Hyperspectral images are remote sensing images containing more than a hundred spectral bands of the same ground space with various wavelengths. It has multiple applications but the random nature of latent data such as correlation, variability, and the number of spectral bands turned classification into a challenging task. These natures can be made to be less discriminatory by using a stand-alone preprocessing approach (dimensionality reduction techniques) with a classifier. A model performs poorly when redundant features are present and spatial-spectral concerns are ignored. A 2D Convolutional Neural Network (CNN) model is treated as a good method for hyperspectral image classification whereas accuracy depends on both spectral-spatial properties. Therefore, 3D CNN can be used as an alternative variant but has high computational complexity due to the large size of the volume and spectral dimension. A selective spectral-spatial HybridNET model that embeds dimensionality reduction and deep learning convolutional approaches are provided for both feature selection and extraction in order to solve these sorts of difficulties. In which both 3D and 2D convolutional networks have been adjusted to make a composite network with selective data preprocessors. Thus, this model is able to resolve time complexity issues as well as handle large amounts of data. Experiments have been performed using selective HybridNET on two available datasets such as Indian Pines and Pavia University, to confirm the stability of the proposed selective HybridNET over different state-of-the-art methods.
{"title":"Selective HybridNET: Spectral-Spatial Dimensionality Reduction for HSI Classification","authors":"Md. Rashedul Islam, Md. Touhid Islam, Md. Sohrawordi","doi":"10.1109/ECCE57851.2023.10101534","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101534","url":null,"abstract":"Hyperspectral images are remote sensing images containing more than a hundred spectral bands of the same ground space with various wavelengths. It has multiple applications but the random nature of latent data such as correlation, variability, and the number of spectral bands turned classification into a challenging task. These natures can be made to be less discriminatory by using a stand-alone preprocessing approach (dimensionality reduction techniques) with a classifier. A model performs poorly when redundant features are present and spatial-spectral concerns are ignored. A 2D Convolutional Neural Network (CNN) model is treated as a good method for hyperspectral image classification whereas accuracy depends on both spectral-spatial properties. Therefore, 3D CNN can be used as an alternative variant but has high computational complexity due to the large size of the volume and spectral dimension. A selective spectral-spatial HybridNET model that embeds dimensionality reduction and deep learning convolutional approaches are provided for both feature selection and extraction in order to solve these sorts of difficulties. In which both 3D and 2D convolutional networks have been adjusted to make a composite network with selective data preprocessors. Thus, this model is able to resolve time complexity issues as well as handle large amounts of data. Experiments have been performed using selective HybridNET on two available datasets such as Indian Pines and Pavia University, to confirm the stability of the proposed selective HybridNET over different state-of-the-art methods.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"10 36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126159185","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101626
Mosammat Israt Jahan, T. S. Sazzad, L. Armstrong
Breast tumor is considered as one of the most familiar tumors among women which cause breast cancer. Breast abrasion is observed as a thickened block of cells which forms tumor cell. In this paper, an improved and efficient breast tumor detection approach has been delineated using MRI images which not only provides faster detection but also has better accuracy compared to other existing available works. Numerous abrasion regions which are not considered as breast tumor surrounded by actual breast tumor causes processing issues and hence analysis and identification becomes challenging. To overcome under or over segmentation issues associated with breast tumor local histogram processing was incorporated. Additionally, instead of using conventional filtering approaches in this work mathematical morphological operation was incorporated followed by identification using shape and size features. The approach used in this study indicates an accuracy of 96.41% for conventional method and 96.67% for machine learning based model (CNN). Both approaches have been accepted by the experts' in the histopathology laboratory.
