Dr.T.M. Thiyagu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Dr.T.M. Thiyagu has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- A. Samson Arun Raj
Department of Computer Science and Engineering, Karunya Institute of Technology and Sciences, Karunya Nagar, Coimbatore, India
1 shared publication - Anjali Rawat
Computer Science Department (CS), SVIIT, Shri Vaishnav Vidyapeeth Vishwavidyalaya, Indore, India
1 shared publication - Antonio Gonzalez-Torres
Costa Rica Institute of Technology, Costa Rica
1 shared publication - Azhagiri Mahendiran
Department of Computer Science and Engineering, SRM Institute of Science and Technology, Chennai, India
1 shared publication - C. Shanmuganathan
Department of Computer Science and Engineering SRM Institute of Science and Technology -600089 1 2 3 Ramapuram Tamil Nadu Chennai India
1 shared publication - Dr.V. Jyothi
Department of ECE, JNTUA, Ananthapuram, India
1 shared publication
Latest Publications
- Open Access
- 25 June 2025
- Published by The Research Publication in Indian Journal of Information Sources and Services
- Vol. 15 (2) , 110-115
Implementation of Voice Search Technology in Digital Library Systems
Incorporating voice search features into digital libraries enhances user interaction and accessibility. Users with disabilities and many others often struggle with typing, making precise keyword entry, which digital libraries require, irritating and frustrating. This project focuses on hands-free walking technology using voice recognition optimized for easy interaction with digital libraries. With NLP and speech recognition, users can provide simple commands, and the system will use the databases to look up and retrieve pertinent information. The system interprets the intent behind queries spoken out loud, processes them, and returns relevant data stored in the digital library. Search operations become less burdensome, and users are able to multitask as complex automation eases effort on the user experience. Furthermore, voice searches need not be restricted to English, expanding access to digital resources. Additionally, the system employs algorithms based on machine learning to understand user preferences better, and automate tasks to enhance accuracy, providing users with tailored responses to queries. This paper discusses the architecture, implementation, and advantages of voice searching in digital libraries, considering background noises, accents, and privacy concerns. This proposed solution aims to foster the adaptability of digital libraries with contemporary technologies, enhancing usability and inclusivity.
- 25 June 2025
- Published by The Research Publication in Indian Journal of Information Sources and Services
- Vol. 15 (2) , 110-115
Open AccessImplementation of Voice Search Technology in Digital Library Systems
Incorporating voice search features into digital libraries enhances user interaction and accessibility. Users with disabilities and many others often struggle with typing, making precise keyword entry, which digital libraries require, irritating and frustrating. This project focuses on hands-free walking technology using voice recognition optimized for easy interaction with digital libraries. With NLP and speech recognition, users can provide simple commands, and the system will use the databases to look up and retrieve pertinent information. The system interprets the intent behind queries spoken out loud, processes them, and returns relevant data stored in the digital library. Search operations become less burdensome, and users are able to multitask as complex automation eases effort on the user experience. Furthermore, voice searches need not be restricted to English, expanding access to digital resources. Additionally, the system employs algorithms based on machine learning to understand user preferences better, and automate tasks to enhance accuracy, providing users with tailored responses to queries. This paper discusses the architecture, implementation, and advantages of voice searching in digital libraries, considering background noises, accents, and privacy concerns. This proposed solution aims to foster the adaptability of digital libraries with contemporary technologies, enhancing usability and inclusivity.
- 14 September 2024
- Published by Springer Nature in SN Computer Science
- Vol. 5 (7) , 1-13
Detection of Dysrhythmias from Electrocardiogram Signals Using Noise Removal and Deep Learning Techniques
- 14 September 2024
- Published by Springer Nature in SN Computer Science
- Vol. 5 (7) , 1-13
Detection of Dysrhythmias from Electrocardiogram Signals Using Noise Removal and Deep Learning Techniques
- 2 March 2024
- Published by SASA Publications in Journal of Internet Services and Information Security
- Vol. 14 (1) , 37-51
A Data Management System for Smart Cities Leveraging Artificial Intelligence Modeling Techniques to Enhance Privacy and Security
Smart cities are metropolitan areas that use sophisticated technology to increase efficiency, sustainability, and overall quality of life. The potential for transformation is tremendous, with applications ranging from Internet of Things (IoT)-driven infrastructure to data-driven governance. Effectively handling the abundant data produced in smart cities requires stringent security and privacy protocols. This research aims to tackle these difficulties by introducing the suggested Artificial Intelligence-based Data Management System (AI-DMS) for Smart Cities. AI-DMS seeks to optimize the data processing pipeline, guaranteeing effectiveness throughout the process, from data extraction to publication. Implementing a Multi-Level Sensitive Model is a notable addition, as it classifies data into three categories: sensitive, quasi-sensitive, and public. This allows for more nuanced sharing of data. Privacy preservation is accomplished using Principal Component Analysis (PCA), a comprehensive technique encompassing feature mapping, selection, normalization, and transformation. The simulation results demonstrate that AI-DMS outperforms other methods. It achieves a Data Quality Score of 95.12% (training) and 93.76% (testing), a Privacy Preservation Rate of 85.23% (training) and 82.76% (testing), a Processing Efficiency of 90.54% (training) and 88.76% (testing), a Sensitivity Model Accuracy of 80.12% (training) and 78.45% (testing), and a Data Access Time of 22.76 ms (training) and 21.32 ms (testing). The results highlight AI-DMS as a reliable and effective system, guaranteeing superior smart city data management that is secure and precise. This contribution aligns with the changing urban scene, offering improvements in decision-making based on data while still ensuring privacy and security.
