Image colorization produces a colored version of a grayscale picture using only a single input im... more Image colorization produces a colored version of a grayscale picture using only a single input image. Given a greyscale picture as input, this study attempts to create a realistic color version of the image. This work presents a method that uses Convolutional Neural Networks to colorize a grayscale image. The proposed system takes a single grayscale image as the input and learns the map between black-and-white image and the colored picture. The proposed approach, which is implemented using Caffe framework, has been tested on grayscale images and has achieved good accuracy and speed. Color mapping using Caffe2 deep learning framework clearly proves to be better suited for human images compared with previous methodologies. This study also presents a novel system for automatically colorizing grayscale images, which has potential uses in applications such as digital painting, image restoration, medical imaging and image editing.
A Chatbot is an artificial intelligence-based software system that imitates human communication w... more A Chatbot is an artificial intelligence-based software system that imitates human communication with Artificial Intelligence. It is formulated to be the eventual personal assistant for amusement, assisting with tasks like respond to queries, obtaining driving instructions, augmenting the sensors in a smart house and playing preferred music among others. Chatbots are becoming increasingly common among companies due to their ability to minimize consumer care costs and manage various users at once. However, to perform these many tasks we need the chatbot to be as efficient as possible. To provide a solution to this, we develop Chatbot incorporated with Artificial Intelligence Markup Language (AIML) and Latent Semantic Analysis (LSA). AIML is utilized to response for standard and common queries namely, such welcoming and greetings. Although LSA can be used to answer at any time for service-related questions, ensuring user satisfaction. Any college (or) academy may utilize this chatbot to answer the questions. Due to the advent of deep learning technology, this study provides research opportunity in Chatbots.
Communications in computer and information science, 2024
Blockchain technology finds diverse applications, encompassing the security and managing the vast... more Blockchain technology finds diverse applications, encompassing the security and managing the vast amounts of data generated by IoT devices, with cure and transparent communication between them. In Supply chain management, blockchain can track products and goods from their origin to destination. A blockchain is an autonomous digital ledger employing Distributed Ledger Technology (DLT) to securely and transparently record transactions in a decentralized manner. It accomplishes this by using a network of computers (nodes) and these components collaborate to verify and register transactions within a communal database. This research proposes a new approach called Segment Blockchain that divides a blockchain into smaller segments and enables nodes to retain only one segment in place of the entire blockchain. This approach can potentially reduce the storage requirements for participating nodes, facilitate the incorporation of addional nodes into the network and maintain a copy of the blockchain. Our proposed methodology aims to address the concern of the risk of a singular vulnerable point, wherein a malicious entity keeps all copies of a specific segment and leaves the system, causing in the irreversible deletion of that segment. The proposed blockchain system can handle big data while reducing storage space demands while ensuring heightened security for user data. Theoretical evidence shows that this is achieved by limiting the number of blocks a malicious entity has the capability to both store and distribute every segment over a cluster of cloud-based blocks, the storage burden is significantly reduced compared to conventional designs. The system was successful in reducing storage space by 33% for large scale data. This makes the proposed segmentation approach more practical for processing and managing large volumes of data.
Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide... more Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide spectrum of symptoms, that varies according to individual, overlapping symptoms with other conditions like thyroid disorders, adrenal gland disorders, and ovarian tumors. Many women tend to overlook the common indications of PCOS and only seek medical attention when they encounter difficulties in conceiving. The integration of Machine Learning (ML) tools into clinical practice in collaboration with healthcare professionals have guaranteed proper interpretation and clinical decision-making. In this research, we explored the feasibility of developing a model that utilizes ML algorithms and techniques to automate the diagnosis of PCOS. To accomplish this, we utilized a dataset containing information from 543 subjects, encompassing 42 features such as metabolic, imaging, hormonal, and biochemical parameters. Initially, data pre-processing is performed, followed by the implementation of feature selection approach to reduce the number of features. Subsequently, various classification algorithms were trained and evaluated using Decision Tree (DT), XG Boost, Ada Boost, Random Forest (RF) and Logistic Regression (LR). After conducting a comprehensive analysis, we determined that the XG Boost outperformed the other algorithms in terms achieving high accuracy 92.45%.
