Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776765
Joshua S. Gulmatico, Julie Ann B. Susa, M. A. Malbog, Aimee G. Acoba, Marte D. Nipas, Jennalyn N. Mindoro
Music consumption patterns could alter due to digitization, and music popularity was redefined in the streaming era. The number of people using Spotify is constantly growing. It has risen to become one of the most popular internet music providers in recent years. People have been listening to my favorite performers and receiving new song recommendations via the Spotify app for the past year. The research looks at the relationship between song data – audio attributes from the Spotify database (for example, key and tempo) – and song popularity, as measured by the number of Spotify streams a song has. To develop a high accuracy model for predicting hit songs, the researcher investigates four machine learning algorithms (MLAs): Linear Regression, Random Forest Classifier, and K-means Clustering. This study presents a prediction model for determining whether a piece of music is popular in the mainstream and using machine learning to classify songs based on their popularity.
{"title":"SpotiPred: A Machine Learning Approach Prediction of Spotify Music Popularity by Audio Features","authors":"Joshua S. Gulmatico, Julie Ann B. Susa, M. A. Malbog, Aimee G. Acoba, Marte D. Nipas, Jennalyn N. Mindoro","doi":"10.1109/ICPC2T53885.2022.9776765","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776765","url":null,"abstract":"Music consumption patterns could alter due to digitization, and music popularity was redefined in the streaming era. The number of people using Spotify is constantly growing. It has risen to become one of the most popular internet music providers in recent years. People have been listening to my favorite performers and receiving new song recommendations via the Spotify app for the past year. The research looks at the relationship between song data – audio attributes from the Spotify database (for example, key and tempo) – and song popularity, as measured by the number of Spotify streams a song has. To develop a high accuracy model for predicting hit songs, the researcher investigates four machine learning algorithms (MLAs): Linear Regression, Random Forest Classifier, and K-means Clustering. This study presents a prediction model for determining whether a piece of music is popular in the mainstream and using machine learning to classify songs based on their popularity.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127862648","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776676
Rohan Gawhade, Lokesh Ramdev Bohara, Jesvin Mathew, Poonam Bari
Machine Learning (ML) has seen a sudden exponential rise in past decades. Numerous resources and documentation allow people to become ML practitioners. Companies make huge profits out of the analysis and predictions they make. ML Engineers are highly paid for their knowledge in this domain. It has become prevalent and much more comprehensible. One best out of the important stages in ML is Data preprocessing, and feature extraction. In Data Preprocessing itself, there are various tasks one needs to perform accurately to make the data provided. From handling missing values to encoding and normalization, each step has its importance and hence a professional must be adept with each of these steps. Data Preprocessing steps depend upon the type of data provided i.e. categorical data, continuous data, an array of images' pixels or even images themselves. With the requirement to deal with all the cleaning steps, it becomes quite strenuous to learn and become an expert. Moreover, it is time-consuming and does not guarantee expected results. Hence, there is a need to handle this issue. We aim to automate this complete process to ease the work of Machine Learning Engineers and make it more productive. Any user will only have to provide the dataset and does not have to manually select the processing techniques as provided by the latest Data Mining tools. The application will observe the dataset and apply the suitable techniques on its own. Since all the steps will be automated and the user will only have to provide the dataset, even the people who are not familiar with concepts of Machine Learning can pre-process the dataset. This allows the opening of opportunities for people from various domains who desire to perform Machine Learning operations.
