Pub Date : 2022-12-10DOI: 10.1109/STCR55312.2022.10009275
Akshath Mahajan, Deap Daru, Aditya Thaker, M. Narvekar, Debajyoti Mukhopadhyay
India is prone to tropical cyclones annually, originating from the North Indian Ocean basin. Tropical cyclones are destructive and sudden natural occurrences that annually wreak havoc by taking a huge toll on human lives and property. This engenders a need for accurately forecasting the scale of such mass-destructive events, to provide us with enough time to take precautionary measures that can reduce the death toll and minimize costs. Using the CyINSAT dataset, which gives a multimodal and temporal resolution for TCs occurred from 2014 to 2022, this paper employs and compares multiple techniques to solve the wind speed forecasting issue. All models involve recurrent networks along with image feature extractors, which are used together to predict the next wind speeds from a sequence of images. The architectural differences between these models mainly focus on the nuances involved in handling the current wind speed. The proposed architecture gives higher importance to the currently recorded wind speeds and performs significantly better than the baseline models. It successfully obtained an RMSE of 6.31, MAE of 0.093 and MAPE of 4.53.
{"title":"Forecasting North Indian Ocean Tropical Cyclone Intensity","authors":"Akshath Mahajan, Deap Daru, Aditya Thaker, M. Narvekar, Debajyoti Mukhopadhyay","doi":"10.1109/STCR55312.2022.10009275","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009275","url":null,"abstract":"India is prone to tropical cyclones annually, originating from the North Indian Ocean basin. Tropical cyclones are destructive and sudden natural occurrences that annually wreak havoc by taking a huge toll on human lives and property. This engenders a need for accurately forecasting the scale of such mass-destructive events, to provide us with enough time to take precautionary measures that can reduce the death toll and minimize costs. Using the CyINSAT dataset, which gives a multimodal and temporal resolution for TCs occurred from 2014 to 2022, this paper employs and compares multiple techniques to solve the wind speed forecasting issue. All models involve recurrent networks along with image feature extractors, which are used together to predict the next wind speeds from a sequence of images. The architectural differences between these models mainly focus on the nuances involved in handling the current wind speed. The proposed architecture gives higher importance to the currently recorded wind speeds and performs significantly better than the baseline models. It successfully obtained an RMSE of 6.31, MAE of 0.093 and MAPE of 4.53.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124427410","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009329
D. R, V. S, A. S, Srinivethaa Pongiannan, Sabareshwaran M, Hareesh T
Damage to the skin is a leading cause of death worldwide. In the event that it is not immediately handled and dissected, it might acquire into interact with numerous organs and tissues. The high turnover of skin cells exposed to sunlight has similar consequences. It is hoped that having early, observable confirmation from a trustworthy automated system for validating skin sores will save time, effort, and human lives. An effective method of treating skin cancer is to combine in-depth information with image alteration. This hints at a mechanical method of depicting skin disorders. We can see the limits and scope of the primary convolutional mind links. The dataset includes information on nine clinical types of skin damage, including actinic keratosis, basal cell carcinoma, benign keratosis, dermatofibroma, melanoma, nevus, seborrhea keratosis, squamous cell carcinoma, and vascular wounds. Our aim is to use a convolutional neural network to categorize a model that classifies skin diseases into distinct groups. The diagnostic method is based on the ideas of thorough image collection and extensive learning. Various picture enhancement methods have also contributed to a rise in the total number of photographs available. The precision of the collecting chores is also addressed by the trade learning method.