Biclustering enables the identification of gene groups that cannot be discovered using classical ... more Biclustering enables the identification of gene groups that cannot be discovered using classical clustering approaches, which typically operate on all experimental conditions simultaneously. A bicluster is a combination of a set of genes and a group of samples, where the genes exhibit similarity within the samples, and vice versa. When a sample shows the activation in multiple pathways, it can belong to different biclusters. The biclustering method processes two types of data matrices: binary and non-binary. Biclustering algorithms that handle binary data often struggle to strike a satisfactory equilibrium between execution speed and efficiency. This paper presents a novel biclustering algorithm for binary matrix for generating, assessing and validating bicluster, by incorporating a straightforward binary reference model for comparison. The proposed methodology is applied to gene expression data, by utilizing the neighbour joining difference matrix to identify similar genes. Adjacency matrix enables the clustering of genes with similar reactions under various conditions into coherent clusters, which plays a crucial influence in the subsequent analysis of genes. This approach exhibits improved time complexity and relevance scores compared to other biclust algorithms like Bibit and Qubic.
The attendance monitoring system provides a convenient way to track the presence of faculty membe... more The attendance monitoring system provides a convenient way to track the presence of faculty members. In the past, faculty attendance was marked using traditional methods under the supervision of higher authorities such as the principal or department head. However, these methods were often outdated and prone to errors, leading to discrepancies in attendance records over the years. To address these issues, a new system was implemented using modern technology, specifically finger-punching and face recognition. This system replaced the old record books and other outdated methods, offering a more efficient and accurate way to manage faculty attendance. Additionally, a mobile application is developed for faculty or staff members to mark their attendance using their smartphones. This application utilizes Wi-Fi and location services to record attendance on campus. The recorded attendance data is securely stored in a database, allowing the administrator or main authority to access and modify the records as needed. This integrated system not only simplifies the attendance process but also ensures data integrity and provides real-time monitoring capabilities.
Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerv... more Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerve cells. After the age of 60, the risk of developing the illness doubles every five years, with estimates that by 2050, the number will have risen to 135 million. Brain structural Imaging with Magnetic Resonance Image (MRI) has been extensively utilized to recognize AD as it can detect morphometric variations and cerebral congenital malformations. Convolutional Neural Networks (CNNs) are extensively used for image receptions and analysis because of their capacity to handle enormous amounts of unstructured data and retrieve significant characteristics automatically. A new approach involving pre-trained CNN model, VGG16 with fine tuning has been proposed for automatic detection and classification of brain MRI images for AD. The results show that the performance of the proposed modelin terms of accuracy, f1score, recall and precision is above 90%.
Image colorization produces a colored version of a grayscale picture using only a single input im... more Image colorization produces a colored version of a grayscale picture using only a single input image. Given a greyscale picture as input, this study attempts to create a realistic color version of the image. This work presents a method that uses Convolutional Neural Networks to colorize a grayscale image. The proposed system takes a single grayscale image as the input and learns the map between black-and-white image and the colored picture. The proposed approach, which is implemented using Caffe framework, has been tested on grayscale images and has achieved good accuracy and speed. Color mapping using Caffe2 deep learning framework clearly proves to be better suited for human images compared with previous methodologies. This study also presents a novel system for automatically colorizing grayscale images, which has potential uses in applications such as digital painting, image restoration, medical imaging and image editing.
