Devi Thirupathi | Bharathiar University, Coimbatore, Tamilnadu, INDIA (original) (raw)
Papers by Devi Thirupathi
The increasing complexity, heterogeneity and dynamism of Web and its applications have made Web i... more The increasing complexity, heterogeneity and dynamism of Web and its applications have made Web information retrieval less recent, less relevant and unmanageable. The search engines face the problem of keeping the contents of its repository consistent with the pages present in the global database using optimum resource utilization. This paper proposes a traffic adaptive optimum updating scheme (TAOS) to eliminate
International Journal of Services Technology and Management, 2018
Traditionally business software has been hosted in the client's business, typically referred to a... more Traditionally business software has been hosted in the client's business, typically referred to as 'in-house' or 'on-premises' solution that is hosted on a client-owned server. In SAAS software deployment, for servicing multiple clients, vendors deploy same software instance in multiple machines like multiple instances. The implementation would be done in vendor or client location. This type of deployment model consumes more resources such as software implementation effort for each machine and hardware resources. Multi-tenant web application maintenance issues are usage calculation for application, tenant data security issue in database and file server, authorisation control for application resources access and application deployment platform management of resource usage monitor and control for service level agreement (SLA) for response time maintenance. Multi-tenant web framework (MWF) provides standard set of components and utilities that expedites the process of developing multi-tenant functionalities and converts an ordinary web application in to a multi-tenant enabled application. MWF consists of features that include authentication, authorisation, tenant tracking, application platform controller and securing tenant data.
Acute Abdomen is defined as a syndrome induced by a wide variety of pathological conditions that ... more Acute Abdomen is defined as a syndrome induced by a wide variety of pathological conditions that require emergent medical or more often surgical management. The cardinal presenting symptom is abdominal pain which has many underlying causes. Over the past 10 years, sonography has gained acceptance for examining patients with acute abdominal pain. Sonography is dynamic, noninvasive, rapid, inexpensive, and readily accessible. It is very tedious and time consuming to analyze the sonographic images manually. The authors propose a novel method for diagnosing acute appendicitis using Euclidean distance measures. This paper details the image mining system that automates the diagnosis of acute appendicitis with significant speed up, experimentation methods, real data used for testing and the result.
International Journal of Computer Science, Engineering and Applications, 2012
All textile industries aim to produce competitive materials and the competition enhancement depen... more All textile industries aim to produce competitive materials and the competition enhancement depends mainly on designs and quality of the dresses produced by each industry. Every day, a vast amount of textile images are being generated such as images of shirts, jeans, t-shirts and sarees. A principal driver of innovation is World Wide Web, unleashing publication at the scale of tens and millions of content creators. Images play an important role as a picture is worth thousand words in the field of textile design and marketing. A retrieving of images needs special concepts such as image annotation, context, and image content and image values. This research work aimed at studying the image mining process in detail and analyzes the methods for retrieval. The textile images analyze various methods for clustering the images and developing an algorithm for the same. The retrieval method considered is based on relevance feedback, scalable method, edge histogram and color layout. The image clustering algorithm is designed based on color descriptors and k-means clustering algorithm. A software prototype to prove the proposed algorithm has been developed using net beans integrated development environment and found successful.
2021 5th International Conference on Electronics, Communication and Aerospace Technology (ICECA)
Abstract— If you are a busy person, a scheduler would probably be the most indispensable tool you... more Abstract— If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Schedulers come in all forms – diaries, memo pads, computer based scheduling programs, organizers and PDAs (Personal Digital Assistants). All these devices use handwriting or a keyboard/pointing device as the primary mode of input, which is not always efficient, particularly when the user does not have time to note anything down. The Speech Driven Personal Scheduling System (SDPSS) proposed in this paper uses speech as a mode of input which is very convenient and useful for a quick note of things. Also, the use of a wireless phone for the purpose eliminates the need to carry a PDA in addition to a phone.