{"title":"Computerized Data-Preprocessing To Improve Data Quality","authors":"Rohan Gawhade, Lokesh Ramdev Bohara, Jesvin Mathew, Poonam Bari","doi":"10.1109/ICPC2T53885.2022.9776676","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776676","url":null,"abstract":"Machine Learning (ML) has seen a sudden exponential rise in past decades. Numerous resources and documentation allow people to become ML practitioners. Companies make huge profits out of the analysis and predictions they make. ML Engineers are highly paid for their knowledge in this domain. It has become prevalent and much more comprehensible. One best out of the important stages in ML is Data preprocessing, and feature extraction. In Data Preprocessing itself, there are various tasks one needs to perform accurately to make the data provided. From handling missing values to encoding and normalization, each step has its importance and hence a professional must be adept with each of these steps. Data Preprocessing steps depend upon the type of data provided i.e. categorical data, continuous data, an array of images' pixels or even images themselves. With the requirement to deal with all the cleaning steps, it becomes quite strenuous to learn and become an expert. Moreover, it is time-consuming and does not guarantee expected results. Hence, there is a need to handle this issue. We aim to automate this complete process to ease the work of Machine Learning Engineers and make it more productive. Any user will only have to provide the dataset and does not have to manually select the processing techniques as provided by the latest Data Mining tools. The application will observe the dataset and apply the suitable techniques on its own. Since all the steps will be automated and the user will only have to provide the dataset, even the people who are not familiar with concepts of Machine Learning can pre-process the dataset. This allows the opening of opportunities for people from various domains who desire to perform Machine Learning operations.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121939108","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776880
Sheikh Suhail Mohammad, Sheikh Javed Iqbal
Electric vehicles are currently acting as a replacement for fossil fuel-based vehicles. Electric vehicles are environment friendly, and energy efficient. However, electric vehicles demand research attention to improve system modelling, design, reliability, stability and control issues. Power-sharing is critical for electric vehicles reliable and economical operation; hence, they need to improve the power-sharing techniques and algorithms. A supercapacitor assisted fuel cell-based micro-power system is proposed and studied in this work. A power-sharing technique is proposed to control the power flow between fuel cell and supercapacitor during different vehicle operating modes to improve system reliability, stability, and vehicle dynamics. Supercapacitor state of the charge & voltage, fuel cell response time and motor power demand are critical variables for power-sharing and decision making. The design details give information about the system component types their advantages and disadvantages. An extended discussion is carried out that explains how the motors power rating is selected subjected to road dynamics. Time-domain simulations are performed in MATLAB/Simulink that validate the effectiveness of the proposed power-sharing and control technique during different operating modes.
{"title":"Power Management, Control and Design of Supercapacitor Assisted Fuel Cell-based Micro Power System for Electric Vehicles","authors":"Sheikh Suhail Mohammad, Sheikh Javed Iqbal","doi":"10.1109/ICPC2T53885.2022.9776880","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776880","url":null,"abstract":"Electric vehicles are currently acting as a replacement for fossil fuel-based vehicles. Electric vehicles are environment friendly, and energy efficient. However, electric vehicles demand research attention to improve system modelling, design, reliability, stability and control issues. Power-sharing is critical for electric vehicles reliable and economical operation; hence, they need to improve the power-sharing techniques and algorithms. A supercapacitor assisted fuel cell-based micro-power system is proposed and studied in this work. A power-sharing technique is proposed to control the power flow between fuel cell and supercapacitor during different vehicle operating modes to improve system reliability, stability, and vehicle dynamics. Supercapacitor state of the charge & voltage, fuel cell response time and motor power demand are critical variables for power-sharing and decision making. The design details give information about the system component types their advantages and disadvantages. An extended discussion is carried out that explains how the motors power rating is selected subjected to road dynamics. Time-domain simulations are performed in MATLAB/Simulink that validate the effectiveness of the proposed power-sharing and control technique during different operating modes.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129994304","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776979
Soumya Shastry, P. Dewangan
This paper discusses about order reduction of linear time invariant (LTI) systems based on error minimization by Genetic algorithm. The coefficients of the state space modelmatrices of reduced dimensions are obtained by the proposed method. The reduction procedure is simple and computer oriented. An example system is considered to show the efficacy of the proposed method. The step responses of the higher order system and its models, and validation parameters areused for performance comparison. The results obtained confirm the superiority of the proposed technique.
{"title":"Order Reduction of Linear Time Invariant Systems Using Genetic Algorithm","authors":"Soumya Shastry, P. Dewangan","doi":"10.1109/ICPC2T53885.2022.9776979","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776979","url":null,"abstract":"This paper discusses about order reduction of linear time invariant (LTI) systems based on error minimization by Genetic algorithm. The coefficients of the state space modelmatrices of reduced dimensions are obtained by the proposed method. The reduction procedure is simple and computer oriented. An example system is considered to show the efficacy of the proposed method. The step responses of the higher order system and its models, and validation parameters areused for performance comparison. The results obtained confirm the superiority of the proposed technique.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"95 10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130565071","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776894
Shreya Chakravarty, Shardul Fating, Ishita Jain, Ishika Varun, R. Khandelwal
The all-embracing use of digital images has revamped the quality of life and security to a great extent. Right from finding an item on online shopping websites through a clicked picture, to CCTV cameras being used for road traffic control, the users have learnt to appreciate the existence of technology being as advanced. However, one cannot overlook the gravity of this technology being misused. Although, the digitization has incorporated advanced concepts like Computer Vision and Deep Learning for security-check and crowd control, this has encouraged the advancement of courtroom discussions. Framing people for wrongdoings they are not involved with, on the basis of a fake “digital proof,” is one of the newly faced muddles. False allegations on a person, on the basis of a picture or a video, can potentially put a question on the existence of a person. The need to find the legitimacy of a produced image is therefore, of utmost importance. There have been various studies over the years, wherein a lot of methods were proposed to develop a system that identifies the camera model. Through this paper, we aim to produce a comparative study between four leading architectures, DenseNet, Inception V3, MobileNetV2 and Exception(XCeption), and suggest a the most competent architecture for commercialization of this system.