A Chatbot is an artificial intelligence-based software system that imitates human communication w... more A Chatbot is an artificial intelligence-based software system that imitates human communication with Artificial Intelligence. It is formulated to be the eventual personal assistant for amusement, assisting with tasks like respond to queries, obtaining driving instructions, augmenting the sensors in a smart house and playing preferred music among others. Chatbots are becoming increasingly common among companies due to their ability to minimize consumer care costs and manage various users at once. However, to perform these many tasks we need the chatbot to be as efficient as possible. To provide a solution to this, we develop Chatbot incorporated with Artificial Intelligence Markup Language (AIML) and Latent Semantic Analysis (LSA). AIML is utilized to response for standard and common queries namely, such welcoming and greetings. Although LSA can be used to answer at any time for service-related questions, ensuring user satisfaction. Any college (or) academy may utilize this chatbot to answer the questions. Due to the advent of deep learning technology, this study provides research opportunity in Chatbots.
In book: Soft Computing and Its Engineering Applications, 2024
Blockchain technology finds diverse applications, encompassing the security and managing the vast... more Blockchain technology finds diverse applications, encompassing the security and managing the vast amounts of data generated by IoT devices, with cure and transparent communication between them. In Supply chain management, blockchain can track products and goods from their origin to destination. A blockchain is an autonomous digital ledger employing Distributed Ledger Technology (DLT) to securely and transparently record transactions in a decentralized manner. It accomplishes this by using a network of computers (nodes) and these components collaborate to verify and register transactions within a communal database. This research proposes a new approach called Segment Blockchain that divides a blockchain into smaller segments and enables nodes to retain only one segment in place of the entire blockchain. This approach can potentially reduce the storage requirements for participating nodes, facilitate the incorporation of addional nodes into the network and maintain a copy of the blockchain. Our proposed methodology aims to address the concern of the risk of a singular vulnerable point, wherein a malicious entity keeps all copies of a specific segment and leaves the system, causing in the irreversible deletion of that segment. The proposed blockchain system can handle big data while reducing storage space demands while ensuring heightened security for user data. Theoretical evidence shows that this is achieved by limiting the number of blocks a malicious entity has the capability to both store and distribute every segment over a cluster of cloud-based blocks, the storage burden is significantly reduced compared to conventional designs. The system was successful in reducing storage space by 33% for large scale data. This makes the proposed segmentation approach more practical for processing and managing large volumes of data.
2023 International Conference on Recent Advances in Information Technology for Sustainable Development (ICRAIS), 2023
Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide... more Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide spectrum of symptoms, that varies according to individual, overlapping symptoms with other conditions like thyroid disorders, adrenal gland disorders, and ovarian tumors. Many women tend to overlook the common indications of PCOS and only seek medical attention when they encounter difficulties in conceiving. The integration of Machine Learning (ML) tools into clinical practice in collaboration with healthcare professionals have guaranteed proper interpretation and clinical decision-making. In this research, we explored the feasibility of developing a model that utilizes ML algorithms and techniques to automate the diagnosis of PCOS. To accomplish this, we utilized a dataset containing information from 543 subjects, encompassing 42 features such as metabolic, imaging, hormonal, and biochemical parameters. Initially, data pre-processing is performed, followed by the implementation of feature selection approach to reduce the number of features. Subsequently, various classification algorithms were trained and evaluated using Decision Tree (DT), XG Boost, Ada Boost, Random Forest (RF) and Logistic Regression (LR). After conducting a comprehensive analysis, we determined that the XG Boost outperformed the other algorithms in terms achieving high accuracy 92.45%.
2023 4th IEEE Global Conference for Advancement in Technology (GCAT) Bangalore, India. Oct 6-8, 2023, 2023
Biclustering enables the identification of gene groups that cannot be discovered using classical ... more Biclustering enables the identification of gene groups that cannot be discovered using classical clustering approaches, which typically operate on all experimental conditions simultaneously. A bicluster is a combination of a set of genes and a group of samples, where the genes exhibit similarity within the samples, and vice versa. When a sample shows the activation in multiple pathways, it can belong to different biclusters. The biclustering method processes two types of data matrices: binary and non-binary. Biclustering algorithms that handle binary data often struggle to strike a satisfactory equilibrium between execution speed and efficiency. This paper presents a novel biclustering algorithm for binary matrix for generating, assessing and validating bicluster, by incorporating a straightforward binary reference model for comparison. The proposed methodology is applied to gene expression data, by utilizing the neighbour joining difference matrix to identify similar genes. Adjacency matrix enables the clustering of genes with similar reactions under various conditions into coherent clusters, which plays a crucial influence in the subsequent analysis of genes. This approach exhibits improved time complexity and relevance scores compared to other biclust algorithms like Bibit and Qubic.
