Pub Date : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758351
G. Vishal, J. Pradeep
Humans are moving towards a pollution-free environment, Electrical vehicles (EV) could help to achieve this since one of the major contributors to pollution is Conventional vehicles. Increasing the performance of EV's will promote the use of EVs in human civilization. For any electrical machine, performance depends on Time Domain parameters. By optimizing the time domain parameter, the performance increases drastically. With a simple optimized PID controller, the motor could achieve performance similar to other controllers like the Fuzzy logic system. In many papers, PID is tuned using Particle swarm optimization (PSO). Recently, a new biological metaheuristic technique is determined that is Tunicate swarm Algorithm (TSA). This method is better than many biological metaheuristic techniques. In this paper, the TSA is implemented to the PID controller for the Permanent Magnet Synchronous Motor (PMSM) operation thereby improving the Speed response and comparing with the existing PSO and conventional PID controller.
{"title":"Improved Performance of PMSM using Tunicate Swarm optimization","authors":"G. Vishal, J. Pradeep","doi":"10.1109/ESCI53509.2022.9758351","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758351","url":null,"abstract":"Humans are moving towards a pollution-free environment, Electrical vehicles (EV) could help to achieve this since one of the major contributors to pollution is Conventional vehicles. Increasing the performance of EV's will promote the use of EVs in human civilization. For any electrical machine, performance depends on Time Domain parameters. By optimizing the time domain parameter, the performance increases drastically. With a simple optimized PID controller, the motor could achieve performance similar to other controllers like the Fuzzy logic system. In many papers, PID is tuned using Particle swarm optimization (PSO). Recently, a new biological metaheuristic technique is determined that is Tunicate swarm Algorithm (TSA). This method is better than many biological metaheuristic techniques. In this paper, the TSA is implemented to the PID controller for the Permanent Magnet Synchronous Motor (PMSM) operation thereby improving the Speed response and comparing with the existing PSO and conventional PID controller.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127231543","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758154
Prajwal S. Gaikwad, Shabnam Sayyad
OTT television and film material is a method of delivering television and cinema content on the web at the request of and in accordance with the preferences of the individual user. The word “over-the-top” is an abbreviation for “over-the-top,” which signifies that a content provider is providing services on top of already existing internet services. During the pandemic period, the requirement for this infrastructure has increased by orders of magnitude. In India the two upcoming players Amazon Prime Video and Netflix have become the prior choice as compared to the daily soaps and the movie budgets. This research study focuses on the prime factors as the strategies of these two competitors to build up the two-person game theory model. A survey is conducted among the users of these two players as to find the trends of the identified factors. The regression models Y on X and X on Y are used to find the values of payoff matrix and then Game theory model is solved to find out the optimum strategies. The study is expected to benefit the various competitors of the OTT domain to build their strategies.
{"title":"A Game Theory Model for Optimization of the OTT Platform Strategies","authors":"Prajwal S. Gaikwad, Shabnam Sayyad","doi":"10.1109/ESCI53509.2022.9758154","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758154","url":null,"abstract":"OTT television and film material is a method of delivering television and cinema content on the web at the request of and in accordance with the preferences of the individual user. The word “over-the-top” is an abbreviation for “over-the-top,” which signifies that a content provider is providing services on top of already existing internet services. During the pandemic period, the requirement for this infrastructure has increased by orders of magnitude. In India the two upcoming players Amazon Prime Video and Netflix have become the prior choice as compared to the daily soaps and the movie budgets. This research study focuses on the prime factors as the strategies of these two competitors to build up the two-person game theory model. A survey is conducted among the users of these two players as to find the trends of the identified factors. The regression models Y on X and X on Y are used to find the values of payoff matrix and then Game theory model is solved to find out the optimum strategies. The study is expected to benefit the various competitors of the OTT domain to build their strategies.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123709721","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758222
S. Bakre, A. Shiralkar, S. Shelar, Suchita Ingle
The theft of electricity is a matter of concern for the distribution utility today. The Aggregate Technical and Commercial (AT&C) loss of Maharashtra State Electricity Distribution Company is around 20.72% for the year 2020–21. The main cause of such a higher loss is pilferage or theft of electricity. As per statistics given by various distribution utilities, the theft incidences of three phase HT and LT consumers are under control. However, there is a rising trend in tampering of single phase meters. Various methods of theft detection of single phase meters are in existence, however, tampering of meter by inserting the resistive link in parallel with the meter cannot be detected using these conventional methods. In this paper, a novice technique of tamper detection using Artificial Neural Network is proposed. The proposed method is cost effective and feasible.
