Pub Date : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179738
Chandragiri Vasanth Kumar, Saravanan. M.S, R. Surendran
The research is to study the patient's length of stay in intensive care unit (ICU) admissions each year with their cost and health expenditure. Forecasting in the clinical Decision Support System (DSS) is being developed in the study to anticipate and enhance hospital equipment for patients' health analysis. The most crucial examination is to give appropriate technology and quality drugs to analyze the patient's health, which is then recorded in electronic medical records. To achieve the best exactness, this research study employed the innovative Feed Forward Neural Network and Deep Belief Network to accomplish the operations. The study gathered 47 samples from two groups of calculation with a G-power of 80% and their Patient electronic health records investigations were collected from a variety of online sources, with recent research findings and a 0.05% threshold, confidence interval of 95% mean and standard deviation. The unique Feed Forward Neural Network approach obtained 93.65% accuracy in predicting ICU analysis; consequently, The Deep Belief Network method in machine learning should be upgraded for improved accuracy in health prediction in this study. This study discovered a 90.07% accuracy for ICU analysis utilizing the Deep Belief Network method, with a significant value of two-tailed tests of 0.006 (p0.05) and a 95% confidence range. This study reveals that the innovative Feed Forward Neural Network method outperforms the Deep Belief Network algorithm for ICU analysis of patients.
{"title":"Prediction of Insufficient Accuracy for Patient's Length of Stay using Feed Forward Neural Network by comparing Deep Belief Network","authors":"Chandragiri Vasanth Kumar, Saravanan. M.S, R. Surendran","doi":"10.1109/ICECCT56650.2023.10179738","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179738","url":null,"abstract":"The research is to study the patient's length of stay in intensive care unit (ICU) admissions each year with their cost and health expenditure. Forecasting in the clinical Decision Support System (DSS) is being developed in the study to anticipate and enhance hospital equipment for patients' health analysis. The most crucial examination is to give appropriate technology and quality drugs to analyze the patient's health, which is then recorded in electronic medical records. To achieve the best exactness, this research study employed the innovative Feed Forward Neural Network and Deep Belief Network to accomplish the operations. The study gathered 47 samples from two groups of calculation with a G-power of 80% and their Patient electronic health records investigations were collected from a variety of online sources, with recent research findings and a 0.05% threshold, confidence interval of 95% mean and standard deviation. The unique Feed Forward Neural Network approach obtained 93.65% accuracy in predicting ICU analysis; consequently, The Deep Belief Network method in machine learning should be upgraded for improved accuracy in health prediction in this study. This study discovered a 90.07% accuracy for ICU analysis utilizing the Deep Belief Network method, with a significant value of two-tailed tests of 0.006 (p0.05) and a 95% confidence range. This study reveals that the innovative Feed Forward Neural Network method outperforms the Deep Belief Network algorithm for ICU analysis of patients.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132532112","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179797
Akash Mahanand, Prathibha Prakash, Anjuna Devaraj
Web traffic is a kind of time-series motion, having its highs and lows. The analysis of predicting web traffic has a greater significance for website owners, to make reliable decisions for website users. But the major gripe often faced while exploring concealed and significant details are regarding web users' different usage patterns. In this paper, we apply hybrid-based deep learning algorithms which combine two different architectures of Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), and Gated Recurrent Unit (GRU). The outcome of our hybrid model is acquired by using the ensemble method of stacking. The Web Traffic Time Series Forecasting(WTTSF) dataset by Kaggle is being used to predict future traffic of Wikipedia articles. We use mean squared error, mean absolute error, and $R^{2}$ as major conventional evaluation metrics and it offers less error even though it has data randomness over a large scale.
{"title":"Deep Learning-based Hybrid Technique for Forecasting Web Traffic","authors":"Akash Mahanand, Prathibha Prakash, Anjuna Devaraj","doi":"10.1109/ICECCT56650.2023.10179797","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179797","url":null,"abstract":"Web traffic is a kind of time-series motion, having its highs and lows. The analysis of predicting web traffic has a greater significance for website owners, to make reliable decisions for website users. But the major gripe often faced while exploring concealed and significant details are regarding web users' different usage patterns. In this paper, we apply hybrid-based deep learning algorithms which combine two different architectures of Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), and Gated Recurrent Unit (GRU). The outcome of our hybrid model is acquired by using the ensemble method of stacking. The Web Traffic Time Series Forecasting(WTTSF) dataset by Kaggle is being used to predict future traffic of Wikipedia articles. We use mean squared error, mean absolute error, and $R^{2}$ as major conventional evaluation metrics and it offers less error even though it has data randomness over a large scale.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"331 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132542564","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179820
Pon Selchiya R, M. P, T. S
Melanoma is a type of skin cancer that matures when melanocytes start to grow out of control which leads to a Noxious Disease. Eventually, Genetics are another factor for this Tumor. It's broadly known that growing levels of ultraviolet (UV) exposure are one of the most motives for this fast rise in the variety of skin cancer cases. Melanoma can be detected or classified using the Artificial Intelligence techniques either using deep learning or machine learning algorithms. In condition to classify the cancer detection, the deep learning algorithms are applied in Melanoma skin cancer dataset of 10000 images in this project.