{"title":"Automated Breast Tumor Detection Using MRI Images","authors":"Mosammat Israt Jahan, T. S. Sazzad, L. Armstrong","doi":"10.1109/ECCE57851.2023.10101626","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101626","url":null,"abstract":"Breast tumor is considered as one of the most familiar tumors among women which cause breast cancer. Breast abrasion is observed as a thickened block of cells which forms tumor cell. In this paper, an improved and efficient breast tumor detection approach has been delineated using MRI images which not only provides faster detection but also has better accuracy compared to other existing available works. Numerous abrasion regions which are not considered as breast tumor surrounded by actual breast tumor causes processing issues and hence analysis and identification becomes challenging. To overcome under or over segmentation issues associated with breast tumor local histogram processing was incorporated. Additionally, instead of using conventional filtering approaches in this work mathematical morphological operation was incorporated followed by identification using shape and size features. The approach used in this study indicates an accuracy of 96.41% for conventional method and 96.67% for machine learning based model (CNN). Both approaches have been accepted by the experts' in the histopathology laboratory.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122296947","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101553
Nusrat Tasnim, S. Mamun
The mortality prediction model in the Intensive Care Unit (ICU) can be a great tool for assisting physicians in decision-making for the optimal allocation of ICU according to the patient's health conditions. Traditional scoring-based systems for mortality prediction don't provide good predictive performance in the case of a large dataset. Moreover, machine learning models can also provide poor performance for the lack of proper feature selection. A comparison of the performance of machine learning models with and without feature selection was explored in this study. Principal Component Analysis (PCA) was used to choose features for this investigation. For the classification job, the most widely used and diversified classifiers from the literature were used, including Logistic Regression(LR), Decision Tree (DT), K Nearest Neighbours (KNN), and Support Vector Machine (SVM). The Medical Information Mart for Intensive Care III (MIMIC-III) dataset was used to collect data on heart failure patients. On the MIMIC-III dataset, it was discovered that feature selection significantly improved the performance of the described machine learning models. Without feature selection, the accuracy of LR, DT, KNN, and SVM models was 86.66%, 80.12%, 85.13%, and 86.49%, respectively, however with PCA, the accuracy was improved to 88.0%, 80.46%, 86.83%, and 87.34%, respectively with only 5 principal components. Finally, the model's decision-making process was analyzed with explainable artificial intelligence using Local Interpretable Model-agnostic Explanations (LIME). This analysis can help to understand the feature's contribution to the model's prediction process. It was also observed that the features involved in the prediction process were mostly common with the first 15 features found in feature importance hierarchy.
重症监护病房(ICU)死亡率预测模型可以帮助医生根据患者的健康状况做出最佳的ICU配置决策。传统的基于评分的死亡率预测系统在大型数据集的情况下不能提供良好的预测性能。此外，由于缺乏适当的特征选择，机器学习模型也会提供较差的性能。本研究探讨了带特征选择和不带特征选择的机器学习模型的性能比较。本研究采用主成分分析(PCA)来选择特征。对于分类工作，使用了文献中最广泛使用和多样化的分类器，包括逻辑回归(LR)，决策树(DT)， K近邻(KNN)和支持向量机(SVM)。重症监护医学信息市场III (MIMIC-III)数据集用于收集心力衰竭患者的数据。在MIMIC-III数据集上，发现特征选择显著提高了所描述的机器学习模型的性能。在没有特征选择的情况下，LR、DT、KNN和SVM模型的准确率分别为86.66%、80.12%、85.13%和86.49%，而在只有5个主成分的情况下，PCA的准确率分别提高到88.0%、80.46%、86.83%和87.34%。最后，利用局部可解释模型不可知论解释(Local Interpretable model -agnostic Explanations, LIME)分析了可解释人工智能模型的决策过程。这种分析有助于理解特征对模型预测过程的贡献。我们还观察到，在预测过程中涉及的特征与特征重要性层次中发现的前15个特征最常见。