{"title":"Optimized Skin Cancer Detection using Web Technology","authors":"D. R, V. S, A. S, Srinivethaa Pongiannan, Sabareshwaran M, Hareesh T","doi":"10.1109/STCR55312.2022.10009329","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009329","url":null,"abstract":"Damage to the skin is a leading cause of death worldwide. In the event that it is not immediately handled and dissected, it might acquire into interact with numerous organs and tissues. The high turnover of skin cells exposed to sunlight has similar consequences. It is hoped that having early, observable confirmation from a trustworthy automated system for validating skin sores will save time, effort, and human lives. An effective method of treating skin cancer is to combine in-depth information with image alteration. This hints at a mechanical method of depicting skin disorders. We can see the limits and scope of the primary convolutional mind links. The dataset includes information on nine clinical types of skin damage, including actinic keratosis, basal cell carcinoma, benign keratosis, dermatofibroma, melanoma, nevus, seborrhea keratosis, squamous cell carcinoma, and vascular wounds. Our aim is to use a convolutional neural network to categorize a model that classifies skin diseases into distinct groups. The diagnostic method is based on the ideas of thorough image collection and extensive learning. Various picture enhancement methods have also contributed to a rise in the total number of photographs available. The precision of the collecting chores is also addressed by the trade learning method.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122694762","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009052
S. Rangeetha, C. Ganesh Babu, N. Nagarajan, M. Madhumalini
In spacecraft electronics realization, while testing cards, there is a need for the calculation of speed and accuracy to be high. But when the calculations are made manually, there are more chances of error which reduces the accuracy and also for the reduction in calculation speed. The manual process requires more time for calculating and analyzing the results. To overcome the errors, to reduce the time consumption and also to increase the speed, an automated card testing method has to be designed. In the automated card testing method, the pulse, data, signal that are obtained will be analyzed automatically to provide accurate and quick results. In the automated card testing method, there will be very less error and consumes less amount of time when compared to the manual process. The pulse width, the data transferred and the difference in the time period can be obtained quickly with the help of automated card testing method.
{"title":"Automated Card Testing using Onboard Test Interface Simulator","authors":"S. Rangeetha, C. Ganesh Babu, N. Nagarajan, M. Madhumalini","doi":"10.1109/STCR55312.2022.10009052","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009052","url":null,"abstract":"In spacecraft electronics realization, while testing cards, there is a need for the calculation of speed and accuracy to be high. But when the calculations are made manually, there are more chances of error which reduces the accuracy and also for the reduction in calculation speed. The manual process requires more time for calculating and analyzing the results. To overcome the errors, to reduce the time consumption and also to increase the speed, an automated card testing method has to be designed. In the automated card testing method, the pulse, data, signal that are obtained will be analyzed automatically to provide accurate and quick results. In the automated card testing method, there will be very less error and consumes less amount of time when compared to the manual process. The pulse width, the data transferred and the difference in the time period can be obtained quickly with the help of automated card testing method.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"136 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116720671","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009421
Padmini M S, S. Kuzhalivaimozhi, Pruthu V Simha, Pulkit Singh, Abhinandan A
Recently, there has been an increase in the integration of good devices and platforms like unmanned Aerial Vehicles (UAVs) and drones into the universal network of the net of Things (IoT). Unmanned aerial vehicles (UAVs) that are autonomous use navigation and control software that is powered by artificial intelligence (AI) and do not need a human pilot to fly them. As technology is advancing day by day, the trend to replace humans with robots & other devices like drones and unmanned Aerial Vehicles (UAVs) will be implemented and will be kept up for quite a long time. Essentially, an associate degree autonomous drone could be a flying vehicle that may be remotely controlled or fly autonomously using software-controlled flight plans that work in conjunction with onboard sensors and a worldwide positioning system (GPS). This work to boot explores the technical efforts toward facultative safe UAV operations exploitation associate degree autonomous nano drone, whereas taking into thought varied challenges like security, precision, and varied challenges thanks to the restrictions of the autonomous drone.
{"title":"An Implementation of Gesture-Controlled Autonomous Drone","authors":"Padmini M S, S. Kuzhalivaimozhi, Pruthu V Simha, Pulkit Singh, Abhinandan A","doi":"10.1109/STCR55312.2022.10009421","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009421","url":null,"abstract":"Recently, there has been an increase in the integration of good devices and platforms like unmanned Aerial Vehicles (UAVs) and drones into the universal network of the net of Things (IoT). Unmanned aerial vehicles (UAVs) that are autonomous use navigation and control software that is powered by artificial intelligence (AI) and do not need a human pilot to fly them. As technology is advancing day by day, the trend to replace humans with robots & other devices like drones and unmanned Aerial Vehicles (UAVs) will be implemented and will be kept up for quite a long time. Essentially, an associate degree autonomous drone could be a flying vehicle that may be remotely controlled or fly autonomously using software-controlled flight plans that work in conjunction with onboard sensors and a worldwide positioning system (GPS). This work to boot explores the technical efforts toward facultative safe UAV operations exploitation associate degree autonomous nano drone, whereas taking into thought varied challenges like security, precision, and varied challenges thanks to the restrictions of the autonomous drone.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126981553","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009230
Poongodi C, D. D., S. k, P. T, M. D
The wireless communication operating in sub 6GHz channel models is facing challenges like user demand and data traffic. So, the Millimeter wave (mmWave) frequencies are introduced in 5G communication, and it is operating in 6GHz to 100GHz band. The propagation nature of the mmWave channel models is more complicated and it is prone to path loss, shadowing effects, and other atmospheric conditions. So the channel modeling for indoor and outdoor communication is a more challenging one. Various channel models are proposed for 5G communication. This paper focuses on the 3GPP TR38.901 channel model in Urban Micro (UMi), Urban Macro (UMa) environments. The pathloss and channel capacity is analyzed up to 100GHz frequency, most preferably 28GHZ, 38GHz, 60GHz and 73GHz. Higher frequency provides high pathloss and less capacity, but it covers wide band of frequencies.