{"title":"Comparative Analysis Based Melanoma Detection In Dermoscopic Images With Deep Learning Techniques","authors":"Pon Selchiya R, M. P, T. S","doi":"10.1109/ICECCT56650.2023.10179820","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179820","url":null,"abstract":"Melanoma is a type of skin cancer that matures when melanocytes start to grow out of control which leads to a Noxious Disease. Eventually, Genetics are another factor for this Tumor. It's broadly known that growing levels of ultraviolet (UV) exposure are one of the most motives for this fast rise in the variety of skin cancer cases. Melanoma can be detected or classified using the Artificial Intelligence techniques either using deep learning or machine learning algorithms. In condition to classify the cancer detection, the deep learning algorithms are applied in Melanoma skin cancer dataset of 10000 images in this project.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130157605","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179631
Satheeswari Damodaran, Leninisha Shanmugam, N. Swaroopan
To ensure the integrity of power lines, electrical transmission towers must be monitored. Monitoring vegetation encroachment, which can lead to power outages, is a significant challenge. The majority of current monitoring techniques rely on manual labor and traditional methods of observation such as unmanned aerial vehicles (UAV) and airborne photography. Monitoring large areas with these methods, however, is expensive and time consuming. Our paper describes a method for monitoring power line corridors with UAV images. A two-stage procedure is proposed. Background clustering was performed using Fuzzy C-means in the first stage. Our second step was to detect the presence of transmission towers using state-of-the-art deep learning technologies AlexNet and DenseNet-121. By comparing the two deep learning architectures, the proposed methodology detects the transmission tower from VAV images with an accuracy of 94.8% for AlexNet and 98.6% for DenseNet - 121 with better precision, recall, and F1-score.
{"title":"Extraction of Overhead Transmission Towers from UAV Images","authors":"Satheeswari Damodaran, Leninisha Shanmugam, N. Swaroopan","doi":"10.1109/ICECCT56650.2023.10179631","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179631","url":null,"abstract":"To ensure the integrity of power lines, electrical transmission towers must be monitored. Monitoring vegetation encroachment, which can lead to power outages, is a significant challenge. The majority of current monitoring techniques rely on manual labor and traditional methods of observation such as unmanned aerial vehicles (UAV) and airborne photography. Monitoring large areas with these methods, however, is expensive and time consuming. Our paper describes a method for monitoring power line corridors with UAV images. A two-stage procedure is proposed. Background clustering was performed using Fuzzy C-means in the first stage. Our second step was to detect the presence of transmission towers using state-of-the-art deep learning technologies AlexNet and DenseNet-121. By comparing the two deep learning architectures, the proposed methodology detects the transmission tower from VAV images with an accuracy of 94.8% for AlexNet and 98.6% for DenseNet - 121 with better precision, recall, and F1-score.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130197737","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179724
Anurita Bose, Deepanjali Pandit, Nidhi Prakash, Ashwini M. Joshi
Sarcasm refers to the use of irony to mock or convey contempt and involves the use of words that mean the opposite of what someone truly intends to convey. Online forums which enable users to express sarcasm as a sentiment tend to induce misunderstandings between different parties and obscure the users' true intentions. This leads to ambiguity being one of the prime challenges in detecting sarcasm. Another challenge in sarcasm detection is the rapidly growing size of language vocabularies with the addition of new slang words every day. Additionally, usage of emojis in online text can greatly influence the polarity of a sentence by inducing a sarcastic tone. These setbacks make sarcasm a particularly demanding sentiment to determine. In this paper, the statistical significance of various deep learning models for the purpose of detecting sarcasm in online comments containing emojis is explored. For the task of binary classification, GRU achieves an accuracy score of 73.44% with an F1-score of 73.96%. The proposed ensemble-based approach yields an accuracy score of 74.41% for the combination of LSTM and GRU, which is comparable to the accuracy achieved with conventional ensemble techniques such as max-voting and averaging. Twenty-six different hybrid combinations of deep learning models were explored and the most optimal performing ones were identified. CNN and Global Average Pooling 1D are two other architectures that were explored.
