Lung cancer (LC) will decrease the yield, which will have a negative impact on the economy. Therefore, primary and accurate the attack finding is a priority for the agro-dependent state. In several modern technologies for early detection of LC, image processing has become a one of the essential tool so that it cannot only early to find the disease accurately, but also successfully measure it. Various approaches have been developed to detect LC based on background modelling. Most of them focus on temporal information but partially or completely ignore spatial information, making it sensitive to noise. In order to overcome these issues an improved hybrid semantic feature descriptor technique is introduced based on Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Pattern (LBP) and histogram of oriented gradients (HOG) feature extraction algorithms. And also to improve the LC segmentation problems a fuzzy c-means clustering algorithm (FCM) is used. Experiments and comparisons on publically available LIDC-IBRI dataset. To evaluate the proposed feature extraction performance three different classifiers are analysed such as artificial neural networks (ANN), recursive neural network and recurrent neural networks (RNNs).
{"title":"Hybrid Semantic Feature Descriptor and Fuzzy C-Means Clustering for Lung Cancer Detection and Classification","authors":"P. Priyadharshini, B. Zoraida","doi":"10.1166/JCTN.2021.9391","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9391","url":null,"abstract":"Lung cancer (LC) will decrease the yield, which will have a negative impact on the economy. Therefore, primary and accurate the attack finding is a priority for the agro-dependent state. In several modern technologies for early detection of LC, image processing has become a one of the\u0000 essential tool so that it cannot only early to find the disease accurately, but also successfully measure it. Various approaches have been developed to detect LC based on background modelling. Most of them focus on temporal information but partially or completely ignore spatial information,\u0000 making it sensitive to noise. In order to overcome these issues an improved hybrid semantic feature descriptor technique is introduced based on Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Pattern (LBP) and histogram of oriented gradients (HOG) feature extraction algorithms. And also\u0000 to improve the LC segmentation problems a fuzzy c-means clustering algorithm (FCM) is used. Experiments and comparisons on publically available LIDC-IBRI dataset. To evaluate the proposed feature extraction performance three different classifiers are analysed such as artificial neural networks\u0000 (ANN), recursive neural network and recurrent neural networks (RNNs).","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46028819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, a linear two-dimensional Volterra integral equation of the second kind with the discontinuous kernel is considered. The conditions for ensuring the existence of a unique continuous solution are mentioned. The product Nystrom method, as a well-known method of solving singular integral equations, is presented. Therefore, the Nystrom method is applied to the linear Volterra integral equation with the discontinuous kernel to convert it to a linear algebraic system. Some formulas are expanded in two dimensions. Weights’ functions of the Nystrom method are obtained for kernels of logarithmic and Carleman types. Some numerical applications are presented to show the efficiency and accuracy of the proposed method. Maple18 is used to compute numerical solutions. The estimated error is calculated in each case. The Nystrom method is useful and effective in treating the two-dimensional singular Volterra integral equation. Finally, we conclude that the time factor and the parameter v have a clear effect on the results.
{"title":"Nyström Method to Solve Two-Dimensional Volterra Integral Equation with Discontinuous Kernel","authors":"S. Raad, Mariam Mohammed Al-Atawi","doi":"10.1166/JCTN.2021.9718","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9718","url":null,"abstract":"In this paper, a linear two-dimensional Volterra integral equation of the second kind with the discontinuous kernel is considered. The conditions for ensuring the existence of a unique continuous solution are mentioned. The product Nystrom method, as a well-known method of solving singular\u0000 integral equations, is presented. Therefore, the Nystrom method is applied to the linear Volterra integral equation with the discontinuous kernel to convert it to a linear algebraic system. Some formulas are expanded in two dimensions. Weights’ functions of the Nystrom method are obtained\u0000 for kernels of logarithmic and Carleman types. Some numerical applications are presented to show the efficiency and accuracy of the proposed method. Maple18 is used to compute numerical solutions. The estimated error is calculated in each case. The Nystrom method is useful and effective in\u0000 treating the two-dimensional singular Volterra integral equation. Finally, we conclude that the time factor and the parameter v have a clear effect on the results.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47699477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Resource management is a significant task of scheduling and allocating resources to applications to meet the required Quality of Service (QoS) limitations by the minimization of overhead with an effective resource utilization. This paper presents a Fog-enabled Cloud computing resource management model for smart homes by the Improved Grey Wolf Optimization Strategy. Besides, Kernel Support Vector Machine (KSVM) model is applied for series forecasting of time and also of processing load of a distributed server and determine the proper resources which should be allocated for the optimization of the service response time. The presented IGWO-KSVM model has been simulated under several aspects and the outcome exhibited the outstanding performance of the presented model.