{"title":"Comparative Performance Analysis of Feature Selection for Mortality Prediction in ICU with Explainable Artificial Intelligence","authors":"Nusrat Tasnim, S. Mamun","doi":"10.1109/ECCE57851.2023.10101553","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101553","url":null,"abstract":"The mortality prediction model in the Intensive Care Unit (ICU) can be a great tool for assisting physicians in decision-making for the optimal allocation of ICU according to the patient's health conditions. Traditional scoring-based systems for mortality prediction don't provide good predictive performance in the case of a large dataset. Moreover, machine learning models can also provide poor performance for the lack of proper feature selection. A comparison of the performance of machine learning models with and without feature selection was explored in this study. Principal Component Analysis (PCA) was used to choose features for this investigation. For the classification job, the most widely used and diversified classifiers from the literature were used, including Logistic Regression(LR), Decision Tree (DT), K Nearest Neighbours (KNN), and Support Vector Machine (SVM). The Medical Information Mart for Intensive Care III (MIMIC-III) dataset was used to collect data on heart failure patients. On the MIMIC-III dataset, it was discovered that feature selection significantly improved the performance of the described machine learning models. Without feature selection, the accuracy of LR, DT, KNN, and SVM models was 86.66%, 80.12%, 85.13%, and 86.49%, respectively, however with PCA, the accuracy was improved to 88.0%, 80.46%, 86.83%, and 87.34%, respectively with only 5 principal components. Finally, the model's decision-making process was analyzed with explainable artificial intelligence using Local Interpretable Model-agnostic Explanations (LIME). This analysis can help to understand the feature's contribution to the model's prediction process. It was also observed that the features involved in the prediction process were mostly common with the first 15 features found in feature importance hierarchy.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122181177","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101523
Nahrin Jannat, S. M. Mahedy Hasan, Anwar Hossain Efat, Md Fakrul Taraque, Mostarina Mitu, Md. Al Mamun, Md. Farukuzzaman Faruk
Precise early detection of diseases can reduce the worsening and lethality, but it is not a spontaneous act to deal with complex medical data. Machine Learning (ML) can help the research community extensively in this aspect by playing a vast role in predicting the status of diseases at early stages. The study intended to develop a generalized model based on ML techniques that can classify frequently occurring diseases with better performance and reliability. In this research, four datasets collected from different repositories, such as the MRI and Alzheimer's Dataset (MAD), the SPECTF Heart Dataset (SHD), the Early Stage Diabetes Dataset (ESDD), and Lower Back Pain Dataset (LBPD), followed by analyzing and evaluating according to their performances to propose the prediction model. Numerous studies on this aspect conducted by others are available, but there is still scope for prosperity. To overcome the shortcomings of previous research, we have driven the first step with data preprocessing followed by six classification techniques such as Logistic regression (LR), Support Vector Machine (SVM), Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), and Extra tree (ET) are performed with 10-fold cross-validation as evaluation measure after assigning the best parameters manually by randomized search. In addition, the three best-performing classifiers (LR, RF, and SVM) are selected with their hyper-parameters to create an ensemble model through the stacking ensemble technique. After all, our generalized stacking ensemble model outperformed all other classifiers used in this study as well as other researchers in terms of accuracy that 96.97% in MAD, 95.08% in SHD, 98.90% in ESDD and 91.34% in LBPD are obtained.