Video observation is fundamental for guaranteeing public well-being and security in different env... more Video observation is fundamental for guaranteeing public well-being and security in different environments such as air terminals, train stations, retail outlets, and local residential locations. Existing video examination techniques face significant restrictions and difficulties, like low precision, high computational intricacy, and restricted flexibility to evolving environmental conditions. To address these difficulties, this paper proposes an original way to deal with improving video examination for distant observation by consolidating progressed object location, following and following behavioural algorithms into a unified structure. Faster deep learning object detection Algorithms like R-CNN, YOLO, and SSD are used here to precisely recognize and restrict objects of interest in surveillance videos. We investigated a few benchmark datasets and contrasted their presentation and cuttingedge techniques. The outcomes show that the proposed approach beats existing precision, strength, and efficiency strategies.
One special feature of wireless networks is their capacity to keep people in contact even when th... more One special feature of wireless networks is their capacity to keep people in contact even when they switch locations. The adaptability and agility needed for wireless communications are provided. The fields of business, education, defense, homebased and industrial applications, and the military environment have all found use for wireless and mobile communications. The world has changed significantly as a result of wireless networks since they have made it easier, more reliable, and more efficient to send information abroad or behind enemy lines. An essential productivity tool for today's mobile workforce is wireless networking. With wireless networking, we may practically access corporate and other information resources at any time and from any location. In recent times, wireless networks have become more commonplace in several fields, such as satellite broadcasting, mobile analog, and digital cellular telephones. Supporting a good quality of service (QoS) in this environment for the delivery of voice, video, and data has therefore become one of the major challenges of the twenty-first century. In this research article, we provide an overview of wireless networks, their classification, and quality of service (QOS) parameters such as bandwidth, delay, delay variation (jitter), throughput, and energy. Lastly, we provide security information regarding various parameters for end-to-end communication over wireless networks.
The attendance monitoring system provides a convenient way to track the presence of faculty membe... more The attendance monitoring system provides a convenient way to track the presence of faculty members. In the past, faculty attendance was marked using traditional methods under the supervision of higher authorities such as the principal or department head. However, these methods were often outdated and prone to errors, leading to discrepancies in attendance records over the years. To address these issues, a new system was implemented using modern technology, specifically finger-punching and face recognition. This system replaced the old record books and other outdated methods, offering a more efficient and accurate way to manage faculty attendance. Additionally, a mobile application is developed for faculty or staff members to mark their attendance using their smartphones. This application utilizes Wi-Fi and location services to record attendance on campus. The recorded attendance data is securely stored in a database, allowing the administrator or main authority to access and modify the records as needed. This integrated system not only simplifies the attendance process but also ensures data integrity and provides real-time monitoring capabilities.
Image colorization produces a colored version of a grayscale picture using only a single input im... more Image colorization produces a colored version of a grayscale picture using only a single input image. Given a greyscale picture as input, this study attempts to create a realistic color version of the image. This work presents a method that uses Convolutional Neural Networks to colorize a grayscale image. The proposed system takes a single grayscale image as the input and learns the map between black-and-white image and the colored picture. The proposed approach, which is implemented using Caffe framework, has been tested on grayscale images and has achieved good accuracy and speed. Color mapping using Caffe2 deep learning framework clearly proves to be better suited for human images compared with previous methodologies. This study also presents a novel system for automatically colorizing grayscale images, which has potential uses in applications such as digital painting, image restoration, medical imaging and image editing.