{"title":"Fog Enabled Cloud Based Intelligent Resource Management Approach Using Improved Grey Wolf Optimization Strategy and Kernel Support Vector Machine","authors":"R. Sudha, G. Indirani, S. Selvamuthukumaran","doi":"10.1166/JCTN.2021.9401","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9401","url":null,"abstract":"Resource management is a significant task of scheduling and allocating resources to applications to meet the required Quality of Service (QoS) limitations by the minimization of overhead with an effective resource utilization. This paper presents a Fog-enabled Cloud computing resource\u0000 management model for smart homes by the Improved Grey Wolf Optimization Strategy. Besides, Kernel Support Vector Machine (KSVM) model is applied for series forecasting of time and also of processing load of a distributed server and determine the proper resources which should be allocated for\u0000 the optimization of the service response time. The presented IGWO-KSVM model has been simulated under several aspects and the outcome exhibited the outstanding performance of the presented model.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48187604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Durai Pandurangan, R. S. Kumar, Lukas Gebremariam, L. Arulmurugan, S. Tamilselvan
Insufficient and poor lightning conditions affect the quality of videos and images captured by the camcorders. The low quality images decrease the performances of computer vision systems in smart traffic, video surveillance, and other imaging systems applications. In this paper, combined gray level transformation technique is proposed to enhance the less quality of illuminated images. This technique is composed of log transformation, power law transformation and adaptive histogram equalization process to improve the low light illumination image estimated using HIS color model. Finally, the enhanced illumination image is blended with original reflectance image to get enhanced color image. This paper shows that the proposed algorithm on various weakly illuminated images is enhanced better and has taken reduced computation time than previous image processing techniques.
{"title":"Combined Gray Level Transformation Technique for Low Light Color Image Enhancement","authors":"Durai Pandurangan, R. S. Kumar, Lukas Gebremariam, L. Arulmurugan, S. Tamilselvan","doi":"10.1166/JCTN.2021.9392","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9392","url":null,"abstract":"Insufficient and poor lightning conditions affect the quality of videos and images captured by the camcorders. The low quality images decrease the performances of computer vision systems in smart traffic, video surveillance, and other imaging systems applications. In this paper, combined\u0000 gray level transformation technique is proposed to enhance the less quality of illuminated images. This technique is composed of log transformation, power law transformation and adaptive histogram equalization process to improve the low light illumination image estimated using HIS color model.\u0000 Finally, the enhanced illumination image is blended with original reflectance image to get enhanced color image. This paper shows that the proposed algorithm on various weakly illuminated images is enhanced better and has taken reduced computation time than previous image processing techniques.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42357328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
R. Ferraz, Raiff Sales da Fonseca, Igor Thonke Rodrigues, Cláudio Bastos da Silva, H. T. Filho
The main goal of this paper is to present the design of a surface electromyography acquisition, processing and amplification system with low power consumption. Based on a micro-controller and a Bluetooth module, it must send the data to a cell phone in real time. The main topology is based on an operational amplifier and passive components in order to produce filters and an instrumentation amplifier applied to Electromyography (EMG). This paper also shows the equations used during design and describes each step of development, from simulations and testing to acquired data and microcontroller programming. In order to produce a low-cost circuit that can be later used as an acquisition tool for portable mechanisms and prosthesis, the design of the main circuit considers the lowest number of components while it does not compromise efficiency.
