Pub Date : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10125698
Nakayiza Hellen, Ggaliwango Marvin
The ongoing digital revolution is having a significant impact on homes and communities worldwide, affecting access to information, communication, learning, and sports. One of the most significant changes brought about by this revolution is the shift from traditional classroom-based education to virtual and hybrid online learning environments. Higher education institutions, in particular, are recognizing the value of online educational programs, which allow them to expand their digital pre se n ce, increase access to their programs, and reach students beyond their physical borders. The advancements in educational technology made possible by the 4th Industrial Revolution are also allowing for more flexible, engaging, and accessible learning experiences for both students and teachers. However, there remains a significant gap in terms of education planning, access to digital learning tools, and engagement among stakeholders. This research uses data analytics to examine cl oud-based digital learning tools, education stakeholder engagement, and education access. The findings provide insight for academic stakeholders, particularly governments, private sector, and educational investors, on ways to bridge the gaps between access and engagement for students and teachers.
{"title":"Learning Analytics for Cloud-based Education Planning","authors":"Nakayiza Hellen, Ggaliwango Marvin","doi":"10.1109/ICOEI56765.2023.10125698","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125698","url":null,"abstract":"The ongoing digital revolution is having a significant impact on homes and communities worldwide, affecting access to information, communication, learning, and sports. One of the most significant changes brought about by this revolution is the shift from traditional classroom-based education to virtual and hybrid online learning environments. Higher education institutions, in particular, are recognizing the value of online educational programs, which allow them to expand their digital pre se n ce, increase access to their programs, and reach students beyond their physical borders. The advancements in educational technology made possible by the 4th Industrial Revolution are also allowing for more flexible, engaging, and accessible learning experiences for both students and teachers. However, there remains a significant gap in terms of education planning, access to digital learning tools, and engagement among stakeholders. This research uses data analytics to examine cl oud-based digital learning tools, education stakeholder engagement, and education access. The findings provide insight for academic stakeholders, particularly governments, private sector, and educational investors, on ways to bridge the gaps between access and engagement for students and teachers.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114762025","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10126052
A. Palaniammal, P. Anandababu
Sarcasm is a procedure of verbal irony that is planned to convey ridicule, contempt or mockery with the aid of words that expresses the opposite of what is meant or through facial expression, tone of voice, or inflection. In another word, it is a way of saying something but meaning the opposite, often intending to be critical or humorous. Sarcasm is widely applied in social media, humour, and casual conversation. Sarcasm detection using deep learning (DL) includes training a machine learning (ML) algorithm for identifying instances of sarcasm and recognizing the pattern in language. The study presents a new Chaos Sine Cosine Algorithm with Graph Convolution Network for Sarcasm Detection (CSCA-GCNSD) technique in Social Media. The presented CSCA-GCNSD technique aims to recognize and categorize various kinds of sarcasm. Primarily, the CSCA-GCNSD technique involves different stages of data pre-processing. Next, the CSCA-GCNSD technique applies the GCN model for the detection and classification of various kinds of sarcasm. Finally, the CSCA technique is used to optimally choose the hyperparameter values of the GCN model and thereby resulting in improved detection outcomes. The simulation outcomes of the CSCA-GCNSD methodology was tested on different sarcasm datasets and the outcomes reported the betterment of the CSCA-GCNSD algorithms over other models.
{"title":"Chaos Sine Cosine Algorithm with Graph Convolution Network for Sarcasm Detection in Social Media","authors":"A. Palaniammal, P. Anandababu","doi":"10.1109/ICOEI56765.2023.10126052","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10126052","url":null,"abstract":"Sarcasm is a procedure of verbal irony that is planned to convey ridicule, contempt or mockery with the aid of words that expresses the opposite of what is meant or through facial expression, tone of voice, or inflection. In another word, it is a way of saying something but meaning the opposite, often intending to be critical or humorous. Sarcasm is widely applied in social media, humour, and casual conversation. Sarcasm detection using deep learning (DL) includes training a machine learning (ML) algorithm for identifying instances of sarcasm and recognizing the pattern in language. The study presents a new Chaos Sine Cosine Algorithm with Graph Convolution Network for Sarcasm Detection (CSCA-GCNSD) technique in Social Media. The presented CSCA-GCNSD technique aims to recognize and categorize various kinds of sarcasm. Primarily, the CSCA-GCNSD technique involves different stages of data pre-processing. Next, the CSCA-GCNSD technique applies the GCN model for the detection and classification of various kinds of sarcasm. Finally, the CSCA technique is used to optimally choose the hyperparameter values of the GCN model and thereby resulting in improved detection outcomes. The simulation outcomes of the CSCA-GCNSD methodology was tested on different sarcasm datasets and the outcomes reported the betterment of the CSCA-GCNSD algorithms over other models.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"681 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116108508","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10126055
Parveen Akhther. A, A. Maryposonia, P. S.