{"title":"Millimeter Wave Channel in Urban Micro / Urban Macro Environments: Path Loss Model and its Effect on Channel Capacity","authors":"Poongodi C, D. D., S. k, P. T, M. D","doi":"10.1109/STCR55312.2022.10009230","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009230","url":null,"abstract":"The wireless communication operating in sub 6GHz channel models is facing challenges like user demand and data traffic. So, the Millimeter wave (mmWave) frequencies are introduced in 5G communication, and it is operating in 6GHz to 100GHz band. The propagation nature of the mmWave channel models is more complicated and it is prone to path loss, shadowing effects, and other atmospheric conditions. So the channel modeling for indoor and outdoor communication is a more challenging one. Various channel models are proposed for 5G communication. This paper focuses on the 3GPP TR38.901 channel model in Urban Micro (UMi), Urban Macro (UMa) environments. The pathloss and channel capacity is analyzed up to 100GHz frequency, most preferably 28GHZ, 38GHz, 60GHz and 73GHz. Higher frequency provides high pathloss and less capacity, but it covers wide band of frequencies.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114228037","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009519
B. S, H. K, S. M
In this digital era, identifying human facial expressions and responding accordingly is an emerging need. On the other hand, similar data can be used for surveillance activities and relative crime detection. Human Facial Expression Recognition (HFER) based on human face expression variations in real-time is proposed in this paper. Here two CNN- models are cascaded to produce the facial expression recognition output. YOLO V5 is used for people detection and custom trained CNN model is used for expression recognition. The suggested model provides better accuracy of 95.57% for seven different facial expressions than the existing models.
{"title":"Expression Recognition using YOLO and Shallow CNN Model","authors":"B. S, H. K, S. M","doi":"10.1109/STCR55312.2022.10009519","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009519","url":null,"abstract":"In this digital era, identifying human facial expressions and responding accordingly is an emerging need. On the other hand, similar data can be used for surveillance activities and relative crime detection. Human Facial Expression Recognition (HFER) based on human face expression variations in real-time is proposed in this paper. Here two CNN- models are cascaded to produce the facial expression recognition output. YOLO V5 is used for people detection and custom trained CNN model is used for expression recognition. The suggested model provides better accuracy of 95.57% for seven different facial expressions than the existing models.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133780984","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009366
Suhail Ashaq, Mir Nazish, Mehvish Ali, Ishfaq Sultan, M. Tariq Banday
Conventional cryptographic techniques such as Advanced Encryption Standard (AES) being resource intensive are not feasible for low-end Internet of Things (IoT) devices. As such, several lightweight crypto primitives have been designed to offer an optimum level of security along with reduced resource utilisation. Also, because of the trade-offs between different metrics, lightweight cryptography often targets a specific parameter, making it a good fit for a particular field of IoT application. This paper aims to reduce the hardware footprint of the PRESENT block cypher with the area-efficient hardware design of Substitution-Box, which is the most resource-consuming part of the PRESENT cypher. The proposed hardware design for S-Box is implemented in the state-of-the-art architectures reported in the literature. The designs are implemented on the FPGAs to assess resource consumption and performance. The original designs and their implementation with the proposed hardware for S-Box have been compared in terms of resource consumption, maximum achievable throughput, and throughput per slice. The results indicate a 13.67% improvement in resource consumption by adopting the proposed S-Box in the architecture. Moreover, throughput has been increased for certain PRESENT architectures, thus enhancing their overall performance.