{"title":"Surface Electromyographic Signal Acquisition System for Real Time Monitoring of Upper Limbs Muscles","authors":"R. Ferraz, Raiff Sales da Fonseca, Igor Thonke Rodrigues, Cláudio Bastos da Silva, H. T. Filho","doi":"10.1166/JCTN.2021.9716","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9716","url":null,"abstract":"The main goal of this paper is to present the design of a surface electromyography acquisition, processing and amplification system with low power consumption. Based on a micro-controller and a Bluetooth module, it must send the data to a cell phone in real time. The main topology is\u0000 based on an operational amplifier and passive components in order to produce filters and an instrumentation amplifier applied to Electromyography (EMG). This paper also shows the equations used during design and describes each step of development, from simulations and testing to acquired data\u0000 and microcontroller programming. In order to produce a low-cost circuit that can be later used as an acquisition tool for portable mechanisms and prosthesis, the design of the main circuit considers the lowest number of components while it does not compromise efficiency.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42756888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The 5G mobile wireless network systems faces a lot of security issues due to the opening of network and its insecurity. The insecure network prone to various attacks and it disrupts secure data communications between legitimate users. Many works have addressed the security problems in 3G and 4G networks in efficient way through authentication and cryptographic techniques. But, the security in 5G networks during data communication was not improved. Subtractive Gradient Boost Clustered Node Authentication (SGBCNA) Method is introduced to perform secure data communication. The subtractive gradient boost clustering technique is applied to authenticate the mobile node as normal nodes and malicious nodes based on the selected features. The designed ensemble clustering model combines the weak learners to make final strong clustering results with minimum loss. Finally, the malicious nodes are eliminated and normal mobile nodes are taken for performing the secured communication in 5G networks. Simulation is carried out on factors such as authentication accuracy, computation overhead and security level with respect to a number of mobile nodes and data packets. The observed outcomes clearly illustrate that the SGBCNA Method efficiently improves node authentication accuracy, security level with minimum overhead than the state-of-the-art-methods.
{"title":"Subtractive Gradient Boost Clustering for Mobile Node Authentication in Internet of Things Aware 5G Networks","authors":"M. Haripriya, P. Venkadesh","doi":"10.1166/JCTN.2021.9394","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9394","url":null,"abstract":"The 5G mobile wireless network systems faces a lot of security issues due to the opening of network and its insecurity. The insecure network prone to various attacks and it disrupts secure data communications between legitimate users. Many works have addressed the security problems\u0000 in 3G and 4G networks in efficient way through authentication and cryptographic techniques. But, the security in 5G networks during data communication was not improved. Subtractive Gradient Boost Clustered Node Authentication (SGBCNA) Method is introduced to perform secure data communication.\u0000 The subtractive gradient boost clustering technique is applied to authenticate the mobile node as normal nodes and malicious nodes based on the selected features. The designed ensemble clustering model combines the weak learners to make final strong clustering results with minimum loss. Finally,\u0000 the malicious nodes are eliminated and normal mobile nodes are taken for performing the secured communication in 5G networks. Simulation is carried out on factors such as authentication accuracy, computation overhead and security level with respect to a number of mobile nodes and data packets.\u0000 The observed outcomes clearly illustrate that the SGBCNA Method efficiently improves node authentication accuracy, security level with minimum overhead than the state-of-the-art-methods.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46052184","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The task of recognizing human activities directs extensive divergence of various functions and applications. Despite analysing the intricate activity it endures demanding requirements in contemporary field of research. A subject performs a definite task at a particular time by determining the activity by using sensor data. In this research task we appraise a unique way by using data with supervised learning techniques by placing sensors on the human body by contingent upon classification process at different stages. The State-of-art machine learning approach random forests are widely discussed in terms of covering practical and theoretical aspects of body sensing. The eventual target is the superior rate of accurate predictions effecting Human Activity Recognition further effective for behavioural monitoring, medical and healthcare sectors. Classification processes are deployed for pairs of activities that are distracted often and this work attempts to analyse the essential sensors for the improved prediction. The results shows the best accuracy scores and the remaining of our findings we expose the outline, exhibiting the degree of distraction between features of ranking and human activities which renders back to sensor ranking.