{"title":"Comparative Study between Leading Transfer Learning Architectures for Source Camera Identification","authors":"Shreya Chakravarty, Shardul Fating, Ishita Jain, Ishika Varun, R. Khandelwal","doi":"10.1109/ICPC2T53885.2022.9776894","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776894","url":null,"abstract":"The all-embracing use of digital images has revamped the quality of life and security to a great extent. Right from finding an item on online shopping websites through a clicked picture, to CCTV cameras being used for road traffic control, the users have learnt to appreciate the existence of technology being as advanced. However, one cannot overlook the gravity of this technology being misused. Although, the digitization has incorporated advanced concepts like Computer Vision and Deep Learning for security-check and crowd control, this has encouraged the advancement of courtroom discussions. Framing people for wrongdoings they are not involved with, on the basis of a fake “digital proof,” is one of the newly faced muddles. False allegations on a person, on the basis of a picture or a video, can potentially put a question on the existence of a person. The need to find the legitimacy of a produced image is therefore, of utmost importance. There have been various studies over the years, wherein a lot of methods were proposed to develop a system that identifies the camera model. Through this paper, we aim to produce a comparative study between four leading architectures, DenseNet, Inception V3, MobileNetV2 and Exception(XCeption), and suggest a the most competent architecture for commercialization of this system.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123106977","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776982
Vijay Mohale, T. Chelliah
The extra high voltage transmission line 765k V, connected to a doubly-fed induction machine (DFIM) for variable speed pumped storage plant prone to sub-synchronous oscillation (SSO). Hence, the cost-effective and most popular approach to increase the power transfer ability in long transmission lines is to use series capacitive compensation. However, SSO is a significant problem that can cause electrical instability and generator shaft failure. The main motive of this paper is to investigate sub-synchronous oscillations caused by series compensation in a transmission line connected to a DFIM. The simulation is carried out and results are validated in MATLAB/Simulink to analyze SSO in case of different series compensation levels of 30%, 50%, and 90%. The experimental validation is obtained in the laboratory on scale down model of Tehri (PSPP to be commissioned) to Meerut EHV transmission line.
{"title":"Impact of Series Compensated High Voltage Transmission Lines in the Operation of DFIM Based Hydro Unit","authors":"Vijay Mohale, T. Chelliah","doi":"10.1109/ICPC2T53885.2022.9776982","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776982","url":null,"abstract":"The extra high voltage transmission line 765k V, connected to a doubly-fed induction machine (DFIM) for variable speed pumped storage plant prone to sub-synchronous oscillation (SSO). Hence, the cost-effective and most popular approach to increase the power transfer ability in long transmission lines is to use series capacitive compensation. However, SSO is a significant problem that can cause electrical instability and generator shaft failure. The main motive of this paper is to investigate sub-synchronous oscillations caused by series compensation in a transmission line connected to a DFIM. The simulation is carried out and results are validated in MATLAB/Simulink to analyze SSO in case of different series compensation levels of 30%, 50%, and 90%. The experimental validation is obtained in the laboratory on scale down model of Tehri (PSPP to be commissioned) to Meerut EHV transmission line.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"315 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116596136","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776999
S. V. Krishna
Grid connected PV solar generates DC power utilizing the solar energy as input and the generated DC power is converted to AC power using DC-DC converter and DC-AC Inverter. The output AC current from inverter consists of harmonic currents along with fundamental current. This current is fed to Grid through Inverter Duty Transformer (IDT) and Power Transformer (PT). The harmonics generated by solar plant are measured at LV & HV of IDT by using a harmonic analyser at site and presented in this paper. Apart from it, simulation model of PV solar generation with DC-DC converter & Inverter is also developed in PSCAD and the generated harmonics are compared with the measured harmonics and presented in this paper.
