification of data flows to prevent unauthorized access. The proposed technique attains 8.26%, 3.06%, and 3.00% higher accuracy, 3.48%, 3.13%, and 3.39% higher precision over state-of-the-art methods including QIFL, ZTA-IoT, and XAI-based detection systems. The proposed approach thus delivers a scalable, intelligent, and resilient solution for securing heterogeneous IoT-edge ecosystems.
- 19 June 2026
- Published by World Scientific Pub Co Pte Ltd in International Journal of Information Technology & Decision Making
- p. 1-32
Autonomous Cyber-Defense for IoT-Edge Networks Using a Starfish Algorithm Optimized Bayesian Constitutive Neural Networks with Cross-Domain Zero Trust
This paper presents an advanced autonomous cyber-defense framework for IoT-edge networks, leveraging a novel integration of Starfish Algorithm-Optimized Bayesian Constitutive Neural Networks (BCNN) within Cross-Domain Zero Trust architecture. The system begins with input data sourced from two comprehensive and diverse datasets, IoT-24 and TON_IoT-v2, which reflect a wide range of IoT traffic patterns and attack scenarios. To enhance data quality, a Reverse Lognormal Kalman Filter (RLKF) is employed for pre-processing, effectively normalizing and denoising the input data. The cleaned data is then analyzed by BCNNs, which provide probabilistic modeling to account for uncertainties in cyber threat detection. To overcome the inherent limitations of BCNNs in parameter tuning, the Starfish Optimization Algorithm (SFOA) is used to optimize the model, significantly improving detection accuracy while minimizing computational demands—an essential feature for deployment in resource-constrained edge environments. Security is further reinforced through a Cross-Domain Zero Trust model, enabling continuous authentication and verification of data flows to prevent unauthorized access. The proposed technique attains 8.26%, 3.06%, and 3.00% higher accuracy, 3.48%, 3.13%, and 3.39% higher precision over state-of-the-art methods including QIFL, ZTA-IoT, and XAI-based detection systems. The proposed approach thus delivers a scalable, intelligent, and resilient solution for securing heterogeneous IoT-edge ecosystems.
- 28 November 2025
- Published by IGI Global Scientific Publishing
- p. 333-372
Smart IoT Health Monitor for Early TB Diagnosis Using Optimized Hyperbolic NasNet Attention Sea-Horse
Tuberculosis (TB) remains a major global health challenge, particularly in low-resource settings where traditional diagnostic methods are slow, often leading to disease progression and increased transmission. To overcome these limitations, this study introduces an IoT-based patient monitoring system for early TB diagnosis, employing an Optimized Hyperbolic NasNet Attention Sea-horse Graph Network (OpHyp-NASH-Gra-Net). The system uses a wearable IoT sensor to collect patient data, which is transmitted via mobile device to cloud storage. Data is then clustered using an Adaptively Regularized Kernel-Based Fuzzy C-Means algorithm, filtered via a Fast Guided Median Filter (FGMF) for enhanced quality, and processed with a Scale-Aware Modulation Meet Transformer (SA2MT) for feature extraction. The OpHyp-NASH-Gra-Net model classifies the data to distinguish between TB and non-TB cases, achieving 99% accuracy in simulations conducted with Python. This approach enables real-time, high-accuracy diagnosis, reducing delays and improving outcomes in resource-constrained environments.
- 28 November 2025
- Published by IGI Global Scientific Publishing
- p. 333-372
Smart IoT Health Monitor for Early TB Diagnosis Using Optimized Hyperbolic NasNet Attention Sea-Horse
Tuberculosis (TB) remains a major global health challenge, particularly in low-resource settings where traditional diagnostic methods are slow, often leading to disease progression and increased transmission. To overcome these limitations, this study introduces an IoT-based patient monitoring system for early TB diagnosis, employing an Optimized Hyperbolic NasNet Attention Sea-horse Graph Network (OpHyp-NASH-Gra-Net). The system uses a wearable IoT sensor to collect patient data, which is transmitted via mobile device to cloud storage. Data is then clustered using an Adaptively Regularized Kernel-Based Fuzzy C-Means algorithm, filtered via a Fast Guided Median Filter (FGMF) for enhanced quality, and processed with a Scale-Aware Modulation Meet Transformer (SA2MT) for feature extraction. The OpHyp-NASH-Gra-Net model classifies the data to distinguish between TB and non-TB cases, achieving 99% accuracy in simulations conducted with Python. This approach enables real-time, high-accuracy diagnosis, reducing delays and improving outcomes in resource-constrained environments.