The thesis proposes an integrated product introduction (PI) information model for managing the pr... more The thesis proposes an integrated product introduction (PI) information model for managing the product introduction process in an efficient manner. Through the process of product introduction, ideas and needs are converted to the information from which technical systems and products can be made and sold. Two critical factors for its success are the management of the product introduction activities, and the quality and functionality of its output (i.e. the product) which itself depends on the quality of the PI process. The process is as effective as the decisions made within it, and as efficient as the speed with which the information required for each decision is made available. In order to improve the efficiency of the management of the project in meeting its diverse targets (project time, project cost, product cost and uparrow product functionality), a model that integrates the targets would be essential in relating the activities of the project to their outcomes. Existing informa...
This research is concerned with the study and analysis of product functionality and its associati... more This research is concerned with the study and analysis of product functionality and its association with the physical structure of the product and to develop a product functionality information model (PFIM) for effective management of the product development process. PFIM will enable the designer to make good use of the existing product data and bring more control of the design to the proposed new product. The information model will use the function diagram, which depicts function, responses, failure modes and the critical parameters and noise factors that control these. product and which of these creates most customer dissatisfaction. This leads the designer into an understanding of which of the functions of the mature product are least robust or most prone to failure. The model will meet the requirement to link the failure modes with the critical parameters that drive them. The critical parameters provide the basis for deciding where the design effort should be applied to enable m...
2019 IEEE Tenth International Conference on Technology for Education (T4E), 2019
Digital Learning provides flexibility for learning in terms of space, time and age. This paper ad... more Digital Learning provides flexibility for learning in terms of space, time and age. This paper addresses the need for digital learning, the profile required of a digital learner and a teacher, the pedagogy used in the digital learning process and the technological devices that could make this transformation happen in education. Case studies showcasing the recent trends used to facilitate digital learning are detailed.
The acute appendicitis necessitates emergency abdominal surgery and in the past decade, sonograph... more The acute appendicitis necessitates emergency abdominal surgery and in the past decade, sonography has gained acceptance for examining patients with acute abdominal pain. Sonographic imaging is dynamic, noninvasive, rapid, inexpensive and readily accessible. Manual analysis of sonographic image is a tedious process and consumes enormous time. The authors aim at design and development of an automatic system that would detect acute appendicitis by taking sonographic images as input. This paper describes the image mining system that automates the diagnosis of acute appendicitis with significant time reduction. The experimentation methods, results of the testing using real data are detailed in this paper. The data set of 21 patients’ sonographic images collected from a reputed hospital in India has been used as input. It is concluded that an algorithm integrating region based segmentation and euclidean distance method yields accurate results in diagnosing
If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Sc... more If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Schedulers come in all forms – diaries, memo pads, computer based scheduling programs, organizers and PDAs (Personal Digital Assistants). All these devices use handwriting or a keyboard/pointing device as the primary mode of input, which is not always efficient, particularly when the user does not have time to note anything down. The Speech Driven Personal Scheduling System (SDPSS) proposed in this paper uses speech as a mode of input which is very convenient and useful for a quick note of things. Also, the use of a wireless phone for the purpose eliminates the need to carry a PDA in addition to a phone.
All textile industries aim to produce competitive materials and the competition enhancement depen... more All textile industries aim to produce competitive materials and the competition enhancement depends mainly on designs and quality of the dresses produced by each industry. Every day, a vast amount of textile images are being generated such as images of shirts, jeans, t-shirts and sarees. A principal driver of innovation is World Wide Web, unleashing publication at the scale of tens and millions of content creators. Images play an important role as a picture is worth thousand words in the field of textile design and marketing. A retrieving of images needs special concepts such as image annotation, context, and image content and image values. This research work aimed at studying the image mining process in detail and analyzes the methods for retrieval. The textile images analyze various methods for clustering the images and developing an algorithm for the same. The retrieval method considered is based on relevance feedback, scalable method, edge histogram and color layout. The image c...
The thesis proposes an integrated product introduction (PI) information model for managing the pr... more The thesis proposes an integrated product introduction (PI) information model for managing the product introduction process in an efficient manner. Through the process of product introduction, ideas and needs are converted to the information from which technical systems and products can be made and sold. Two critical factors for its success are the management of the product introduction activities, and the quality and functionality of its output (i.e. the product) which itself depends on the quality of the PI process. The process is as effective as the decisions made within it, and as efficient as the speed with which the information required for each decision is made available. In order to improve the efficiency of the management of the project in meeting its diverse targets (project time, project cost, product cost and uparrow product functionality), a model that integrates the targets would be essential in relating the activities of the project to their outcomes. Existing informa...