- 2 March 2024
- Published by SASA Publications in Journal of Internet Services and Information Security
- Vol. 14 (1) , 37-51
A Data Management System for Smart Cities Leveraging Artificial Intelligence Modeling Techniques to Enhance Privacy and Security
Smart cities are metropolitan areas that use sophisticated technology to increase efficiency, sustainability, and overall quality of life. The potential for transformation is tremendous, with applications ranging from Internet of Things (IoT)-driven infrastructure to data-driven governance. Effectively handling the abundant data produced in smart cities requires stringent security and privacy protocols. This research aims to tackle these difficulties by introducing the suggested Artificial Intelligence-based Data Management System (AI-DMS) for Smart Cities. AI-DMS seeks to optimize the data processing pipeline, guaranteeing effectiveness throughout the process, from data extraction to publication. Implementing a Multi-Level Sensitive Model is a notable addition, as it classifies data into three categories: sensitive, quasi-sensitive, and public. This allows for more nuanced sharing of data. Privacy preservation is accomplished using Principal Component Analysis (PCA), a comprehensive technique encompassing feature mapping, selection, normalization, and transformation. The simulation results demonstrate that AI-DMS outperforms other methods. It achieves a Data Quality Score of 95.12% (training) and 93.76% (testing), a Privacy Preservation Rate of 85.23% (training) and 82.76% (testing), a Processing Efficiency of 90.54% (training) and 88.76% (testing), a Sensitivity Model Accuracy of 80.12% (training) and 78.45% (testing), and a Data Access Time of 22.76 ms (training) and 21.32 ms (testing). The results highlight AI-DMS as a reliable and effective system, guaranteeing superior smart city data management that is secure and precise. This contribution aligns with the changing urban scene, offering improvements in decision-making based on data while still ensuring privacy and security.
- Open Access
- 1 January 2024
- Published by GN1 Sistemas e Publicacoes Ltd. in ITEGAM-Journal of Engineering and Technology for Industrial Applications (ITEGAM-JETIA)
- Vol. 10 (50) , 128-137
EFR-Net: Enhanced Fracture Prediction in Osteoporosis with U-Net-Based Analysis
Osteoporosis, a prevalent bone disease, is characterized by the equation , where is bone density, is maximum bone density, and is osteoporosis rate. Conventional imaging techniques, governed by the formula where accuracy is, is image thresholding, and is scan resolution), often yield a detection accuracy of merely 75%. In this work, we introduce the EFR-Net: a U-Net-based deep learning model. Its efficacy is represented by the equation , where is the new accuracy, is the fraction of fracture-prone regions detected, is the Dice coefficient, and is the noise reduction factor. Leveraging a comprehensive dataset of 10,000 bone scans, our model, adhering to the above equation, achieved a commendable accuracy rate of 89%. This translates to a mathematical improvement represented by , yielding a 14% enhancement over traditional methods. Moreover, the reduction in false negatives, a critical metric in medical diagnoses, can be quantified by , where and are the old and new false negatives respectively. EFR-Net's innovative approach and promising results underline its potential in revolutionizing osteoporosis-related fracture prediction, offering a robust bridge between computational advancements and clinical necessities.
- 1 January 2024
- Published by GN1 Sistemas e Publicacoes Ltd. in ITEGAM-Journal of Engineering and Technology for Industrial Applications (ITEGAM-JETIA)
- Vol. 10 (50) , 128-137
Open AccessEFR-Net: Enhanced Fracture Prediction in Osteoporosis with U-Net-Based Analysis
Osteoporosis, a prevalent bone disease, is characterized by the equation , where is bone density, is maximum bone density, and is osteoporosis rate. Conventional imaging techniques, governed by the formula where accuracy is, is image thresholding, and is scan resolution), often yield a detection accuracy of merely 75%. In this work, we introduce the EFR-Net: a U-Net-based deep learning model. Its efficacy is represented by the equation , where is the new accuracy, is the fraction of fracture-prone regions detected, is the Dice coefficient, and is the noise reduction factor. Leveraging a comprehensive dataset of 10,000 bone scans, our model, adhering to the above equation, achieved a commendable accuracy rate of 89%. This translates to a mathematical improvement represented by , yielding a 14% enhancement over traditional methods. Moreover, the reduction in false negatives, a critical metric in medical diagnoses, can be quantified by , where and are the old and new false negatives respectively. EFR-Net's innovative approach and promising results underline its potential in revolutionizing osteoporosis-related fracture prediction, offering a robust bridge between computational advancements and clinical necessities.