- 9 October 2023
- Published by IGI Global Scientific Publishing
- p. 88-123
Ethical Considerations for Wearable Sensors and AI Algorithms
Fetal heart rate monitoring is a crucial aspect of prenatal care that has limitations with traditional methods. Recent advancements in wearable sensors and AI algorithms have introduced new approaches that address some of these limitations, including accuracy, comfort, and mobility. However, the use of these technologies raises ethical concerns related to privacy, informed consent, accuracy, and reliability of AI algorithms, potential for misdiagnosis or overdiagnosis, and equity in access to technology and healthcare. While recent studies demonstrate the potential benefits of wearable sensors and AI algorithms for fetal heart rate monitoring, they also highlight the need for continued dialogue and collaboration between healthcare providers, technology developers, and policymakers to ensure the ethical use of these technologies. This chapter will examine the ethical considerations related to the use of wearable sensors and AI algorithms in fetal heart rate monitoring, emphasizing the need for responsible and ethical use of these technologies.
- 9 October 2023
- Published by IGI Global Scientific Publishing
- p. 88-123
Ethical Considerations for Wearable Sensors and AI Algorithms
Fetal heart rate monitoring is a crucial aspect of prenatal care that has limitations with traditional methods. Recent advancements in wearable sensors and AI algorithms have introduced new approaches that address some of these limitations, including accuracy, comfort, and mobility. However, the use of these technologies raises ethical concerns related to privacy, informed consent, accuracy, and reliability of AI algorithms, potential for misdiagnosis or overdiagnosis, and equity in access to technology and healthcare. While recent studies demonstrate the potential benefits of wearable sensors and AI algorithms for fetal heart rate monitoring, they also highlight the need for continued dialogue and collaboration between healthcare providers, technology developers, and policymakers to ensure the ethical use of these technologies. This chapter will examine the ethical considerations related to the use of wearable sensors and AI algorithms in fetal heart rate monitoring, emphasizing the need for responsible and ethical use of these technologies.
- 2 March 2023
- Published by Springer Nature
- p. 239-249
Analysis of Crop Yield Prediction Using Random Forest Regression Model
- 2 March 2023
- Published by Springer Nature
- p. 239-249
Analysis of Crop Yield Prediction Using Random Forest Regression Model
- Open Access
- 6 December 2022
- Published by Univ. of Malaya in Malaysian Journal of Computer Science
- p. 43-52
MOBILE PHONE RECOMMENDER USING MULTI CRITERIA DECISION MAKING ALGORITHM
A study found that the depression rate is growing at an alarming rate among everyone. Many people who report symptoms of depression mostly have not been diagnosed or underwent treatments for it. If they do not get proper treatments like medication, therapy, guidance or counselling, it would be difficult for them to lead a happy and stress-free lifestyle. India is already on the cusp of a health crisis, and we urgently require a long-term solution to the problem of depression. People are more inclined to open it up to a smart machine than to a human, according to a recent study. Digital interfaces are gaining traction as feasible options for closing the gap and making mental diagnosis and treatment more accessible and inexpensive to everybody. The aim of the project is to develop a chatbot called Therapy Bot using sentiment analysis and cognitive behavioral therapy to predict the mental health status of an individual. Moreover, the chatbot can serve as a good companion to the affected by communicating with friendly manner and help them recover. The chatbot will personalize its responses based on the user's answer to keep the conversation interesting. The chatbots can be used as a complement to treatment or as a kind of interim support while waiting for an appointment. The benefit of such a method is that, rather than reaching a point where a trip to a psychologist is required, an online free version will reach a large number of people, mitigate the negative effects of depression, and contribute to a better of society.
- 6 December 2022
- Published by Univ. of Malaya in Malaysian Journal of Computer Science
- p. 43-52
Open AccessMOBILE PHONE RECOMMENDER USING MULTI CRITERIA DECISION MAKING ALGORITHM
A study found that the depression rate is growing at an alarming rate among everyone. Many people who report symptoms of depression mostly have not been diagnosed or underwent treatments for it. If they do not get proper treatments like medication, therapy, guidance or counselling, it would be difficult for them to lead a happy and stress-free lifestyle. India is already on the cusp of a health crisis, and we urgently require a long-term solution to the problem of depression. People are more inclined to open it up to a smart machine than to a human, according to a recent study. Digital interfaces are gaining traction as feasible options for closing the gap and making mental diagnosis and treatment more accessible and inexpensive to everybody. The aim of the project is to develop a chatbot called Therapy Bot using sentiment analysis and cognitive behavioral therapy to predict the mental health status of an individual. Moreover, the chatbot can serve as a good companion to the affected by communicating with friendly manner and help them recover. The chatbot will personalize its responses based on the user's answer to keep the conversation interesting. The chatbots can be used as a complement to treatment or as a kind of interim support while waiting for an appointment. The benefit of such a method is that, rather than reaching a point where a trip to a psychologist is required, an online free version will reach a large number of people, mitigate the negative effects of depression, and contribute to a better of society.