A Chatbot is an artificial intelligence-based software system that imitates human communication w... more A Chatbot is an artificial intelligence-based software system that imitates human communication with Artificial Intelligence. It is formulated to be the eventual personal assistant for amusement, assisting with tasks like respond to queries, obtaining driving instructions, augmenting the sensors in a smart house and playing preferred music among others. Chatbots are becoming increasingly common among companies due to their ability to minimize consumer care costs and manage various users at once. However, to perform these many tasks we need the chatbot to be as efficient as possible. To provide a solution to this, we develop Chatbot incorporated with Artificial Intelligence Markup Language (AIML) and Latent Semantic Analysis (LSA). AIML is utilized to response for standard and common queries namely, such welcoming and greetings. Although LSA can be used to answer at any time for service-related questions, ensuring user satisfaction. Any college (or) academy may utilize this chatbot to answer the questions. Due to the advent of deep learning technology, this study provides research opportunity in Chatbots.
Communications in computer and information science, 2024
Blockchain technology finds diverse applications, encompassing the security and managing the vast... more Blockchain technology finds diverse applications, encompassing the security and managing the vast amounts of data generated by IoT devices, with cure and transparent communication between them. In Supply chain management, blockchain can track products and goods from their origin to destination. A blockchain is an autonomous digital ledger employing Distributed Ledger Technology (DLT) to securely and transparently record transactions in a decentralized manner. It accomplishes this by using a network of computers (nodes) and these components collaborate to verify and register transactions within a communal database. This research proposes a new approach called Segment Blockchain that divides a blockchain into smaller segments and enables nodes to retain only one segment in place of the entire blockchain. This approach can potentially reduce the storage requirements for participating nodes, facilitate the incorporation of addional nodes into the network and maintain a copy of the blockchain. Our proposed methodology aims to address the concern of the risk of a singular vulnerable point, wherein a malicious entity keeps all copies of a specific segment and leaves the system, causing in the irreversible deletion of that segment. The proposed blockchain system can handle big data while reducing storage space demands while ensuring heightened security for user data. Theoretical evidence shows that this is achieved by limiting the number of blocks a malicious entity has the capability to both store and distribute every segment over a cluster of cloud-based blocks, the storage burden is significantly reduced compared to conventional designs. The system was successful in reducing storage space by 33% for large scale data. This makes the proposed segmentation approach more practical for processing and managing large volumes of data.
Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide... more Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide spectrum of symptoms, that varies according to individual, overlapping symptoms with other conditions like thyroid disorders, adrenal gland disorders, and ovarian tumors. Many women tend to overlook the common indications of PCOS and only seek medical attention when they encounter difficulties in conceiving. The integration of Machine Learning (ML) tools into clinical practice in collaboration with healthcare professionals have guaranteed proper interpretation and clinical decision-making. In this research, we explored the feasibility of developing a model that utilizes ML algorithms and techniques to automate the diagnosis of PCOS. To accomplish this, we utilized a dataset containing information from 543 subjects, encompassing 42 features such as metabolic, imaging, hormonal, and biochemical parameters. Initially, data pre-processing is performed, followed by the implementation of feature selection approach to reduce the number of features. Subsequently, various classification algorithms were trained and evaluated using Decision Tree (DT), XG Boost, Ada Boost, Random Forest (RF) and Logistic Regression (LR). After conducting a comprehensive analysis, we determined that the XG Boost outperformed the other algorithms in terms achieving high accuracy 92.45%.
Biclustering enables the identification of gene groups that cannot be discovered using classical ... more Biclustering enables the identification of gene groups that cannot be discovered using classical clustering approaches, which typically operate on all experimental conditions simultaneously. A bicluster is a combination of a set of genes and a group of samples, where the genes exhibit similarity within the samples, and vice versa. When a sample shows the activation in multiple pathways, it can belong to different biclusters. The biclustering method processes two types of data matrices: binary and non-binary. Biclustering algorithms that handle binary data often struggle to strike a satisfactory equilibrium between execution speed and efficiency. This paper presents a novel biclustering algorithm for binary matrix for generating, assessing and validating bicluster, by incorporating a straightforward binary reference model for comparison. The proposed methodology is applied to gene expression data, by utilizing the neighbour joining difference matrix to identify similar genes. Adjacency matrix enables the clustering of genes with similar reactions under various conditions into coherent clusters, which plays a crucial influence in the subsequent analysis of genes. This approach exhibits improved time complexity and relevance scores compared to other biclust algorithms like Bibit and Qubic.