{"title":"Artificial Neural Network Based Electricity Theft Detection","authors":"S. Bakre, A. Shiralkar, S. Shelar, Suchita Ingle","doi":"10.1109/ESCI53509.2022.9758222","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758222","url":null,"abstract":"The theft of electricity is a matter of concern for the distribution utility today. The Aggregate Technical and Commercial (AT&C) loss of Maharashtra State Electricity Distribution Company is around 20.72% for the year 2020–21. The main cause of such a higher loss is pilferage or theft of electricity. As per statistics given by various distribution utilities, the theft incidences of three phase HT and LT consumers are under control. However, there is a rising trend in tampering of single phase meters. Various methods of theft detection of single phase meters are in existence, however, tampering of meter by inserting the resistive link in parallel with the meter cannot be detected using these conventional methods. In this paper, a novice technique of tamper detection using Artificial Neural Network is proposed. The proposed method is cost effective and feasible.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122189009","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758290
Salim Shamsher, Manikandan Thirumalaisamy, P. Tyagi, Deepa Muthiah, Nakirekanti Suvarna
The Internet of Things (IoT) is now growing dramatically on various levels and helps to digitize various vital industries quickly. The most difficult obstacle for BCIs to overcome is the fact that not everyone has the same brain. Every new session requires the BCI to learn from the user's brain, which is accomplished via the use of Machine Learning. However, this learning process is time-consuming. Calibration time refers to the amount of time it takes for the BCI to adapt to the user's brain in order to properly categorise their thoughts and determine their meaning. The patient has had to wait an arduous and tiresome length of time for the system to be completely functioning up until now because of this calibration, which may take up to 20 - 30 minutes. The aim of this thesis was to find a way to decrease the amount of time required for calibration to the smallest amount feasible. In the first section of this paper, a first effort is made to determine the optimum number of features required for the BCI to operate reasonably, taking into consideration all of the calibration data provided. When the results were averaged across five participants, the percentage of properly identified thoughts was just 67.15 percent. Transfer learning was used in order to improve the performance of the BCI while simultaneously decreasing the calibration time. It is feasible to decrease the amount of calibration required for the categorization of thoughts coming from a new target subject by using knowledge collected from previously recorded subjects to the greatest extent possible in Transfer Learning. It was determined that existing methods were superior, and a new methodology was created that required just 24 seconds of calibration data while accurately identifying 86.8% of the thoughts. In order to alleviate mental stress and anger, the system suggested fits effectively with a deep learning network. This paper proposes a brain learning framework that uses a neural network model that is complex in nature and uses IoT for data collection from various wearable devices and the same can be used for modelling the brain functions.
{"title":"Detection of Epileptic Seizure using Improved Adaptive Neuro Fuzzy Inference System with Machine Learning Techniques","authors":"Salim Shamsher, Manikandan Thirumalaisamy, P. Tyagi, Deepa Muthiah, Nakirekanti Suvarna","doi":"10.1109/ESCI53509.2022.9758290","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758290","url":null,"abstract":"The Internet of Things (IoT) is now growing dramatically on various levels and helps to digitize various vital industries quickly. The most difficult obstacle for BCIs to overcome is the fact that not everyone has the same brain. Every new session requires the BCI to learn from the user's brain, which is accomplished via the use of Machine Learning. However, this learning process is time-consuming. Calibration time refers to the amount of time it takes for the BCI to adapt to the user's brain in order to properly categorise their thoughts and determine their meaning. The patient has had to wait an arduous and tiresome length of time for the system to be completely functioning up until now because of this calibration, which may take up to 20 - 30 minutes. The aim of this thesis was to find a way to decrease the amount of time required for calibration to the smallest amount feasible. In the first section of this paper, a first effort is made to determine the optimum number of features required for the BCI to operate reasonably, taking into consideration all of the calibration data provided. When the results were averaged across five participants, the percentage of properly identified thoughts was just 67.15 percent. Transfer learning was used in order to improve the performance of the BCI while simultaneously decreasing the calibration time. It is feasible to decrease the amount of calibration required for the categorization of thoughts coming from a new target subject by using knowledge collected from previously recorded subjects to the greatest extent possible in Transfer Learning. It was determined that existing methods were superior, and a new methodology was created that required just 24 seconds of calibration data while accurately identifying 86.8% of the thoughts. In order to alleviate mental stress and anger, the system suggested fits effectively with a deep learning network. This paper proposes a brain learning framework that uses a neural network model that is complex in nature and uses IoT for data collection from various wearable devices and the same can be used for modelling the brain functions.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121680261","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758339
Suneel Kumar Rath, M. Sahu, S. P. Das, S. Mohapatra
The primary purpose of the software industry is to provide high-quality software. Software system failure is caused by faulty software components. The goal of reliable software is to reduce the amount of software programme failures. Software defect prediction is a crucial aspect of developing high-quality software. One can predict software failures by implement essential prediction metrics and previous fault information. A good software fault prediction model makes testing easier while also improving the quality and consistency of software. For defect prediction systems based on diverse parameters, several methodologies have been proposed. However, none of the models meet the criteria for software reliability defect prediction. So in this article we proposed a hybrid software reliability model using feature selection and support vector classifier. In terms of software reliability defect prediction, the provided methodology is acceptable for different software metrics with experimental approvals utilizing a standard dataset. In the methodology, the NASA Metrics Data Program datasets are used for real-time verification and validation.