The task to ensure security in a network that is distributed over several nodes is a significant and challenging one. Since the primary objective of a DDoS attack is to prevent authorized nodes from gaining access to the service, this type of attack presents a significant threat to distributed networks. It is highly important that a modular and dependable NIDS must be created for handling DOS attacks in the distributed environment effectively, and in turn, all the nodes are available in the distributed network.The high need for modular techniques required in the detection phase for collecting, storing and analyzing the big data from the nodes in the distributed network poses significant hurdles in finding out the Distributed DOS attack.This research proposes a Big Data-based Distributed Denial of Service Network Intrusion Detection System to address these issues. Important features of the proposed intrusion detection system include a module for detecting network traffic and another for collecting data on that traffic. In this study, micro-batch data processing is employed for traffic feature gathering in the Network collection module and Random Forest (RF) algorithm-based classification technique is used in the traffic detection module for feature selection. For Storing a large number of wary attacks, Hadoop File System (HDFS) is used, and for accelerating the speed of data processing, S park is used as a suggested solution.The method was assessed using the NSL-KDD benchmark dataset to find the accuracy and many other parameters. Experimental results for Accuracy, Recall, F1-Measure and Precision, from the proposed work are compared to those from the machine learning techniques, DT(Decision Tree), PCA RF(Principal Component Analysis Random Forest), NB(Naive Bayes), SVM(Support Vector Machine), and LR (Logistic Regression). According to the experimental findings, the suggested detection algorithm achieved an Accuracy of 99.89%, respectively.
{"title":"An Efficient Network Intrusion Detection System for Distributed Networks using Machine Learning Technique","authors":"Parveen Akhther. A, A. Maryposonia, P. S.","doi":"10.1109/ICOEI56765.2023.10126055","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10126055","url":null,"abstract":"The task to ensure security in a network that is distributed over several nodes is a significant and challenging one. Since the primary objective of a DDoS attack is to prevent authorized nodes from gaining access to the service, this type of attack presents a significant threat to distributed networks. It is highly important that a modular and dependable NIDS must be created for handling DOS attacks in the distributed environment effectively, and in turn, all the nodes are available in the distributed network.The high need for modular techniques required in the detection phase for collecting, storing and analyzing the big data from the nodes in the distributed network poses significant hurdles in finding out the Distributed DOS attack.This research proposes a Big Data-based Distributed Denial of Service Network Intrusion Detection System to address these issues. Important features of the proposed intrusion detection system include a module for detecting network traffic and another for collecting data on that traffic. In this study, micro-batch data processing is employed for traffic feature gathering in the Network collection module and Random Forest (RF) algorithm-based classification technique is used in the traffic detection module for feature selection. For Storing a large number of wary attacks, Hadoop File System (HDFS) is used, and for accelerating the speed of data processing, S park is used as a suggested solution.The method was assessed using the NSL-KDD benchmark dataset to find the accuracy and many other parameters. Experimental results for Accuracy, Recall, F1-Measure and Precision, from the proposed work are compared to those from the machine learning techniques, DT(Decision Tree), PCA RF(Principal Component Analysis Random Forest), NB(Naive Bayes), SVM(Support Vector Machine), and LR (Logistic Regression). According to the experimental findings, the suggested detection algorithm achieved an Accuracy of 99.89%, respectively.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123173835","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10126051
Sowmen Mitra, P. Kanungoe
Recognition of human activities is essential for many applications, and the widespread availability of low-cost sensors on smartphones and wearables has enabled the development of mobile apps capable of tracking user activities “in the wild.” However, dealing with heterogeneous data from different devices and real-time scenarios presents significant challenges. In this study, a novel learning framework is proposed for Human Activity Recognition (HAR) that combines a Convolutional Neural Network (CNN) with an autoencoder for feature extraction. The study also investigates the importance of preprocessing techniques, including orientation-independent transformation, to mitigate heterogeneity when dealing with multiple types of smartphones. The results show that the proposed approach outperforms state-of-the-art methods in HAR, with an accuracy of 95.74% on the heterogeneous dataset used in this study. Furthermore, the study demonstrates that proposed framework can be effectively deployed on smartphones with limited computational resources, making it suitable for real-world applications.