讽刺指的是用讽刺的方式来嘲笑或表达蔑视，包括使用与某人真正想表达的意思相反的词语。允许用户将讽刺作为一种情感表达的网络论坛，容易引起各方之间的误解，模糊用户的真实意图。这导致歧义成为检测讽刺的主要挑战之一。讽刺检测的另一个挑战是语言词汇量的快速增长，每天都有新的俚语词汇增加。此外，在网络文本中使用表情符号可以通过诱导讽刺语气来极大地影响句子的极性。这些挫折使讽刺成为一种特别需要判断的情绪。本文探讨了各种深度学习模型用于检测包含表情符号的在线评论中的讽刺的统计意义。对于二值分类任务，GRU的准确率得分为73.44%，f1得分为73.96%。基于集成的LSTM和GRU组合方法的准确率为74.41%，与传统集成技术(如max-voting和average)的准确率相当。探索了26种不同的深度学习模型混合组合，并确定了性能最优的模型。CNN和Global Average Pooling 1D是我们探索的另外两种架构。
{"title":"A Deviation based Ensemble Algorithm for Sarcasm Detection in Online Comments","authors":"Anurita Bose, Deepanjali Pandit, Nidhi Prakash, Ashwini M. Joshi","doi":"10.1109/ICECCT56650.2023.10179724","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179724","url":null,"abstract":"Sarcasm refers to the use of irony to mock or convey contempt and involves the use of words that mean the opposite of what someone truly intends to convey. Online forums which enable users to express sarcasm as a sentiment tend to induce misunderstandings between different parties and obscure the users' true intentions. This leads to ambiguity being one of the prime challenges in detecting sarcasm. Another challenge in sarcasm detection is the rapidly growing size of language vocabularies with the addition of new slang words every day. Additionally, usage of emojis in online text can greatly influence the polarity of a sentence by inducing a sarcastic tone. These setbacks make sarcasm a particularly demanding sentiment to determine. In this paper, the statistical significance of various deep learning models for the purpose of detecting sarcasm in online comments containing emojis is explored. For the task of binary classification, GRU achieves an accuracy score of 73.44% with an F1-score of 73.96%. The proposed ensemble-based approach yields an accuracy score of 74.41% for the combination of LSTM and GRU, which is comparable to the accuracy achieved with conventional ensemble techniques such as max-voting and averaging. Twenty-six different hybrid combinations of deep learning models were explored and the most optimal performing ones were identified. CNN and Global Average Pooling 1D are two other architectures that were explored.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133952657","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179646
Subramanyam Shashi Kumar, Prakash Ramachandran
The basic functioning of heart can be read through Electrocardiogram (ECG) Signal, this signal gives an idea whether the functioning of heart is normal or abnormal and type abnormality can also be identified, which helps to diagnose the patients in time. This work investigates a deep-learning model using 2DCNN to classify various category of ECG signal. This proposed CNN model is trained and tested to classify three different classes of heart arrhythmia such as cardiac arrhythmia (ARR), congestive heart failure (CHF) and normal sinus rhythms (NSR). The time domain ECG signal is preprocessed and further it is transformed in to time-frequency scalogram by utilizing continuous wavelet transform (CWT), these scalogram is remodeled and saved as RGB images with necessary dimensions. Later these converted RGB images are fed to the input of various 2DCNN models such as alexnet, vgg16, squeezenet and googlenet to classify arrhythmia type. ECG Recordings from MIT BIH database were chosen and used for training and testing dataset. The performance of proposed scheme is evaluated on various CNN networks, a reasonable classification accuracy of 99.33 % was acheived by alex net.