{"title":"Power Quality Analysis of Solar Plant by Measurement at Site and Through Simulation in PSCAD","authors":"S. V. Krishna","doi":"10.1109/ICPC2T53885.2022.9776999","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776999","url":null,"abstract":"Grid connected PV solar generates DC power utilizing the solar energy as input and the generated DC power is converted to AC power using DC-DC converter and DC-AC Inverter. The output AC current from inverter consists of harmonic currents along with fundamental current. This current is fed to Grid through Inverter Duty Transformer (IDT) and Power Transformer (PT). The harmonics generated by solar plant are measured at LV & HV of IDT by using a harmonic analyser at site and presented in this paper. Apart from it, simulation model of PV solar generation with DC-DC converter & Inverter is also developed in PSCAD and the generated harmonics are compared with the measured harmonics and presented in this paper.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134137109","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776784
D. Singh, N. Verma
This paper provides an approach for design of Fuzzy Model Based (FMB) control system for generic aircraft and UAV related application. The FMB control is an evolving nonlinear control strategy which consists of a fuzzy model and a fuzzy controller connected in a closed-loop. In proposed work an application-oriented research, in which an emerging soft computing-based technique (fuzzy system) is applied for design of flight control system of Generic aircraft. The existing theoretical base developed in fuzzy systems literature is explored/customized for aircraft related application. The short period mode of longitudinal aircraft dynamics is considered for simulation and demonstration purpose. The fuzzy model of longitudinal aircraft dynamics are obtained from nonlinear dynamics equations about various representative points (equilibrium points) of flight envelop with some fuzzy rules. The aircraft flight envelope parameters i.e operating altitude and Mach Number are characterized as premise parameters and elements of stability and control derivative matrix are identified as consequent parameters of fuzzy model. The decay rate fuzzy controller with constraint on state and control input parameters is considered and its feedback gains are obtained by solving the LMI stability conditions. The closed-loop response of FMB controller is presented at three initial flight conditions. The simulation result reveals that proposed FMB controller is well suited at various identified operating points of the flight envelop. It not only stabilizes the aircraft dynamics but also provides improved transient performance. This demonstrates the utility of FMB control system for aircraft / UAVs related application.
{"title":"Design of Fuzzy Control System for Generic Aircraft/UAVs","authors":"D. Singh, N. Verma","doi":"10.1109/ICPC2T53885.2022.9776784","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776784","url":null,"abstract":"This paper provides an approach for design of Fuzzy Model Based (FMB) control system for generic aircraft and UAV related application. The FMB control is an evolving nonlinear control strategy which consists of a fuzzy model and a fuzzy controller connected in a closed-loop. In proposed work an application-oriented research, in which an emerging soft computing-based technique (fuzzy system) is applied for design of flight control system of Generic aircraft. The existing theoretical base developed in fuzzy systems literature is explored/customized for aircraft related application. The short period mode of longitudinal aircraft dynamics is considered for simulation and demonstration purpose. The fuzzy model of longitudinal aircraft dynamics are obtained from nonlinear dynamics equations about various representative points (equilibrium points) of flight envelop with some fuzzy rules. The aircraft flight envelope parameters i.e operating altitude and Mach Number are characterized as premise parameters and elements of stability and control derivative matrix are identified as consequent parameters of fuzzy model. The decay rate fuzzy controller with constraint on state and control input parameters is considered and its feedback gains are obtained by solving the LMI stability conditions. The closed-loop response of FMB controller is presented at three initial flight conditions. The simulation result reveals that proposed FMB controller is well suited at various identified operating points of the flight envelop. It not only stabilizes the aircraft dynamics but also provides improved transient performance. This demonstrates the utility of FMB control system for aircraft / UAVs related application.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115557735","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776987
Franklin M. Miranda, Arnel C. Fajardo, Ruji P. Medina
Many countries have been facing problems concerning Pulmonary Tuberculosis. These illnesses have a great toll on human lives whether young or old. Medicines and health awareness have driven Health and Medical institutions to embrace the advancement in the medical field. It aimed to maximize the ability of Medical science to combat this illness. With medical and computer science combined led Artificial Intelligence and Neural networks to produce a drastic product in the early detection of Pulmonary Tuberculosis. Moreover, Image processing of captured photographs of Chest X-ray results was processed using a technique. The Contrast Low Adaptive Histogram Equalization and Grab Cut was used to concentrate on the region of interest for lungs. The Radial basis function was utilized as the network model and part of the program is to use Scikit learn in determining the confusion matrix, precision, recall, f1-score, and support. The concept was the first step to provide medical “diagnosis”, especially in low and hard-up far-flung communities that were rarely visited by a specialist for Chest X-ray interpretation.