The attendance monitoring system provides a convenient way to track the presence of faculty membe... more The attendance monitoring system provides a convenient way to track the presence of faculty members. In the past, faculty attendance was marked using traditional methods under the supervision of higher authorities such as the principal or department head. However, these methods were often outdated and prone to errors, leading to discrepancies in attendance records over the years. To address these issues, a new system was implemented using modern technology, specifically finger-punching and face recognition. This system replaced the old record books and other outdated methods, offering a more efficient and accurate way to manage faculty attendance. Additionally, a mobile application is developed for faculty or staff members to mark their attendance using their smartphones. This application utilizes Wi-Fi and location services to record attendance on campus. The recorded attendance data is securely stored in a database, allowing the administrator or main authority to access and modify the records as needed. This integrated system not only simplifies the attendance process but also ensures data integrity and provides real-time monitoring capabilities.
Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerv... more Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerve cells. After the age of 60, the risk of developing the illness doubles every five years, with estimates that by 2050, the number will have risen to 135 million. Brain structural Imaging with Magnetic Resonance Image (MRI) has been extensively utilized to recognize AD as it can detect morphometric variations and cerebral congenital malformations. Convolutional Neural Networks (CNNs) are extensively used for image receptions and analysis because of their capacity to handle enormous amounts of unstructured data and retrieve significant characteristics automatically. A new approach involving pre-trained CNN model, VGG16 with fine tuning has been proposed for automatic detection and classification of brain MRI images for AD. The results show that the performance of the proposed modelin terms of accuracy, f1score, recall and precision is above 90%.
Image colorization produces a colored version of a grayscale picture using only a single input im... more Image colorization produces a colored version of a grayscale picture using only a single input image. Given a greyscale picture as input, this study attempts to create a realistic color version of the image. This work presents a method that uses Convolutional Neural Networks to colorize a grayscale image. The proposed system takes a single grayscale image as the input and learns the map between black-and-white image and the colored picture. The proposed approach, which is implemented using Caffe framework, has been tested on grayscale images and has achieved good accuracy and speed. Color mapping using Caffe2 deep learning framework clearly proves to be better suited for human images compared with previous methodologies. This study also presents a novel system for automatically colorizing grayscale images, which has potential uses in applications such as digital painting, image restoration, medical imaging and image editing.
A Chatbot is an artificial intelligence-based software system that imitates human communication w... more A Chatbot is an artificial intelligence-based software system that imitates human communication with Artificial Intelligence. It is formulated to be the eventual personal assistant for amusement, assisting with tasks like respond to queries, obtaining driving instructions, augmenting the sensors in a smart house and playing preferred music among others. Chatbots are becoming increasingly common among companies due to their ability to minimize consumer care costs and manage various users at once. However, to perform these many tasks we need the chatbot to be as efficient as possible. To provide a solution to this, we develop Chatbot incorporated with Artificial Intelligence Markup Language (AIML) and Latent Semantic Analysis (LSA). AIML is utilized to response for standard and common queries namely, such welcoming and greetings. Although LSA can be used to answer at any time for service-related questions, ensuring user satisfaction. Any college (or) academy may utilize this chatbot to answer the questions. Due to the advent of deep learning technology, this study provides research opportunity in Chatbots.