{"title":"Hybrid Software Reliability Prediction Model Using Feature Selection and Support Vector Classifier","authors":"Suneel Kumar Rath, M. Sahu, S. P. Das, S. Mohapatra","doi":"10.1109/ESCI53509.2022.9758339","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758339","url":null,"abstract":"The primary purpose of the software industry is to provide high-quality software. Software system failure is caused by faulty software components. The goal of reliable software is to reduce the amount of software programme failures. Software defect prediction is a crucial aspect of developing high-quality software. One can predict software failures by implement essential prediction metrics and previous fault information. A good software fault prediction model makes testing easier while also improving the quality and consistency of software. For defect prediction systems based on diverse parameters, several methodologies have been proposed. However, none of the models meet the criteria for software reliability defect prediction. So in this article we proposed a hybrid software reliability model using feature selection and support vector classifier. In terms of software reliability defect prediction, the provided methodology is acceptable for different software metrics with experimental approvals utilizing a standard dataset. In the methodology, the NASA Metrics Data Program datasets are used for real-time verification and validation.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131443418","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758381
Rahul B. Diwate, Atharva Zagade, M. Khodaskar, Varsha R. Dange
Object Detection is one of the important entities in the field of Computer Vision with a large number of applications. This project demonstrates Object detection using You Only Look Once (YOLO) Algorithm, version 3. YOLOv3 method is prominently used in object detection methods which are based on Deep Learning. It uses k-means cluster method for creating bounding boxes of specific height and width, which are used for predicting output. The model training is based on the Common Object in Context (COCO) Dataset. The dataset has around 164K images based on 80 categories, also called as classes. Thus, this object detection model takes an image from the user and then with the help of YOLO algorithm, predicts the types of objects present in that image and marks them accurately., the lower complex CNN model achieves an accuracy of 0.93.
{"title":"Optimization in Object Detection Model using YOLO.v3","authors":"Rahul B. Diwate, Atharva Zagade, M. Khodaskar, Varsha R. Dange","doi":"10.1109/ESCI53509.2022.9758381","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758381","url":null,"abstract":"Object Detection is one of the important entities in the field of Computer Vision with a large number of applications. This project demonstrates Object detection using You Only Look Once (YOLO) Algorithm, version 3. YOLOv3 method is prominently used in object detection methods which are based on Deep Learning. It uses k-means cluster method for creating bounding boxes of specific height and width, which are used for predicting output. The model training is based on the Common Object in Context (COCO) Dataset. The dataset has around 164K images based on 80 categories, also called as classes. Thus, this object detection model takes an image from the user and then with the help of YOLO algorithm, predicts the types of objects present in that image and marks them accurately., the lower complex CNN model achieves an accuracy of 0.93.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116719401","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758188
Harsh Sharma, Pooja Singh, Ayush Bhardwaj
In the contemporary world, the early detection of any disease has become imperative. With an accelerating rate of population, the chance of fatality by breast cancer is growing exponentially. A reliable and effective detection system helps the medical personnel in fast detection of cancer. In the course of the present study, we have presented a comparative analysis of recent state-of the-art machine learning techniques that are being extensively used in cancer detection especially Breast Cancer by using the breast cancer dataset named Wisconsin dataset. We have statistically and comparatively scrutinized and compared the machine learning techniques that are used in classification like Naïve Bayes (NB), K-Nearest Neighbor (KNN), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), XGboost (XG) and Decision Tree (DT) for computing the accuracy in the light of performance metrics like recall, precision F1 score and accuracy percentage. Moreover, these classification techniques were also projected on ROC Curve. As a result, this research paper evaluates that the accuracy obtained by XGboost is 98.24% whereas in SVM the accuracy is 96.49%.