{"title":"FPGA Implementation of PRESENT Block Cypher with Optimised Substitution Box","authors":"Suhail Ashaq, Mir Nazish, Mehvish Ali, Ishfaq Sultan, M. Tariq Banday","doi":"10.1109/STCR55312.2022.10009366","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009366","url":null,"abstract":"Conventional cryptographic techniques such as Advanced Encryption Standard (AES) being resource intensive are not feasible for low-end Internet of Things (IoT) devices. As such, several lightweight crypto primitives have been designed to offer an optimum level of security along with reduced resource utilisation. Also, because of the trade-offs between different metrics, lightweight cryptography often targets a specific parameter, making it a good fit for a particular field of IoT application. This paper aims to reduce the hardware footprint of the PRESENT block cypher with the area-efficient hardware design of Substitution-Box, which is the most resource-consuming part of the PRESENT cypher. The proposed hardware design for S-Box is implemented in the state-of-the-art architectures reported in the literature. The designs are implemented on the FPGAs to assess resource consumption and performance. The original designs and their implementation with the proposed hardware for S-Box have been compared in terms of resource consumption, maximum achievable throughput, and throughput per slice. The results indicate a 13.67% improvement in resource consumption by adopting the proposed S-Box in the architecture. Moreover, throughput has been increased for certain PRESENT architectures, thus enhancing their overall performance.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128641735","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009345
Saranya S, J. M, Sakthivel V, Seema Aashikab A, S. P
The world has suffered enough in the aspect of COVID-19. From the year 2019, all we have in our hearts is a constant fear and terror of becoming prey to this deadly virus that has almost taken five lakh and twenty-five thousand lives to date within India, as the statistics show. The way to ensure that you maintain proper public hygiene is by ensuring that you wear masks in public places. There have been many algorithms that provide quicker results. We have tested our model in K-Nearest Neighbors (KNN), Support Vector Machine (SVM) algorithms and using deep learning technique Convolution Neural Networks (CNN). Comparing others, CNN provides more accuracy and has a shorter latency. Thus, we have implemented human face mask detector using CNN. The body temperature of the individual entering a room is monitored by the support of myDAQ, NI Instruments. If the body temperature is higher than 99F, then the person entering the space is not permitted inside. We have designed a device that monitors the temperature of the person entering the room along with the monitoring of face masks using the webcam.
{"title":"Human Body Temperature and Face Mask Audit System for COVID Protocol","authors":"Saranya S, J. M, Sakthivel V, Seema Aashikab A, S. P","doi":"10.1109/STCR55312.2022.10009345","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009345","url":null,"abstract":"The world has suffered enough in the aspect of COVID-19. From the year 2019, all we have in our hearts is a constant fear and terror of becoming prey to this deadly virus that has almost taken five lakh and twenty-five thousand lives to date within India, as the statistics show. The way to ensure that you maintain proper public hygiene is by ensuring that you wear masks in public places. There have been many algorithms that provide quicker results. We have tested our model in K-Nearest Neighbors (KNN), Support Vector Machine (SVM) algorithms and using deep learning technique Convolution Neural Networks (CNN). Comparing others, CNN provides more accuracy and has a shorter latency. Thus, we have implemented human face mask detector using CNN. The body temperature of the individual entering a room is monitored by the support of myDAQ, NI Instruments. If the body temperature is higher than 99F, then the person entering the space is not permitted inside. We have designed a device that monitors the temperature of the person entering the room along with the monitoring of face masks using the webcam.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132082331","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009301
Nathiya S, Sumitha J, V. M, S. S., Sathana G
The second most widespread cancer in women, breast cancer, is the most lethal and is responsible for an increasing number of daily deaths in females. Breast cancer patients who receive an advanced diagnosis have a higher probability of dying and reduces the probability of surviving. Numerous research projects are being implemented in order to improve breast cancer rapid identification. Despite the existence of numerous medical diagnostic techniques for predicting breast cancer, it remains difficult to anticipate breast cancer at its earliest stages. The major objective of this research is to establish a predictive model that could really identify breast cancer earlier on and increase survival rates. The Wisconsin Breast Cancer (Original) Dataset is utilized in this research to forecast breast cancer utilizing techniques from data mining like K-Nearest Neighbor Algorithm(KNN), Support Vector Machine Algorithm(SVM), and Artificial Neural Network Algorithm(ANN), a new model SVM-ANN optimized algorithm is also proposed. A variety of parameters which including Accuracy, Precision, and Recall, have been employed to compare the results of these algorithms in order to fully evaluate their effectiveness. Finally, with a 97 percent accuracy rate, the proposed algorithm SVM-ANN optimized algorithm significantly outperformed other current algorithms.