Procedia Computer Science, 2015
Procedia Computer Science, 2015
Journal of Mechanics in Medicine and Biology, 2012
Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV)... more Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV) has been proven to be a non-invasive marker of the autonomic modulation of the heart. Nonlinear analyses of HRV signals have shown that the HRV is reduced signi¯cantly in patients with CAD. Therefore, in this work, we extracted nonlinear features from the HRV signals using the following techniques: recurrence plots (RP), Poincare plots, and detrended°uctuation analysis (DFA). We also extracted three types of entropy, namely, Shannon entropy (ShanEn), approximation entropy (ApEn), and sample entropy (SampEn). These features were subjected to principal component analysis (PCA). The signi¯cant principal components were evaluated using eight classi¯cation techniques, and the performances of these techniques were evaluated to determine which presented the highest accuracy in classifying normal and CAD classes. We observed that the multilayer perceptron (MLP) method resulted in the highest classi¯cation accuracy (89.5%) using our proposed technique.
Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV)... more Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV) has been proven to be a non-invasive marker of the autonomic modulation of the heart. Nonlinear analyses of HRV signals have shown that the HRV is reduced signi¯cantly in patients with CAD. Therefore, in this work, we extracted nonlinear features from the HRV signals using the following techniques: recurrence plots (RP), Poincare plots, and detrended°uctuation analysis (DFA). We also extracted three types of entropy, namely, Shannon entropy (ShanEn), approximation entropy (ApEn), and sample entropy (SampEn). These features were subjected to principal component analysis (PCA). The signi¯cant principal components were evaluated using eight classi¯cation techniques, and the performances of these techniques were evaluated to determine which presented the highest accuracy in classifying normal and CAD classes. We observed that the multilayer perceptron (MLP) method resulted in the highest classi¯cation accuracy (89.5%) using our proposed technique.
The increasing complexity, heterogeneity and dynamism of Web and its applications have made Web i... more The increasing complexity, heterogeneity and dynamism of Web and its applications have made Web information retrieval less recent, less relevant and unmanageable. The search engines face the problem of keeping the contents of its repository consistent with the pages present in the global database using optimum resource utilization. This paper proposes a traffic adaptive optimum updating scheme (TAOS) to eliminate
International Journal of Services Technology and Management, 2018
Traditionally business software has been hosted in the client's business, typically referred to a... more Traditionally business software has been hosted in the client's business, typically referred to as 'in-house' or 'on-premises' solution that is hosted on a client-owned server. In SAAS software deployment, for servicing multiple clients, vendors deploy same software instance in multiple machines like multiple instances. The implementation would be done in vendor or client location. This type of deployment model consumes more resources such as software implementation effort for each machine and hardware resources. Multi-tenant web application maintenance issues are usage calculation for application, tenant data security issue in database and file server, authorisation control for application resources access and application deployment platform management of resource usage monitor and control for service level agreement (SLA) for response time maintenance. Multi-tenant web framework (MWF) provides standard set of components and utilities that expedites the process of developing multi-tenant functionalities and converts an ordinary web application in to a multi-tenant enabled application. MWF consists of features that include authentication, authorisation, tenant tracking, application platform controller and securing tenant data.
Acute Abdomen is defined as a syndrome induced by a wide variety of pathological conditions that ... more Acute Abdomen is defined as a syndrome induced by a wide variety of pathological conditions that require emergent medical or more often surgical management. The cardinal presenting symptom is abdominal pain which has many underlying causes. Over the past 10 years, sonography has gained acceptance for examining patients with acute abdominal pain. Sonography is dynamic, noninvasive, rapid, inexpensive, and readily accessible. It is very tedious and time consuming to analyze the sonographic images manually. The authors propose a novel method for diagnosing acute appendicitis using Euclidean distance measures. This paper details the image mining system that automates the diagnosis of acute appendicitis with significant speed up, experimentation methods, real data used for testing and the result.