{"title":"Breast Cancer Detection: Comparative Analysis of Machine Learning Classification Techniques","authors":"Harsh Sharma, Pooja Singh, Ayush Bhardwaj","doi":"10.1109/ESCI53509.2022.9758188","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758188","url":null,"abstract":"In the contemporary world, the early detection of any disease has become imperative. With an accelerating rate of population, the chance of fatality by breast cancer is growing exponentially. A reliable and effective detection system helps the medical personnel in fast detection of cancer. In the course of the present study, we have presented a comparative analysis of recent state-of the-art machine learning techniques that are being extensively used in cancer detection especially Breast Cancer by using the breast cancer dataset named Wisconsin dataset. We have statistically and comparatively scrutinized and compared the machine learning techniques that are used in classification like Naïve Bayes (NB), K-Nearest Neighbor (KNN), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), XGboost (XG) and Decision Tree (DT) for computing the accuracy in the light of performance metrics like recall, precision F1 score and accuracy percentage. Moreover, these classification techniques were also projected on ROC Curve. As a result, this research paper evaluates that the accuracy obtained by XGboost is 98.24% whereas in SVM the accuracy is 96.49%.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117263889","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758327
B. Behera, M. Mohanty
The authors show how to keep a flattened and low dispersion in a graded-index depressed core fiber for a few-mode operation in this paper. The proposed design supports seven linearly polarized modes over the C band. However, this design yields a dispersion flattened characteristic over the C-band of the optical communication spectrum. Moreover, this proposed few-mode structure exhibits zero-dispersion for the fundamental LP01 mode at 1550 nm first time according to our knowledge. The results show the dispersion slope of 0.007ps/nm2 km and the flatness of dispersion is about 1ps/nm km over the C-band. The proposed few-mode fiber is designed with an inner-core of 13.5% GeO2 doped silica, a trench with 2% F doped silica, and an outer-core with 3.5% GeO2 doped silica. The cladding is assumed to be fused Silica to maintain low optical losses and dispersion. Simulation is used to select the design parameters and the molar percentage of the dopants. The proposed fiber is suitably designed to guide 7 linearly polarized modes namely, LP01, LP11, LP21, LP02, LP31, LP12, and LP22. The proposed few-mode fiber exhibits large mode separation, flattened dispersion, low bending loss, low and flat differential mode delay, and large effective mode-area over the C-band. In summary, the proposed depressed core few-mode fiber is a prospective aspirant for new-generation mode-division multiplexing transmission.
{"title":"Design of Depressed-core Four-ring Few-mode Fiber for Next-Generation Communication","authors":"B. Behera, M. Mohanty","doi":"10.1109/ESCI53509.2022.9758327","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758327","url":null,"abstract":"The authors show how to keep a flattened and low dispersion in a graded-index depressed core fiber for a few-mode operation in this paper. The proposed design supports seven linearly polarized modes over the C band. However, this design yields a dispersion flattened characteristic over the C-band of the optical communication spectrum. Moreover, this proposed few-mode structure exhibits zero-dispersion for the fundamental LP01 mode at 1550 nm first time according to our knowledge. The results show the dispersion slope of 0.007ps/nm2 km and the flatness of dispersion is about 1ps/nm km over the C-band. The proposed few-mode fiber is designed with an inner-core of 13.5% GeO2 doped silica, a trench with 2% F doped silica, and an outer-core with 3.5% GeO2 doped silica. The cladding is assumed to be fused Silica to maintain low optical losses and dispersion. Simulation is used to select the design parameters and the molar percentage of the dopants. The proposed fiber is suitably designed to guide 7 linearly polarized modes namely, LP01, LP11, LP21, LP02, LP31, LP12, and LP22. The proposed few-mode fiber exhibits large mode separation, flattened dispersion, low bending loss, low and flat differential mode delay, and large effective mode-area over the C-band. In summary, the proposed depressed core few-mode fiber is a prospective aspirant for new-generation mode-division multiplexing transmission.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123840721","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758390
Sudeshna Baliarsingh, Saumendra Kumar Mohapatra, Prakash Kumar Panda, M. Mohanty
The Internet of things (IoT) has a great role to provide the recent technology including the applications in the area of Health care, Engineering, smart societies and many different human activities. It needs the compliant for the Artificial Intelligence of Medical things (AIoMT). Among all the processes including the signal processing, communication and Machine Learning, data compression playing an important role to satisfy all these applications. We have considered the Arrhythmia ECG data for compression using different transforms. The data is collected from the physio-net data base. Its performance using wavelet transform found suitable in terms of noise suppression, multiband filtering and compression encoding. Further to satisfy IoT based communication wavelet Packet Transform (WPT) is utilised to develop the model for multicarrier communication. From the result it is observed that it can be useful for Medical professionals as well as the patients from the remote places.