In book: Soft Computing and Its Engineering Applications, 2024
Blockchain technology finds diverse applications, encompassing the security and managing the vast... more Blockchain technology finds diverse applications, encompassing the security and managing the vast amounts of data generated by IoT devices, with cure and transparent communication between them. In Supply chain management, blockchain can track products and goods from their origin to destination. A blockchain is an autonomous digital ledger employing Distributed Ledger Technology (DLT) to securely and transparently record transactions in a decentralized manner. It accomplishes this by using a network of computers (nodes) and these components collaborate to verify and register transactions within a communal database. This research proposes a new approach called Segment Blockchain that divides a blockchain into smaller segments and enables nodes to retain only one segment in place of the entire blockchain. This approach can potentially reduce the storage requirements for participating nodes, facilitate the incorporation of addional nodes into the network and maintain a copy of the blockchain. Our proposed methodology aims to address the concern of the risk of a singular vulnerable point, wherein a malicious entity keeps all copies of a specific segment and leaves the system, causing in the irreversible deletion of that segment. The proposed blockchain system can handle big data while reducing storage space demands while ensuring heightened security for user data. Theoretical evidence shows that this is achieved by limiting the number of blocks a malicious entity has the capability to both store and distribute every segment over a cluster of cloud-based blocks, the storage burden is significantly reduced compared to conventional designs. The system was successful in reducing storage space by 33% for large scale data. This makes the proposed segmentation approach more practical for processing and managing large volumes of data.
2023 International Conference on Recent Advances in Information Technology for Sustainable Development (ICRAIS), 2023
Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide... more Identifying Poly Cystic Ovary Syndrome (PCOS) can be challenging due to several factors like wide spectrum of symptoms, that varies according to individual, overlapping symptoms with other conditions like thyroid disorders, adrenal gland disorders, and ovarian tumors. Many women tend to overlook the common indications of PCOS and only seek medical attention when they encounter difficulties in conceiving. The integration of Machine Learning (ML) tools into clinical practice in collaboration with healthcare professionals have guaranteed proper interpretation and clinical decision-making. In this research, we explored the feasibility of developing a model that utilizes ML algorithms and techniques to automate the diagnosis of PCOS. To accomplish this, we utilized a dataset containing information from 543 subjects, encompassing 42 features such as metabolic, imaging, hormonal, and biochemical parameters. Initially, data pre-processing is performed, followed by the implementation of feature selection approach to reduce the number of features. Subsequently, various classification algorithms were trained and evaluated using Decision Tree (DT), XG Boost, Ada Boost, Random Forest (RF) and Logistic Regression (LR). After conducting a comprehensive analysis, we determined that the XG Boost outperformed the other algorithms in terms achieving high accuracy 92.45%.
2023 4th IEEE Global Conference for Advancement in Technology (GCAT) Bangalore, India. Oct 6-8, 2023, 2023
Biclustering enables the identification of gene groups that cannot be discovered using classical ... more Biclustering enables the identification of gene groups that cannot be discovered using classical clustering approaches, which typically operate on all experimental conditions simultaneously. A bicluster is a combination of a set of genes and a group of samples, where the genes exhibit similarity within the samples, and vice versa. When a sample shows the activation in multiple pathways, it can belong to different biclusters. The biclustering method processes two types of data matrices: binary and non-binary. Biclustering algorithms that handle binary data often struggle to strike a satisfactory equilibrium between execution speed and efficiency. This paper presents a novel biclustering algorithm for binary matrix for generating, assessing and validating bicluster, by incorporating a straightforward binary reference model for comparison. The proposed methodology is applied to gene expression data, by utilizing the neighbour joining difference matrix to identify similar genes. Adjacency matrix enables the clustering of genes with similar reactions under various conditions into coherent clusters, which plays a crucial influence in the subsequent analysis of genes. This approach exhibits improved time complexity and relevance scores compared to other biclust algorithms like Bibit and Qubic.
Video observation is fundamental for guaranteeing public well-being and security in different env... more Video observation is fundamental for guaranteeing public well-being and security in different environments such as air terminals, train stations, retail outlets, and local residential locations. Existing video examination techniques face significant restrictions and difficulties, like low precision, high computational intricacy, and restricted flexibility to evolving environmental conditions. To address these difficulties, this paper proposes an original way to deal with improving video examination for distant observation by consolidating progressed object location, following and following behavioural algorithms into a unified structure. Faster deep learning object detection Algorithms like R-CNN, YOLO, and SSD are used here to precisely recognize and restrict objects of interest in surveillance videos. We investigated a few benchmark datasets and contrasted their presentation and cuttingedge techniques. The outcomes show that the proposed approach beats existing precision, strength, and efficiency strategies.