{"title":"Smartphone based Human Activity Recognition using CNNs and Autoencoder Features","authors":"Sowmen Mitra, P. Kanungoe","doi":"10.1109/ICOEI56765.2023.10126051","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10126051","url":null,"abstract":"Recognition of human activities is essential for many applications, and the widespread availability of low-cost sensors on smartphones and wearables has enabled the development of mobile apps capable of tracking user activities “in the wild.” However, dealing with heterogeneous data from different devices and real-time scenarios presents significant challenges. In this study, a novel learning framework is proposed for Human Activity Recognition (HAR) that combines a Convolutional Neural Network (CNN) with an autoencoder for feature extraction. The study also investigates the importance of preprocessing techniques, including orientation-independent transformation, to mitigate heterogeneity when dealing with multiple types of smartphones. The results show that the proposed approach outperforms state-of-the-art methods in HAR, with an accuracy of 95.74% on the heterogeneous dataset used in this study. Furthermore, the study demonstrates that proposed framework can be effectively deployed on smartphones with limited computational resources, making it suitable for real-world applications.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123684391","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10125652
Phanitha Sai Lakshmi Veeranki, Gaja Lakshmi Banavath, P. R. Devi
According to statistics from WHO, brain tumors will account for roughly 9.5 million deaths globally in the next few decades. Early identification and treatment are the best ways to stop deaths from brain cancer. Brain tumors fall into two categories: benign, which is not cancerous, and malignant, which is cancerous. A brain tumor that originates in a specific location and then metastasizes to other regions of the body, including other areas of the brain, is referred to as a primary tumor. Secondary tumors, commonly referred to as metastatic tumors, arise from primary tumors. It is now possible to more easily analyze medical pictures thanks to the quick development of image processing and soft computing technologies that aid in early detection and therapy. The use of computer-aided diagnostic (CAD) technology for diagnosing illnesses, predicting prognoses, and determining the likelihood of recurrence is expanding as a result of technological improvements. The main area of investigation in this study is the utilization of feature extraction and tumor cell classification for the automatic identification and categorization of brain tumors in magnetic resonance imaging (MRI) scans. Brain tumor detection and classification are done using CNN, and VGG-16 models. Accuracy is obtained by doing a comparative study of these two models. VGG-16 is the best-trained model.
{"title":"Detection and Classification of Brain Tumors using Convolutional Neural Network","authors":"Phanitha Sai Lakshmi Veeranki, Gaja Lakshmi Banavath, P. R. Devi","doi":"10.1109/ICOEI56765.2023.10125652","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125652","url":null,"abstract":"According to statistics from WHO, brain tumors will account for roughly 9.5 million deaths globally in the next few decades. Early identification and treatment are the best ways to stop deaths from brain cancer. Brain tumors fall into two categories: benign, which is not cancerous, and malignant, which is cancerous. A brain tumor that originates in a specific location and then metastasizes to other regions of the body, including other areas of the brain, is referred to as a primary tumor. Secondary tumors, commonly referred to as metastatic tumors, arise from primary tumors. It is now possible to more easily analyze medical pictures thanks to the quick development of image processing and soft computing technologies that aid in early detection and therapy. The use of computer-aided diagnostic (CAD) technology for diagnosing illnesses, predicting prognoses, and determining the likelihood of recurrence is expanding as a result of technological improvements. The main area of investigation in this study is the utilization of feature extraction and tumor cell classification for the automatic identification and categorization of brain tumors in magnetic resonance imaging (MRI) scans. Brain tumor detection and classification are done using CNN, and VGG-16 models. Accuracy is obtained by doing a comparative study of these two models. VGG-16 is the best-trained model.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121711147","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10125708
S. Ragul, Y. S, V. Vijayabalan, B. Venkatasamy, L. Kalaivani, F. A. Jeffrey Vaz
This work reviews the design of controlling the ceiling Fan speed by using android mobile. This product is to provide comfortable sleeping at midnight during the winter and rainy seasons. It automatically controls the speed of the Fan based on the factors such as temperature and humidity. The speed control can be fully automatic or semi-automatic. The proposed smart controller is implemented between Fan and E.B. mains. There is no need to disturb the existing Fan arrangement. The main aim of this study is to replace the existing Fan regulator alone. The mode of operation of the Fan can be controlled by using IoT/Bluetooth/Manual. PIR sensor is incorporated; it makes the Fan run only when the people are present inside the room. This study also includes a night visible digital clock and wakeup alarm system. The proposed controller is made as two variants; one is to control a single ceiling fan with additional features that make a complete bedroom solution cost-effective. Another is a high-power controller that controls a large number of Fans based on the factors, which conserves a lot of energy.