{"title":"Stacking Ensemble Technique for Multiple Medical Datasets Classification: A Generalized Prediction Model","authors":"Nahrin Jannat, S. M. Mahedy Hasan, Anwar Hossain Efat, Md Fakrul Taraque, Mostarina Mitu, Md. Al Mamun, Md. Farukuzzaman Faruk","doi":"10.1109/ECCE57851.2023.10101523","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101523","url":null,"abstract":"Precise early detection of diseases can reduce the worsening and lethality, but it is not a spontaneous act to deal with complex medical data. Machine Learning (ML) can help the research community extensively in this aspect by playing a vast role in predicting the status of diseases at early stages. The study intended to develop a generalized model based on ML techniques that can classify frequently occurring diseases with better performance and reliability. In this research, four datasets collected from different repositories, such as the MRI and Alzheimer's Dataset (MAD), the SPECTF Heart Dataset (SHD), the Early Stage Diabetes Dataset (ESDD), and Lower Back Pain Dataset (LBPD), followed by analyzing and evaluating according to their performances to propose the prediction model. Numerous studies on this aspect conducted by others are available, but there is still scope for prosperity. To overcome the shortcomings of previous research, we have driven the first step with data preprocessing followed by six classification techniques such as Logistic regression (LR), Support Vector Machine (SVM), Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), and Extra tree (ET) are performed with 10-fold cross-validation as evaluation measure after assigning the best parameters manually by randomized search. In addition, the three best-performing classifiers (LR, RF, and SVM) are selected with their hyper-parameters to create an ensemble model through the stacking ensemble technique. After all, our generalized stacking ensemble model outperformed all other classifiers used in this study as well as other researchers in terms of accuracy that 96.97% in MAD, 95.08% in SHD, 98.90% in ESDD and 91.34% in LBPD are obtained.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128799969","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101550
Mst. Alema Khatun, M. Yousuf, M. Moni
This article describes a method to Human Activity Recognition (HAR) challenges based on data from wearable and smartphone sensors. We introduced a deep learning model and recognition system that is a combination of CNN (Convolutional Neural Network) and GRU (Gated Recurrent Unit) to improve results. Preferably, the data have been collected from several wearables as the participants go about their everyday activities. The convolutional neural network (CNN) deployed to improve the extraction of features at various scales. The derived attributes are then inserted into the gated recurrent unit (GRU), which labels features and enhances feature representation by understanding temporal connections. The CNN-GRU model uses a fully inte-grated (FC) layer, which is employed to hook up the feature maps with the classification standard. Three publicly accessible datasets, UCIHAR, OPPORTUNITY, and MHEALTH, were used to test the model's performance, with accuracy rates of 98.74%, 99.05%, and 99.53%, respectively. The outcomes show that the proposed model transcends some of the notified results in terms of activity detection.
{"title":"Deep CNN-GRU Based Human Activity Recognition with Automatic Feature Extraction Using Smartphone and Wearable Sensors","authors":"Mst. Alema Khatun, M. Yousuf, M. Moni","doi":"10.1109/ECCE57851.2023.10101550","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101550","url":null,"abstract":"This article describes a method to Human Activity Recognition (HAR) challenges based on data from wearable and smartphone sensors. We introduced a deep learning model and recognition system that is a combination of CNN (Convolutional Neural Network) and GRU (Gated Recurrent Unit) to improve results. Preferably, the data have been collected from several wearables as the participants go about their everyday activities. The convolutional neural network (CNN) deployed to improve the extraction of features at various scales. The derived attributes are then inserted into the gated recurrent unit (GRU), which labels features and enhances feature representation by understanding temporal connections. The CNN-GRU model uses a fully inte-grated (FC) layer, which is employed to hook up the feature maps with the classification standard. Three publicly accessible datasets, UCIHAR, OPPORTUNITY, and MHEALTH, were used to test the model's performance, with accuracy rates of 98.74%, 99.05%, and 99.53%, respectively. The outcomes show that the proposed model transcends some of the notified results in terms of activity detection.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114302755","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101610
Israt Zahan Nishu, M. F. Samad
Micromixers are devices with microchannels that effectively mix fluids across a tiny area and a limited flow route. One of the crucial components of a microfluidic system is a micromixer that should produce the maximum mixing with the smallest pressure drop. In this paper, a passive micromixer with kite-shaped chambers with rectangular bridges, and vortex-inducing inlets is proposed. The vertical separation of the fluid streams across the bridges and their recombination in the chambers has improved the mixing performance in here. The aim is to maximize the result by optimizing the design using the Taguchi approach and Grey Relational Analysis (GRA). Three factors, each with three level values, are employed in the Taguchi Design of Experiment, which produce a L9 orthogonal array with nine (9) trials. The optimal micromixer and the most influential parameter are derived from the analyses. COMSOL Multiphysics software is used to conduct the numerical simulation, which includes a Reynolds number range of 0.1 to 100. In its 5.8 mm length, the optimized micromixer produces 98% mixing and a maximum 9.8 kPa pressure drop. From the simulated analyses, it can be said that the proposed micromixer could be appropriate for the practical uses in the chemical and the biomedical sectors.