International Journal of Computer Science, Engineering and Applications, 2012
All textile industries aim to produce competitive materials and the competition enhancement depen... more All textile industries aim to produce competitive materials and the competition enhancement depends mainly on designs and quality of the dresses produced by each industry. Every day, a vast amount of textile images are being generated such as images of shirts, jeans, t-shirts and sarees. A principal driver of innovation is World Wide Web, unleashing publication at the scale of tens and millions of content creators. Images play an important role as a picture is worth thousand words in the field of textile design and marketing. A retrieving of images needs special concepts such as image annotation, context, and image content and image values. This research work aimed at studying the image mining process in detail and analyzes the methods for retrieval. The textile images analyze various methods for clustering the images and developing an algorithm for the same. The retrieval method considered is based on relevance feedback, scalable method, edge histogram and color layout. The image clustering algorithm is designed based on color descriptors and k-means clustering algorithm. A software prototype to prove the proposed algorithm has been developed using net beans integrated development environment and found successful.
2021 5th International Conference on Electronics, Communication and Aerospace Technology (ICECA)
Abstract— If you are a busy person, a scheduler would probably be the most indispensable tool you... more Abstract— If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Schedulers come in all forms – diaries, memo pads, computer based scheduling programs, organizers and PDAs (Personal Digital Assistants). All these devices use handwriting or a keyboard/pointing device as the primary mode of input, which is not always efficient, particularly when the user does not have time to note anything down. The Speech Driven Personal Scheduling System (SDPSS) proposed in this paper uses speech as a mode of input which is very convenient and useful for a quick note of things. Also, the use of a wireless phone for the purpose eliminates the need to carry a PDA in addition to a phone.
The thesis proposes an integrated product introduction (PI) information model for managing the pr... more The thesis proposes an integrated product introduction (PI) information model for managing the product introduction process in an efficient manner. Through the process of product introduction, ideas and needs are converted to the information from which technical systems and products can be made and sold. Two critical factors for its success are the management of the product introduction activities, and the quality and functionality of its output (i.e. the product) which itself depends on the quality of the PI process. The process is as effective as the decisions made within it, and as efficient as the speed with which the information required for each decision is made available. In order to improve the efficiency of the management of the project in meeting its diverse targets (project time, project cost, product cost and uparrow product functionality), a model that integrates the targets would be essential in relating the activities of the project to their outcomes. Existing informa...
This research is concerned with the study and analysis of product functionality and its associati... more This research is concerned with the study and analysis of product functionality and its association with the physical structure of the product and to develop a product functionality information model (PFIM) for effective management of the product development process. PFIM will enable the designer to make good use of the existing product data and bring more control of the design to the proposed new product. The information model will use the function diagram, which depicts function, responses, failure modes and the critical parameters and noise factors that control these. product and which of these creates most customer dissatisfaction. This leads the designer into an understanding of which of the functions of the mature product are least robust or most prone to failure. The model will meet the requirement to link the failure modes with the critical parameters that drive them. The critical parameters provide the basis for deciding where the design effort should be applied to enable m...
2019 IEEE Tenth International Conference on Technology for Education (T4E), 2019
Digital Learning provides flexibility for learning in terms of space, time and age. This paper ad... more Digital Learning provides flexibility for learning in terms of space, time and age. This paper addresses the need for digital learning, the profile required of a digital learner and a teacher, the pedagogy used in the digital learning process and the technological devices that could make this transformation happen in education. Case studies showcasing the recent trends used to facilitate digital learning are detailed.