{"title":"Pulmonary Tuberculosis Detection using Digitally Photographed Chest X-RAY Images","authors":"Franklin M. Miranda, Arnel C. Fajardo, Ruji P. Medina","doi":"10.1109/ICPC2T53885.2022.9776987","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776987","url":null,"abstract":"Many countries have been facing problems concerning Pulmonary Tuberculosis. These illnesses have a great toll on human lives whether young or old. Medicines and health awareness have driven Health and Medical institutions to embrace the advancement in the medical field. It aimed to maximize the ability of Medical science to combat this illness. With medical and computer science combined led Artificial Intelligence and Neural networks to produce a drastic product in the early detection of Pulmonary Tuberculosis. Moreover, Image processing of captured photographs of Chest X-ray results was processed using a technique. The Contrast Low Adaptive Histogram Equalization and Grab Cut was used to concentrate on the region of interest for lungs. The Radial basis function was utilized as the network model and part of the program is to use Scikit learn in determining the confusion matrix, precision, recall, f1-score, and support. The concept was the first step to provide medical “diagnosis”, especially in low and hard-up far-flung communities that were rarely visited by a specialist for Chest X-ray interpretation.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116166271","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}
Pub Date : 2022-03-01DOI: 10.1109/ICPC2T53885.2022.9776918
Samiran Das, Chirag Kyal, S. Pratiher
Traditional clustering methods groups data points according to attributes such as similarity, continuity, neighbor-hood information, etc. overlooks the structural properties of the data. Consequently, prevalent clustering approaches to below-par performance in real-world applications. Unlike traditional clustering approaches, subspace clustering methods attempt to group datapoints keeping the inherent structure and rank-related properties of the data into account. Despite the rapid growth in deep learning-based approaches, very few works have utilized deep learning for the subspace clustering task. This work introduced an auto-encoder-based deep learning architecture consisting of a self-expressive layer for the deep subspace clustering task. We use smoothed L2, L0.5 and Frobenius norms instead of the actual measures for ease of optimization task. We also explored the efficacy of sparsity measures that characterize the self-representation coefficient matrix of the self-expressive layer. The experiments conducted on standard datasets suggest that the application of efficient sparsity measures improves the performance of the subspace clustering approach and results in superior performance compared to the previous deep subspace clustering approaches.
{"title":"On Sparsity Measures In Deep Subspace Clustering","authors":"Samiran Das, Chirag Kyal, S. Pratiher","doi":"10.1109/ICPC2T53885.2022.9776918","DOIUrl":"https://doi.org/10.1109/ICPC2T53885.2022.9776918","url":null,"abstract":"Traditional clustering methods groups data points according to attributes such as similarity, continuity, neighbor-hood information, etc. overlooks the structural properties of the data. Consequently, prevalent clustering approaches to below-par performance in real-world applications. Unlike traditional clustering approaches, subspace clustering methods attempt to group datapoints keeping the inherent structure and rank-related properties of the data into account. Despite the rapid growth in deep learning-based approaches, very few works have utilized deep learning for the subspace clustering task. This work introduced an auto-encoder-based deep learning architecture consisting of a self-expressive layer for the deep subspace clustering task. We use smoothed L2, L0.5 and Frobenius norms instead of the actual measures for ease of optimization task. We also explored the efficacy of sparsity measures that characterize the self-representation coefficient matrix of the self-expressive layer. The experiments conducted on standard datasets suggest that the application of efficient sparsity measures improves the performance of the subspace clustering approach and results in superior performance compared to the previous deep subspace clustering approaches.","PeriodicalId":283298,"journal":{"name":"2022 Second International Conference on Power, Control and Computing Technologies (ICPC2T)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115384107","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：