{"title":"Design and Optimization of a Passive Micromixer with Kite-Shaped Chambers","authors":"Israt Zahan Nishu, M. F. Samad","doi":"10.1109/ECCE57851.2023.10101610","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101610","url":null,"abstract":"Micromixers are devices with microchannels that effectively mix fluids across a tiny area and a limited flow route. One of the crucial components of a microfluidic system is a micromixer that should produce the maximum mixing with the smallest pressure drop. In this paper, a passive micromixer with kite-shaped chambers with rectangular bridges, and vortex-inducing inlets is proposed. The vertical separation of the fluid streams across the bridges and their recombination in the chambers has improved the mixing performance in here. The aim is to maximize the result by optimizing the design using the Taguchi approach and Grey Relational Analysis (GRA). Three factors, each with three level values, are employed in the Taguchi Design of Experiment, which produce a L9 orthogonal array with nine (9) trials. The optimal micromixer and the most influential parameter are derived from the analyses. COMSOL Multiphysics software is used to conduct the numerical simulation, which includes a Reynolds number range of 0.1 to 100. In its 5.8 mm length, the optimized micromixer produces 98% mixing and a maximum 9.8 kPa pressure drop. From the simulated analyses, it can be said that the proposed micromixer could be appropriate for the practical uses in the chemical and the biomedical sectors.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126359262","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101567
M. Ferdous, Rifat Shahriyar
A critical, sometimes fatal medical disease called a stroke happens when the blood flow to a portion of the brain is broken off. In the case of stroke, urgent treatment is very essential. Nowadays, stroke is the main cause of death and impairment globally, according to WHO. In this situation, it will be very helpful if we predict the probability of stroke earlier depending on some most important features. Many researchers use different machine learning algorithms for prediction but very few researchers use stacking methods and CNN. The main contribution of this paper is to develop a stacking classifier of ensemble methods and the CNN model. In this paper, data-set is collected from Kaggle. Stroke data is imbalanced. Random oversampling is used for balancing data-set. Then most important features are find out using feature selection method, then applying different machine learning algorithms such as Logistic Regression, Decision Tree Classifier, Support vector machine, Random forest classifier, KNearest neighbour's classifier, Bernoulli Naïve Bayes, Gaussian Naïve Bayes, Stacking of six algorithms (Decision Tree Classifier, Support vector machine, Random forest classifier, KNearest neighbor's classifier, Bernoulli Naïve Bayes, Gaussian Naïve Bayes) and CNN. Then comparing the performances for predicting the probability of stroke during both the training and testing periods. Results show that the Stacking of six algorithms gives the highest accuracy, which is 99.89% for testing and 100% for training.