The acute appendicitis necessitates emergency abdominal surgery and in the past decade, sonograph... more The acute appendicitis necessitates emergency abdominal surgery and in the past decade, sonography has gained acceptance for examining patients with acute abdominal pain. Sonographic imaging is dynamic, noninvasive, rapid, inexpensive and readily accessible. Manual analysis of sonographic image is a tedious process and consumes enormous time. The authors aim at design and development of an automatic system that would detect acute appendicitis by taking sonographic images as input. This paper describes the image mining system that automates the diagnosis of acute appendicitis with significant time reduction. The experimentation methods, results of the testing using real data are detailed in this paper. The data set of 21 patients’ sonographic images collected from a reputed hospital in India has been used as input. It is concluded that an algorithm integrating region based segmentation and euclidean distance method yields accurate results in diagnosing
If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Sc... more If you are a busy person, a scheduler would probably be the most indispensable tool you carry. Schedulers come in all forms – diaries, memo pads, computer based scheduling programs, organizers and PDAs (Personal Digital Assistants). All these devices use handwriting or a keyboard/pointing device as the primary mode of input, which is not always efficient, particularly when the user does not have time to note anything down. The Speech Driven Personal Scheduling System (SDPSS) proposed in this paper uses speech as a mode of input which is very convenient and useful for a quick note of things. Also, the use of a wireless phone for the purpose eliminates the need to carry a PDA in addition to a phone.
All textile industries aim to produce competitive materials and the competition enhancement depen... more All textile industries aim to produce competitive materials and the competition enhancement depends mainly on designs and quality of the dresses produced by each industry. Every day, a vast amount of textile images are being generated such as images of shirts, jeans, t-shirts and sarees. A principal driver of innovation is World Wide Web, unleashing publication at the scale of tens and millions of content creators. Images play an important role as a picture is worth thousand words in the field of textile design and marketing. A retrieving of images needs special concepts such as image annotation, context, and image content and image values. This research work aimed at studying the image mining process in detail and analyzes the methods for retrieval. The textile images analyze various methods for clustering the images and developing an algorithm for the same. The retrieval method considered is based on relevance feedback, scalable method, edge histogram and color layout. The image c...
The thesis proposes an integrated product introduction (PI) information model for managing the pr... more The thesis proposes an integrated product introduction (PI) information model for managing the product introduction process in an efficient manner. Through the process of product introduction, ideas and needs are converted to the information from which technical systems and products can be made and sold. Two critical factors for its success are the management of the product introduction activities, and the quality and functionality of its output (i.e. the product) which itself depends on the quality of the PI process. The process is as effective as the decisions made within it, and as efficient as the speed with which the information required for each decision is made available. In order to improve the efficiency of the management of the project in meeting its diverse targets (project time, project cost, product cost and uparrow product functionality), a model that integrates the targets would be essential in relating the activities of the project to their outcomes. Existing informa...
Procedia Computer Science, 2015
Procedia Computer Science, 2015
Journal of Mechanics in Medicine and Biology, 2012
Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV)... more Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV) has been proven to be a non-invasive marker of the autonomic modulation of the heart. Nonlinear analyses of HRV signals have shown that the HRV is reduced signi¯cantly in patients with CAD. Therefore, in this work, we extracted nonlinear features from the HRV signals using the following techniques: recurrence plots (RP), Poincare plots, and detrended°uctuation analysis (DFA). We also extracted three types of entropy, namely, Shannon entropy (ShanEn), approximation entropy (ApEn), and sample entropy (SampEn). These features were subjected to principal component analysis (PCA). The signi¯cant principal components were evaluated using eight classi¯cation techniques, and the performances of these techniques were evaluated to determine which presented the highest accuracy in classifying normal and CAD classes. We observed that the multilayer perceptron (MLP) method resulted in the highest classi¯cation accuracy (89.5%) using our proposed technique.
Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV)... more Coronary artery disease (CAD) is a leading cause of death worldwide. Heart rate variability (HRV) has been proven to be a non-invasive marker of the autonomic modulation of the heart. Nonlinear analyses of HRV signals have shown that the HRV is reduced signi¯cantly in patients with CAD. Therefore, in this work, we extracted nonlinear features from the HRV signals using the following techniques: recurrence plots (RP), Poincare plots, and detrended°uctuation analysis (DFA). We also extracted three types of entropy, namely, Shannon entropy (ShanEn), approximation entropy (ApEn), and sample entropy (SampEn). These features were subjected to principal component analysis (PCA). The signi¯cant principal components were evaluated using eight classi¯cation techniques, and the performances of these techniques were evaluated to determine which presented the highest accuracy in classifying normal and CAD classes. We observed that the multilayer perceptron (MLP) method resulted in the highest classi¯cation accuracy (89.5%) using our proposed technique.