{"title":"Contemporary Human Activity Recognition Based Predictions by Sensors Using Random Forest Classifier","authors":"S. Anand, S. Magesh, I. Arockiamary","doi":"10.1166/JCTN.2021.9404","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9404","url":null,"abstract":"The task of recognizing human activities directs extensive divergence of various functions and applications. Despite analysing the intricate activity it endures demanding requirements in contemporary field of research. A subject performs a definite task at a particular time by determining\u0000 the activity by using sensor data. In this research task we appraise a unique way by using data with supervised learning techniques by placing sensors on the human body by contingent upon classification process at different stages. The State-of-art machine learning approach random forests\u0000 are widely discussed in terms of covering practical and theoretical aspects of body sensing. The eventual target is the superior rate of accurate predictions effecting Human Activity Recognition further effective for behavioural monitoring, medical and healthcare sectors. Classification processes\u0000 are deployed for pairs of activities that are distracted often and this work attempts to analyse the essential sensors for the improved prediction. The results shows the best accuracy scores and the remaining of our findings we expose the outline, exhibiting the degree of distraction between\u0000 features of ranking and human activities which renders back to sensor ranking.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41956662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Manjula, R. Karthikeyan, S. Karthick, N. Logesh, M. Logeshkumar
An optimized high gain low power low noise amplifier (LNA) is presented using 90 nm CMOS process at 2.4 GHz frequency for Zigbee applications. For achieving desired design specifications, the LNA is optimized by particle swarm optimization (PSO). The PSO is successfully implemented for optimizing noise figure (NF) when satisfying all the design specifications such as gain, power dissipation, linearity and stability. PSO algorithm is developed in MATLAB to optimize the LNA parameters. The LNA with optimized parameters is simulated using Advanced Design System (ADS) Simulator. The LNA with optimized parameters produces 21.470 dB of voltage gain, 1.031 dB of noise figure at 1.02 mW power consumption with 1.2 V supply voltage. The comparison of designed LNA with and without PSO proves that the optimization improves the LNA results while satisfying all the design constraints.
{"title":"Optimized Design of Low Power Complementary Metal Oxide Semiconductor Low Noise Amplifier for Zigbee Application","authors":"S. Manjula, R. Karthikeyan, S. Karthick, N. Logesh, M. Logeshkumar","doi":"10.1166/JCTN.2021.9387","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9387","url":null,"abstract":"An optimized high gain low power low noise amplifier (LNA) is presented using 90 nm CMOS process at 2.4 GHz frequency for Zigbee applications. For achieving desired design specifications, the LNA is optimized by particle swarm optimization (PSO). The PSO is successfully implemented\u0000 for optimizing noise figure (NF) when satisfying all the design specifications such as gain, power dissipation, linearity and stability. PSO algorithm is developed in MATLAB to optimize the LNA parameters. The LNA with optimized parameters is simulated using Advanced Design System (ADS) Simulator.\u0000 The LNA with optimized parameters produces 21.470 dB of voltage gain, 1.031 dB of noise figure at 1.02 mW power consumption with 1.2 V supply voltage. The comparison of designed LNA with and without PSO proves that the optimization improves the LNA results while satisfying all the design constraints.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47726113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stream is endlessly inbound sequence of information, streamed information is unbounded and every information are often examined one time. Streamed information are often noisy and therefore the variety of clusters within the information and their applied mathematics properties will change over time, wherever random access to the information isn’t possible and storing all the arriving information is impractical. When applying data set processing techniques and specifically stream clustering Algorithms to real time information streams, limitation in execution time and memory have to be oblige to be thought-about carefully. The projected hymenopteran colony stream clustering Algorithmic is a clustering Algorithm which forms cluster according to density variation, in which clusters are separated by high density features from low density feature region with mounted movement of hymenopteran. Result shows that it created denser cluster than antecedently projected Algorithmic program. And with mounted movement of ants conjointly it decreases the loss of data points. And conjointly the changed radius formula of cluster is projected so as to increase performance of model to create it a lot of dynamic with continuous flow of information. And also we changed probability formula for pick up and drop to reduce oulier. Results from hymenopteran experiments conjointly showed that sorting is disbursed in 2 phases, a primary clustering episode followed by a spacing part. In this paper, we have also compared proposed Algorithm with particle swarm optimization and genetic optimization using DBSCAN and k -means clustering.