{"title":"A Comparative Analysis for Stroke Risk Prediction Using Machine Learning Algorithms and Convolutional Neural Network Model","authors":"M. Ferdous, Rifat Shahriyar","doi":"10.1109/ECCE57851.2023.10101567","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101567","url":null,"abstract":"A critical, sometimes fatal medical disease called a stroke happens when the blood flow to a portion of the brain is broken off. In the case of stroke, urgent treatment is very essential. Nowadays, stroke is the main cause of death and impairment globally, according to WHO. In this situation, it will be very helpful if we predict the probability of stroke earlier depending on some most important features. Many researchers use different machine learning algorithms for prediction but very few researchers use stacking methods and CNN. The main contribution of this paper is to develop a stacking classifier of ensemble methods and the CNN model. In this paper, data-set is collected from Kaggle. Stroke data is imbalanced. Random oversampling is used for balancing data-set. Then most important features are find out using feature selection method, then applying different machine learning algorithms such as Logistic Regression, Decision Tree Classifier, Support vector machine, Random forest classifier, KNearest neighbour's classifier, Bernoulli Naïve Bayes, Gaussian Naïve Bayes, Stacking of six algorithms (Decision Tree Classifier, Support vector machine, Random forest classifier, KNearest neighbor's classifier, Bernoulli Naïve Bayes, Gaussian Naïve Bayes) and CNN. Then comparing the performances for predicting the probability of stroke during both the training and testing periods. Results show that the Stacking of six algorithms gives the highest accuracy, which is 99.89% for testing and 100% for training.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126399192","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101566
Md. Sifat Hasan, Md. Safaiat Hossain, Md. Rifat Hayder
In this modern era of technological advancement, the demand for electricity is raising day by day. To cope with the increasing demand for electricity the demand for generation is also raising in the same manner. As the traditional power generation resources are overpriced and most of the resources need to be imported, it is a notable challenge for developing countries like Bangladesh to generate electricity at a low cost. For this reason, it is required to find a noble alternative way of energy generation to fulfill the population demand which is more environment friendly, as well as cost-efficient. Among the energy resources, Waste to Electrical Energy (WTEE) has a bright future in Bangladesh (as these wastes locked a bulk amount of energy). This paper mainly focuses on Electricity generation using Municipal solid waste (MSW) via the incineration process and also a possible solution to avoid the drawbacks of the traditional power generation technique. Enough electricity could be generated from the waste throughout the process, aiding in waste management and helping to meet Bangladesh's yearly power requirements.
{"title":"Converting Municipal Solid Waste into Electrical Energy: A Renewable Solution in Bangladesh","authors":"Md. Sifat Hasan, Md. Safaiat Hossain, Md. Rifat Hayder","doi":"10.1109/ECCE57851.2023.10101566","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101566","url":null,"abstract":"In this modern era of technological advancement, the demand for electricity is raising day by day. To cope with the increasing demand for electricity the demand for generation is also raising in the same manner. As the traditional power generation resources are overpriced and most of the resources need to be imported, it is a notable challenge for developing countries like Bangladesh to generate electricity at a low cost. For this reason, it is required to find a noble alternative way of energy generation to fulfill the population demand which is more environment friendly, as well as cost-efficient. Among the energy resources, Waste to Electrical Energy (WTEE) has a bright future in Bangladesh (as these wastes locked a bulk amount of energy). This paper mainly focuses on Electricity generation using Municipal solid waste (MSW) via the incineration process and also a possible solution to avoid the drawbacks of the traditional power generation technique. Enough electricity could be generated from the waste throughout the process, aiding in waste management and helping to meet Bangladesh's yearly power requirements.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126555074","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10101535
Shantanu Bhattacharjee, Md Belal Uddin Sifat, Jayeed Bin Kibria, N. S. Pathan, Nur Mohammad
Bengali Handwritten Digit Recognition (BHDR) has extensive applications in OCR, voting machines, postal mail sorting, security systems, robotics, and many other fields. BHDR can be performed using various popular machine learning models and deep neural network architectures among which Spiking Neural Network (SNN) is getting increasing attention in recent works. SNN is an emerging machine learning model which mimics the natural processing mechanism of actual neurons of the brain. In this paper, SNN is applied for the recognition of Bangla Handwritten Digits using a popular dataset called ‘NumtaDB’. The images have been brought through various preprocessing operations for the SNN model so that it could better interpret the digits. The performance is analyzed for different values of the parameters of SNN. By systematically changing the parameters, the best combination of the values is selected for getting optimal accuracy. The model gives an accuracy of 91.36% with a comparatively faster training time using fewer computational resources relative to other machine learning models.