{"title":"Multi-Class ECG Signal Processing and Classification using CWT based on various Deep Neural Networks","authors":"Subramanyam Shashi Kumar, Prakash Ramachandran","doi":"10.1109/ICECCT56650.2023.10179646","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179646","url":null,"abstract":"The basic functioning of heart can be read through Electrocardiogram (ECG) Signal, this signal gives an idea whether the functioning of heart is normal or abnormal and type abnormality can also be identified, which helps to diagnose the patients in time. This work investigates a deep-learning model using 2DCNN to classify various category of ECG signal. This proposed CNN model is trained and tested to classify three different classes of heart arrhythmia such as cardiac arrhythmia (ARR), congestive heart failure (CHF) and normal sinus rhythms (NSR). The time domain ECG signal is preprocessed and further it is transformed in to time-frequency scalogram by utilizing continuous wavelet transform (CWT), these scalogram is remodeled and saved as RGB images with necessary dimensions. Later these converted RGB images are fed to the input of various 2DCNN models such as alexnet, vgg16, squeezenet and googlenet to classify arrhythmia type. ECG Recordings from MIT BIH database were chosen and used for training and testing dataset. The performance of proposed scheme is evaluated on various CNN networks, a reasonable classification accuracy of 99.33 % was acheived by alex net.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134103522","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179713
Onima Tigga, Jaya Pal, D. Mustafi
In recent times, Machine Learning methods are widely used to handle large and complex data to generate interesting patterns and trends. Supervised Learning methods are generally used to classify different types of real life datasets. In this paper, the two methods Multiple Linear Regression and K Nearest Neighbours have been used to classify the quality of wine and compare the accuracy. As a result, it is found that K Nearest Neighbours gives the good accuracy. The calculated Mean Squared Error (MSE) and calculated Root Mean Squared Error (RMSE) give the model perfection. Result shows that the value of MSE and RMSE applying K Nearest Neighbours (KNN) is higher than Multiple Linear Regression (MLR). The classification performance of the methods is compared with their accuracy. Based on these methods, the highest accuracy of KNN with K = 5 is 0.9444. Meanwhile, for the Multiple Linear Regression, the accuracy reached to 0.6657. Also, MSE and RMSE are calculated as 0.0555 and 0.2357 for KNN with k=5. Multiple Linear Regression has MSE (0.1692) and RMSE (0.4113). The experimental result shows that KNN can be used as alternative method for predicting the new instances. From UCI Machine Learning Repository, the wine dataset is taken which are tested in this research paper.
{"title":"A Comparative Study of Multiple Linear Regression and K Nearest Neighbours using Machine Learning","authors":"Onima Tigga, Jaya Pal, D. Mustafi","doi":"10.1109/ICECCT56650.2023.10179713","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179713","url":null,"abstract":"In recent times, Machine Learning methods are widely used to handle large and complex data to generate interesting patterns and trends. Supervised Learning methods are generally used to classify different types of real life datasets. In this paper, the two methods Multiple Linear Regression and K Nearest Neighbours have been used to classify the quality of wine and compare the accuracy. As a result, it is found that K Nearest Neighbours gives the good accuracy. The calculated Mean Squared Error (MSE) and calculated Root Mean Squared Error (RMSE) give the model perfection. Result shows that the value of MSE and RMSE applying K Nearest Neighbours (KNN) is higher than Multiple Linear Regression (MLR). The classification performance of the methods is compared with their accuracy. Based on these methods, the highest accuracy of KNN with K = 5 is 0.9444. Meanwhile, for the Multiple Linear Regression, the accuracy reached to 0.6657. Also, MSE and RMSE are calculated as 0.0555 and 0.2357 for KNN with k=5. Multiple Linear Regression has MSE (0.1692) and RMSE (0.4113). The experimental result shows that KNN can be used as alternative method for predicting the new instances. From UCI Machine Learning Repository, the wine dataset is taken which are tested in this research paper.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130698376","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The COVID-19 pandemic has compelled educational institutions worldwide to shift to remote online education. Addressing the growing trend, an Oxford University Press report titled ‘Addressing the Deepening Digital Divide’ states that poor digital access is the most significant barrier to digital learning according to 68 percent of academicians. Students in many remote parts of India frequently have access to limited bandwidth internet, which is insufficient for the modern standards of network-hogging online video conference software solutions. This paper provides an algorithmic compression of image buffers to aid low-cost remote online video education. This compression can be done by translating the teacher's blackboard images to pixel arrays projected on a canvas on the student's dashboard while the instructor constantly communicates via real-time voice. The image is first converted to grayscale and dilated with a square kernel. Using Harris Corner Detector, probable board corners are identified and compared to a geometrical center of the points and the corners recovered by cornerSubPix. An adaptive threshold is employed, distinguishing the board's contents from the backdrop on the cropped picture based on the recovered points. The pixel-mapped array is then transmitted to the students through the webRTC real-time protocol, which includes support for two-way audio, allowing the teacher to deliver lectures. Using Canvas API on the application front-end, the array is projected onto the student's device as a dot matrix display. This paper has achieved an effective rate in the video transmission format, aiding online remote education on low-bandwidth network devices.