{"title":"Hymenopteran Colony Stream Clustering Algorithm and Comparison with Particle Swarm Optimization and Genetic Optimization Clustering","authors":"Nikhil Parafe, M. Venkatesan, Prabhavathy Panner","doi":"10.1166/JCTN.2021.9402","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9402","url":null,"abstract":"Stream is endlessly inbound sequence of information, streamed information is unbounded and every information are often examined one time. Streamed information are often noisy and therefore the variety of clusters within the information and their applied mathematics properties will change\u0000 over time, wherever random access to the information isn’t possible and storing all the arriving information is impractical. When applying data set processing techniques and specifically stream clustering Algorithms to real time information streams, limitation in execution time and memory\u0000 have to be oblige to be thought-about carefully. The projected hymenopteran colony stream clustering Algorithmic is a clustering Algorithm which forms cluster according to density variation, in which clusters are separated by high density features from low density feature region with mounted\u0000 movement of hymenopteran. Result shows that it created denser cluster than antecedently projected Algorithmic program. And with mounted movement of ants conjointly it decreases the loss of data points. And conjointly the changed radius formula of cluster is projected so as to increase performance\u0000 of model to create it a lot of dynamic with continuous flow of information. And also we changed probability formula for pick up and drop to reduce oulier. Results from hymenopteran experiments conjointly showed that sorting is disbursed in 2 phases, a primary clustering episode followed by\u0000 a spacing part. In this paper, we have also compared proposed Algorithm with particle swarm optimization and genetic optimization using DBSCAN and k -means clustering.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49421103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Text clustering has now a days become a very major technique in many fields including data mining, Natural Language Processing etc. It’s also broadly used for information retrieval and assimilation of textual data. Majority of the works which were carried out previously focuses on the clustering algorithms where feature extraction is done without considering the semantic meaning of word based on its context. In the given work, we introduce a double clustering algorithm using K -Means, by using in conjuction, a Bi-directional Long Short-Term Memory and a Convolutional Neural Network for the purpose of feature extraction, so that the semantic meaning is also considered. Recurrent neural network (RNN) has the ability to study long-term dependencies prevailing in input whereas CNN models are for long known to be effective in feature extraction of local features of given input data. Unlike all the works previously carried out, this proposed work considers and carries out extraction of features and clustering of documents as one combined mechanism. Here result of clustering is send back to the model as feedback information thereby optimizing the parameters of the network model dynamically. Clustering in a double-clustering manner is implemented, which increases the time efficiency.
{"title":"Double Clustering Based Neural Feedback Method for Unstructured Text Data","authors":"S. Sarannya, M. Venkatesan, Prabhavathy Panner","doi":"10.1166/JCTN.2021.9385","DOIUrl":"https://doi.org/10.1166/JCTN.2021.9385","url":null,"abstract":"Text clustering has now a days become a very major technique in many fields including data mining, Natural Language Processing etc. It’s also broadly used for information retrieval and assimilation of textual data. Majority of the works which were carried out previously focuses\u0000 on the clustering algorithms where feature extraction is done without considering the semantic meaning of word based on its context. In the given work, we introduce a double clustering algorithm using K -Means, by using in conjuction, a Bi-directional Long Short-Term Memory and a Convolutional\u0000 Neural Network for the purpose of feature extraction, so that the semantic meaning is also considered. Recurrent neural network (RNN) has the ability to study long-term dependencies prevailing in input whereas CNN models are for long known to be effective in feature extraction of local features\u0000 of given input data. Unlike all the works previously carried out, this proposed work considers and carries out extraction of features and clustering of documents as one combined mechanism. Here result of clustering is send back to the model as feedback information thereby optimizing the parameters\u0000 of the network model dynamically. Clustering in a double-clustering manner is implemented, which increases the time efficiency.","PeriodicalId":15416,"journal":{"name":"Journal of Computational and Theoretical Nanoscience","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41550733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：