{"title":"Cardiac Data Compression for Reduced Traffic on Application of IoMT","authors":"Sudeshna Baliarsingh, Saumendra Kumar Mohapatra, Prakash Kumar Panda, M. Mohanty","doi":"10.1109/ESCI53509.2022.9758390","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758390","url":null,"abstract":"The Internet of things (IoT) has a great role to provide the recent technology including the applications in the area of Health care, Engineering, smart societies and many different human activities. It needs the compliant for the Artificial Intelligence of Medical things (AIoMT). Among all the processes including the signal processing, communication and Machine Learning, data compression playing an important role to satisfy all these applications. We have considered the Arrhythmia ECG data for compression using different transforms. The data is collected from the physio-net data base. Its performance using wavelet transform found suitable in terms of noise suppression, multiband filtering and compression encoding. Further to satisfy IoT based communication wavelet Packet Transform (WPT) is utilised to develop the model for multicarrier communication. From the result it is observed that it can be useful for Medical professionals as well as the patients from the remote places.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127218653","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 : 2022-03-09DOI: 10.1109/ESCI53509.2022.9758241
N. Ch., Pendurthi Pallavi Sai, G. Madhuri, Kota Srinath Reddy, Devireddy Venkata BharathSimha Reddy
Cervical cancer growth is the fourth maximum of regular diseases in females. It is brought about by long haul disease in skin cells and mucous film cells of the genital region. The World Health Organization (WHO) considers malignant growth a nonexclusive term for a huge gathering of infections that can influence any piece of the body, which is profoundly risky. In 2018, an expected 5,70,000 females were determined to have cervical malignancy worldwide, and around 3,11,000 females passed on from the illness. Hence proposing a model with high precision and high accuracy for diagnosing at the right phase of contamination will help a lot. This paper aims to develop machine learning(ML) algorithms like Support Vector Machine(SVM), Random Forest(RF) and Deep Learning (DL)models like Convolutional Neural Networks (CNN), Artificial Neural Networks (ANN) using python, which gives more accurate results compared to existing models. The accuracy of each model SVM, CNN, RF and ANN obtained was 97%, 95.3%, 94% and 9 5.2%, respectively, where SVM has higher precision among ML algorithms similarly, CNN has the highest precision among the neural network algorithms, So to anticipate the cervical disease and to help in its initial judgments which can shield women in huge scope from being affected to this disease.
{"title":"Artificial Intelligence based Cervical Cancer Risk Prediction Using M1 Algorithms","authors":"N. Ch., Pendurthi Pallavi Sai, G. Madhuri, Kota Srinath Reddy, Devireddy Venkata BharathSimha Reddy","doi":"10.1109/ESCI53509.2022.9758241","DOIUrl":"https://doi.org/10.1109/ESCI53509.2022.9758241","url":null,"abstract":"Cervical cancer growth is the fourth maximum of regular diseases in females. It is brought about by long haul disease in skin cells and mucous film cells of the genital region. The World Health Organization (WHO) considers malignant growth a nonexclusive term for a huge gathering of infections that can influence any piece of the body, which is profoundly risky. In 2018, an expected 5,70,000 females were determined to have cervical malignancy worldwide, and around 3,11,000 females passed on from the illness. Hence proposing a model with high precision and high accuracy for diagnosing at the right phase of contamination will help a lot. This paper aims to develop machine learning(ML) algorithms like Support Vector Machine(SVM), Random Forest(RF) and Deep Learning (DL)models like Convolutional Neural Networks (CNN), Artificial Neural Networks (ANN) using python, which gives more accurate results compared to existing models. The accuracy of each model SVM, CNN, RF and ANN obtained was 97%, 95.3%, 94% and 9 5.2%, respectively, where SVM has higher precision among ML algorithms similarly, CNN has the highest precision among the neural network algorithms, So to anticipate the cervical disease and to help in its initial judgments which can shield women in huge scope from being affected to this disease.","PeriodicalId":436539,"journal":{"name":"2022 International Conference on Emerging Smart Computing and Informatics (ESCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125436742","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：