{"title":"Recognition of Bengali Handwritten Digits Using Spiking Neural Network Architecture","authors":"Shantanu Bhattacharjee, Md Belal Uddin Sifat, Jayeed Bin Kibria, N. S. Pathan, Nur Mohammad","doi":"10.1109/ECCE57851.2023.10101535","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10101535","url":null,"abstract":"Bengali Handwritten Digit Recognition (BHDR) has extensive applications in OCR, voting machines, postal mail sorting, security systems, robotics, and many other fields. BHDR can be performed using various popular machine learning models and deep neural network architectures among which Spiking Neural Network (SNN) is getting increasing attention in recent works. SNN is an emerging machine learning model which mimics the natural processing mechanism of actual neurons of the brain. In this paper, SNN is applied for the recognition of Bangla Handwritten Digits using a popular dataset called ‘NumtaDB’. The images have been brought through various preprocessing operations for the SNN model so that it could better interpret the digits. The performance is analyzed for different values of the parameters of SNN. By systematically changing the parameters, the best combination of the values is selected for getting optimal accuracy. The model gives an accuracy of 91.36% with a comparatively faster training time using fewer computational resources relative to other machine learning models.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131984973","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 : 2023-02-23DOI: 10.1109/ECCE57851.2023.10100746
Shaira Senjuti Oyshee, Shaharehar Rahaman Anik, Mohammad Jawad Ul Kabir Chowdhury, Md. Ahsan Kabir
With a continual rise in electricity prices, the integration of demand-side management (DSM) techniques such as peak load shifting and load behavior patterns with machine learning-based solutions has become a necessity in modern grid management. In this paper, using the Pearson correlation coefficient (PCC), the correlation between a synthesized dataset consisting of load consumption data from the national grid of Bangladesh, and meteorological data, such as maximum and minimum temperature has been calculated, which resulted in values of 0.84, 0.87 and 0.89, respectively. Next, the yearly load data have been clustered using a k-Means clustering algorithm, to find consumption patterns, and using the labels, temperature ranges are clustered to suggest temperature dependence, in accordance with consumption patterns. Finally, for every cluster, using a hypothetical set of percentages, a peak shaving, and load shifting algorithm has been implemented to show hypothetical approximates of load shifting potential for every year, which produced percentages 8.83, 9.07, and 8.79 for the years 2018, 2019 and 2021, respectively.
{"title":"Machine Learning based Load and Temperature Behavior Clustering and Peak Shifting Implementation on Bangladeshi Grid Data","authors":"Shaira Senjuti Oyshee, Shaharehar Rahaman Anik, Mohammad Jawad Ul Kabir Chowdhury, Md. Ahsan Kabir","doi":"10.1109/ECCE57851.2023.10100746","DOIUrl":"https://doi.org/10.1109/ECCE57851.2023.10100746","url":null,"abstract":"With a continual rise in electricity prices, the integration of demand-side management (DSM) techniques such as peak load shifting and load behavior patterns with machine learning-based solutions has become a necessity in modern grid management. In this paper, using the Pearson correlation coefficient (PCC), the correlation between a synthesized dataset consisting of load consumption data from the national grid of Bangladesh, and meteorological data, such as maximum and minimum temperature has been calculated, which resulted in values of 0.84, 0.87 and 0.89, respectively. Next, the yearly load data have been clustered using a k-Means clustering algorithm, to find consumption patterns, and using the labels, temperature ranges are clustered to suggest temperature dependence, in accordance with consumption patterns. Finally, for every cluster, using a hypothetical set of percentages, a peak shaving, and load shifting algorithm has been implemented to show hypothetical approximates of load shifting potential for every year, which produced percentages 8.83, 9.07, and 8.79 for the years 2018, 2019 and 2021, respectively.","PeriodicalId":131537,"journal":{"name":"2023 International Conference on Electrical, Computer and Communication Engineering (ECCE)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133870688","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：