- Open Access
- 1 September 2022
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 11 (3) , 1085-1093
A comprehensive analysis of consumer decisions on Twitter dataset using machine learning algorithms
An exponential growth posting on the web about the product reviews on social media, there has been a great deal of examination being done on sorting out the purchasing behaviors of the client. This paper depends on utilizing twitter for sentiment analysis to comprehend the customer purchasing behavior. There has been a significant increase in e-commerce, particularly in persons purchasing products on the internet. As a result, it becomes a fertile hotspot for opinion analysis and belief mining. In this investigation, we look at the problem of recognizing and anticipating a client's purchase goal for an item. The sentiment analysis helps to arrive at a more indisputable outcome. In this study, the support vector machine, naive Bayes, and logistic regression methods are investigated for understanding the customer's sentiment or opinion on a specific product. These strategies have been demonstrated to be genuinely for making predictions using the analysis models which examine the client's conclusion/sentiment the most precisely. The exactness for each machine learning algorithm will be analyzed and the calculation which is the most precise would be viewed as ideal.
- 1 September 2022
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 11 (3) , 1085-1093
Open AccessA comprehensive analysis of consumer decisions on Twitter dataset using machine learning algorithms
An exponential growth posting on the web about the product reviews on social media, there has been a great deal of examination being done on sorting out the purchasing behaviors of the client. This paper depends on utilizing twitter for sentiment analysis to comprehend the customer purchasing behavior. There has been a significant increase in e-commerce, particularly in persons purchasing products on the internet. As a result, it becomes a fertile hotspot for opinion analysis and belief mining. In this investigation, we look at the problem of recognizing and anticipating a client's purchase goal for an item. The sentiment analysis helps to arrive at a more indisputable outcome. In this study, the support vector machine, naive Bayes, and logistic regression methods are investigated for understanding the customer's sentiment or opinion on a specific product. These strategies have been demonstrated to be genuinely for making predictions using the analysis models which examine the client's conclusion/sentiment the most precisely. The exactness for each machine learning algorithm will be analyzed and the calculation which is the most precise would be viewed as ideal.
- Open Access
- 6 February 2022
- Published by Poltekkes Kemenkes Surabaya in Indonesian Journal of electronics, electromedical engineering, and medical informatics
- Vol. 4 (1) , 23-27
Application of IoT Using nodeMCU ESP8266 on the Syringe Pump Device to Increase Patient Safety
Nowadays, health care has turned out to be more technology-oriented. Today's technology is demanded to be practical and easy to use. The number of mobile devices based on the Android operating system has increased significantly based on data. The open-source nature of android helps in programming applications easily according to user requirements. The syringe pump is a medical device that functions to enter the medicinal fluid into the patient's body at a specific dose and time automatically in mL/hour. The syringe pump is generally equipped with an alarm. Alarms have a vital function to provide information to nurses or doctors on duty. Medical officers cannot hear alarms outside the patient monitoring room. This paper aims to design a syringe pump that is equipped with a NodeMCU8266 WiFi module to provide notifications via a smartphone so that nurses or doctors can know the alarm even though they are outside the patient monitoring room. So, this is expected can improve patient safety. In addition, this paper also aims to verify the size of the syringe against the drug flow rate. Based on the test results, the syringe pump can control the drug flow rate with a sensor accuracy of 0.0217 and an error rate of 0.6% at a target volume of 5mL. The syringe pump can also send alarm notifications to smartphones in real-time.
- 6 February 2022
- Published by Poltekkes Kemenkes Surabaya in Indonesian Journal of electronics, electromedical engineering, and medical informatics
- Vol. 4 (1) , 23-27
Open AccessApplication of IoT Using nodeMCU ESP8266 on the Syringe Pump Device to Increase Patient Safety
Nowadays, health care has turned out to be more technology-oriented. Today's technology is demanded to be practical and easy to use. The number of mobile devices based on the Android operating system has increased significantly based on data. The open-source nature of android helps in programming applications easily according to user requirements. The syringe pump is a medical device that functions to enter the medicinal fluid into the patient's body at a specific dose and time automatically in mL/hour. The syringe pump is generally equipped with an alarm. Alarms have a vital function to provide information to nurses or doctors on duty. Medical officers cannot hear alarms outside the patient monitoring room. This paper aims to design a syringe pump that is equipped with a NodeMCU8266 WiFi module to provide notifications via a smartphone so that nurses or doctors can know the alarm even though they are outside the patient monitoring room. So, this is expected can improve patient safety. In addition, this paper also aims to verify the size of the syringe against the drug flow rate. Based on the test results, the syringe pump can control the drug flow rate with a sensor accuracy of 0.0217 and an error rate of 0.6% at a target volume of 5mL. The syringe pump can also send alarm notifications to smartphones in real-time.