{"title":"A Novel Pipeline for Compressing Image Buffers in Remote Education Video Conferencing using Harris Corner Detection and Pixel Map Array","authors":"Gita Alekhya Paul, Anshum Sharma, Yashvardhan Jagnani, Abhishek Saxena, P. Supraja","doi":"10.1109/ICECCT56650.2023.10179787","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179787","url":null,"abstract":"The COVID-19 pandemic has compelled educational institutions worldwide to shift to remote online education. Addressing the growing trend, an Oxford University Press report titled ‘Addressing the Deepening Digital Divide’ states that poor digital access is the most significant barrier to digital learning according to 68 percent of academicians. Students in many remote parts of India frequently have access to limited bandwidth internet, which is insufficient for the modern standards of network-hogging online video conference software solutions. This paper provides an algorithmic compression of image buffers to aid low-cost remote online video education. This compression can be done by translating the teacher's blackboard images to pixel arrays projected on a canvas on the student's dashboard while the instructor constantly communicates via real-time voice. The image is first converted to grayscale and dilated with a square kernel. Using Harris Corner Detector, probable board corners are identified and compared to a geometrical center of the points and the corners recovered by cornerSubPix. An adaptive threshold is employed, distinguishing the board's contents from the backdrop on the cropped picture based on the recovered points. The pixel-mapped array is then transmitted to the students through the webRTC real-time protocol, which includes support for two-way audio, allowing the teacher to deliver lectures. Using Canvas API on the application front-end, the array is projected onto the student's device as a dot matrix display. This paper has achieved an effective rate in the video transmission format, aiding online remote education on low-bandwidth network devices.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132671068","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179651
M. Venugopal, V. K. Sharma, Kalpana Sharma
The advancement of intelligent learning algorithms made the researchers to develop the generalized models that can handle heterogeneous data. With the post covid, different people are suffering from different type of diseases. Multi disease detection model is needed to prevent or to diagnosis various disease rather using different single detection platforms. In order to develop multi disease platform, the basic analysis lies in the gene structure of the human. All the existing detection systems find the disease based on either general characteristics or symptoms associated with the diseases. Symptoms based model may sometimes fail because of the thin difference between various diseases like continuous cough in case of covid as well as pneumonia or TB. So the proposed model collects the heterogeneous data associated with gene and predicts 8 multiple diseases using the enhanced MLP. Neural networks can handle heterogeneous data with less resources. When compared to the existing machine learning approaches, this model has achieved $+6.4%$ improvements in terms of accuracy.
{"title":"EMLPGENE: Enhanced MLP Gene Based Multi Disease Detection System Using Heterogeneous Data","authors":"M. Venugopal, V. K. Sharma, Kalpana Sharma","doi":"10.1109/ICECCT56650.2023.10179651","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179651","url":null,"abstract":"The advancement of intelligent learning algorithms made the researchers to develop the generalized models that can handle heterogeneous data. With the post covid, different people are suffering from different type of diseases. Multi disease detection model is needed to prevent or to diagnosis various disease rather using different single detection platforms. In order to develop multi disease platform, the basic analysis lies in the gene structure of the human. All the existing detection systems find the disease based on either general characteristics or symptoms associated with the diseases. Symptoms based model may sometimes fail because of the thin difference between various diseases like continuous cough in case of covid as well as pneumonia or TB. So the proposed model collects the heterogeneous data associated with gene and predicts 8 multiple diseases using the enhanced MLP. Neural networks can handle heterogeneous data with less resources. When compared to the existing machine learning approaches, this model has achieved $+6.4%$ improvements in terms of accuracy.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133180695","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 : 2023-02-22DOI: 10.1109/ICECCT56650.2023.10179827
R. K, D. N
The aim of the study is to detect SMS spam using Support Vector Machine (SVM) and linear regression (LR). The dataset used in the study contains 5573 sentences, and accuracy is measured for SMS spam detection. The classification process is carried out using SVM and LR with sample sizes of N=27, which were obtained using a G-power value of 80%. The accuracy of SVM is found to be 97.67%, which is higher than LR with an accuracy of 92%. The p-value for the significant accuracy difference is 0.02 (p<0.05), indicating that SVM performs better than LR in achieving accuracy.
{"title":"Accurate SMS Spam Detection Using Support Vector Machine In Comparison With Linear Regression","authors":"R. K, D. N","doi":"10.1109/ICECCT56650.2023.10179827","DOIUrl":"https://doi.org/10.1109/ICECCT56650.2023.10179827","url":null,"abstract":"The aim of the study is to detect SMS spam using Support Vector Machine (SVM) and linear regression (LR). The dataset used in the study contains 5573 sentences, and accuracy is measured for SMS spam detection. The classification process is carried out using SVM and LR with sample sizes of N=27, which were obtained using a G-power value of 80%. The accuracy of SVM is found to be 97.67%, which is higher than LR with an accuracy of 92%. The p-value for the significant accuracy difference is 0.02 (p<0.05), indicating that SVM performs better than LR in achieving accuracy.","PeriodicalId":180790,"journal":{"name":"2023 Fifth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127826276","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：