{"title":"SVM-ANN Optimized Algorithm for the Classification of Breast Cancer Data as Benign and Malignant","authors":"Nathiya S, Sumitha J, V. M, S. S., Sathana G","doi":"10.1109/STCR55312.2022.10009301","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009301","url":null,"abstract":"The second most widespread cancer in women, breast cancer, is the most lethal and is responsible for an increasing number of daily deaths in females. Breast cancer patients who receive an advanced diagnosis have a higher probability of dying and reduces the probability of surviving. Numerous research projects are being implemented in order to improve breast cancer rapid identification. Despite the existence of numerous medical diagnostic techniques for predicting breast cancer, it remains difficult to anticipate breast cancer at its earliest stages. The major objective of this research is to establish a predictive model that could really identify breast cancer earlier on and increase survival rates. The Wisconsin Breast Cancer (Original) Dataset is utilized in this research to forecast breast cancer utilizing techniques from data mining like K-Nearest Neighbor Algorithm(KNN), Support Vector Machine Algorithm(SVM), and Artificial Neural Network Algorithm(ANN), a new model SVM-ANN optimized algorithm is also proposed. A variety of parameters which including Accuracy, Precision, and Recall, have been employed to compare the results of these algorithms in order to fully evaluate their effectiveness. Finally, with a 97 percent accuracy rate, the proposed algorithm SVM-ANN optimized algorithm significantly outperformed other current algorithms.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121933702","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 : 2022-12-10DOI: 10.1109/STCR55312.2022.10009074
Sanjay V, S. P.
A neurological disease is Parkinson's disease. It causes trembling in the hands, trouble walking, losing balance, and coordination. In the high-level stage, there is no access to medical care. Blood test reports, CT scan results, and X-ray reports are not accessible early enough. Early Parkinson’s disease detection is crucial to implement effective treatment. The purpose of the proposed effort was to identify Parkinson’s disease in early prediction using clinical imaging and machine learning technologies. Despite the fact that there are numerous methods for detecting Parkinson’s disease, using MRI scan images still it is a big challenge. In this study, an Adaboost classifier is used with a hybrid PSO algorithm to propose a novel technique for detecting Parkinson’s disease. Adaboost acted as the best classifier among other classifiers. Initially, MRI image best features are extracted and identified by the curvelet transform and principal component analysis. This Ad boost classifier receives optimal features as input. Finally, Adaboost classifieds the MRI images and gave excellent classification accuracy. To evaluate the proposed method three methods metrics namely accuracy, specificity, and sensitivity are used. Based on the results the proposed methods yield greater accuracy than the existing systems.
{"title":"Machine Learning Techniques for Parkinson's Disease Detection","authors":"Sanjay V, S. P.","doi":"10.1109/STCR55312.2022.10009074","DOIUrl":"https://doi.org/10.1109/STCR55312.2022.10009074","url":null,"abstract":"A neurological disease is Parkinson's disease. It causes trembling in the hands, trouble walking, losing balance, and coordination. In the high-level stage, there is no access to medical care. Blood test reports, CT scan results, and X-ray reports are not accessible early enough. Early Parkinson’s disease detection is crucial to implement effective treatment. The purpose of the proposed effort was to identify Parkinson’s disease in early prediction using clinical imaging and machine learning technologies. Despite the fact that there are numerous methods for detecting Parkinson’s disease, using MRI scan images still it is a big challenge. In this study, an Adaboost classifier is used with a hybrid PSO algorithm to propose a novel technique for detecting Parkinson’s disease. Adaboost acted as the best classifier among other classifiers. Initially, MRI image best features are extracted and identified by the curvelet transform and principal component analysis. This Ad boost classifier receives optimal features as input. Finally, Adaboost classifieds the MRI images and gave excellent classification accuracy. To evaluate the proposed method three methods metrics namely accuracy, specificity, and sensitivity are used. Based on the results the proposed methods yield greater accuracy than the existing systems.","PeriodicalId":338691,"journal":{"name":"2022 Smart Technologies, Communication and Robotics (STCR)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115282889","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：