One special feature of wireless networks is their capacity to keep people in contact even when th... more One special feature of wireless networks is their capacity to keep people in contact even when they switch locations. The adaptability and agility needed for wireless communications are provided. The fields of business, education, defense, homebased and industrial applications, and the military environment have all found use for wireless and mobile communications. The world has changed significantly as a result of wireless networks since they have made it easier, more reliable, and more efficient to send information abroad or behind enemy lines. An essential productivity tool for today's mobile workforce is wireless networking. With wireless networking, we may practically access corporate and other information resources at any time and from any location. In recent times, wireless networks have become more commonplace in several fields, such as satellite broadcasting, mobile analog, and digital cellular telephones. Supporting a good quality of service (QoS) in this environment for the delivery of voice, video, and data has therefore become one of the major challenges of the twenty-first century. In this research article, we provide an overview of wireless networks, their classification, and quality of service (QOS) parameters such as bandwidth, delay, delay variation (jitter), throughput, and energy. Lastly, we provide security information regarding various parameters for end-to-end communication over wireless networks.
The attendance monitoring system provides a convenient way to track the presence of faculty membe... more The attendance monitoring system provides a convenient way to track the presence of faculty members. In the past, faculty attendance was marked using traditional methods under the supervision of higher authorities such as the principal or department head. However, these methods were often outdated and prone to errors, leading to discrepancies in attendance records over the years. To address these issues, a new system was implemented using modern technology, specifically finger-punching and face recognition. This system replaced the old record books and other outdated methods, offering a more efficient and accurate way to manage faculty attendance. Additionally, a mobile application is developed for faculty or staff members to mark their attendance using their smartphones. This application utilizes Wi-Fi and location services to record attendance on campus. The recorded attendance data is securely stored in a database, allowing the administrator or main authority to access and modify the records as needed. This integrated system not only simplifies the attendance process but also ensures data integrity and provides real-time monitoring capabilities.
Fuzzy Logic Applications in Computer Science and Mathematics, 2023
Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerve cells.... more Alzheimer's disease (AD), a neurocognitive disorder and it evolves into the death of nerve cells. After the age of 60, the risk of developing the illness doubles every five years, with estimates that by 2050, the number will have risen to 135 million. Brain structural Imaging with Magnetic Resonance Image (MRI) has been extensively utilized to recognize AD as it can detect morphometric variations and cerebral congenital malformations. Convolutional Neural Networks (CNNs) are extensively used for image receptions and analysis because of their capacity to handle enormous amounts of unstructured data and retrieve significant characteristics automatically. A new approach involving pre-trained CNN model, VGG16 with fine tuning has been proposed for automatic detection and classification of brain MRI images for AD. The results show that the performance of the proposed modelin terms of accuracy, f1score, recall and precision is above 90%.
International Conference on Information, Communication and Computing Technology, 2023
Protein sequence motifs are defined as short, frequent fixedlength patterns in nucleic acid and p... more Protein sequence motifs are defined as short, frequent fixedlength patterns in nucleic acid and protein sequences that may reflect significant structural or functional characteristics. Although finding the gapped consensus motif is crucial, and is difficult due to the wide variety of combinatorial depictions and spaces. The proposed method, De bruijn based Pareto optimization for Emerging Substring (DPES) follows enumerative approach using Bruijn as a structure for developing the emerging substring, based on pangenome and pareto optimization on protein sequences for cervical cancer and detects the different instances of (l, d) motif instances. Multiple values for (l,d) were considered and de Bruijn node ensures that all the possible subsets are considered from the emerging substrings. Experimental results on the data set exhibit that i) DPES can identify (l,d) motifs effectively and is faster than other contemporary state of art (l,d) motif uncovering algorithms ii) DPES can identify subtle consensus that is concealed by the stronger patterns in the data.
Uploads