{"title":"IoT based Smart Controller for Ceiling Fan","authors":"S. Ragul, Y. S, V. Vijayabalan, B. Venkatasamy, L. Kalaivani, F. A. Jeffrey Vaz","doi":"10.1109/ICOEI56765.2023.10125708","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125708","url":null,"abstract":"This work reviews the design of controlling the ceiling Fan speed by using android mobile. This product is to provide comfortable sleeping at midnight during the winter and rainy seasons. It automatically controls the speed of the Fan based on the factors such as temperature and humidity. The speed control can be fully automatic or semi-automatic. The proposed smart controller is implemented between Fan and E.B. mains. There is no need to disturb the existing Fan arrangement. The main aim of this study is to replace the existing Fan regulator alone. The mode of operation of the Fan can be controlled by using IoT/Bluetooth/Manual. PIR sensor is incorporated; it makes the Fan run only when the people are present inside the room. This study also includes a night visible digital clock and wakeup alarm system. The proposed controller is made as two variants; one is to control a single ceiling fan with additional features that make a complete bedroom solution cost-effective. Another is a high-power controller that controls a large number of Fans based on the factors, which conserves a lot of energy.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122763630","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 use of digital photographs has increased along with the development of digital technologies. Due to the vast amounts of information it contains, digital photographs need a lot of storage space, as well as bigger transmission bandwidths and longer transmission times. Therefore, on compressing the images all the redundant bits of information present in the image under test are removed while keeping only the essential information needed to reconstruct the image later on. In this study, DWT-SPIHT technique is introduced, which may be used to compress and reconstruct images at various degrees of wavelet decomposition across wavelet families that were initially a subdivision of the MATLAB wavelet family. Simulations have been conducted on Cameraman Image during this work of different resolution at different levels of decomposition and for different types of thresholding techniques to prove that this algorithm works well and provide us with the good reconstruction quality of the image. The simulation results demonstrate that, when compared to the DWT-EZW algorithm, the proposed DWT-SPIHT algorithm performs significantly better in terms of evaluation parameters like peak signal to noise ratio (PSNR), mean square error (MSE), and visual perception at higher compression ratios (CR) and low bit per pixel values (BPP).
{"title":"Comparative Study and Analysis of DWT-SPIHT with DWT-EZW Method for Image Compression","authors":"Dikendra K. Verma, Garima Singh, Saurabh Pargaien, Purushottam Das, Sashank Chaube, Upendra Bhatt","doi":"10.1109/ICOEI56765.2023.10125692","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125692","url":null,"abstract":"The use of digital photographs has increased along with the development of digital technologies. Due to the vast amounts of information it contains, digital photographs need a lot of storage space, as well as bigger transmission bandwidths and longer transmission times. Therefore, on compressing the images all the redundant bits of information present in the image under test are removed while keeping only the essential information needed to reconstruct the image later on. In this study, DWT-SPIHT technique is introduced, which may be used to compress and reconstruct images at various degrees of wavelet decomposition across wavelet families that were initially a subdivision of the MATLAB wavelet family. Simulations have been conducted on Cameraman Image during this work of different resolution at different levels of decomposition and for different types of thresholding techniques to prove that this algorithm works well and provide us with the good reconstruction quality of the image. The simulation results demonstrate that, when compared to the DWT-EZW algorithm, the proposed DWT-SPIHT algorithm performs significantly better in terms of evaluation parameters like peak signal to noise ratio (PSNR), mean square error (MSE), and visual perception at higher compression ratios (CR) and low bit per pixel values (BPP).","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122932001","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}
Multinational companies are taking advantage of the services provided through a cloud service provider (CS P). It is generally observed that, the companies provide customized services to the user as an added benefit rather than using the initial services. The motive of the proposed study is to build a trusted relationship between the user and service provider. This study analyzes several parameters to scale up an approach by adopting advanced technologies. In this study, a matrix has been prepared by including the utility value for the fruit factors used by the user. Compatibility connection between multiple customers are also measured by obtaining services from the particular company. The proposed matrix can identify the actual use of the significant cloud computing features.
{"title":"A Novel Matrix for Analyzing Cloud Services in Top MNCs","authors":"Nidhi Bansal, Archana Jain, Manoj kumar Sharma, Manish Kumar","doi":"10.1109/ICOEI56765.2023.10125916","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125916","url":null,"abstract":"Multinational companies are taking advantage of the services provided through a cloud service provider (CS P). It is generally observed that, the companies provide customized services to the user as an added benefit rather than using the initial services. The motive of the proposed study is to build a trusted relationship between the user and service provider. This study analyzes several parameters to scale up an approach by adopting advanced technologies. In this study, a matrix has been prepared by including the utility value for the fruit factors used by the user. Compatibility connection between multiple customers are also measured by obtaining services from the particular company. The proposed matrix can identify the actual use of the significant cloud computing features.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125283554","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 : 2023-04-11DOI: 10.1109/ICOEI56765.2023.10125786
Rishabh Kothari, B. Rangwala, Kush Patel
One of the major challenges during online examinations is the assessment of answers, particularly of the subjective type. Subjective answers test a student's ability to retain information and express it in natural language. While objective questions have a correct fixed answer, subjective questions can have multiple correct answers. These answers can convey the same information while using a completely different language and grammatical syntax. This makes it difficult to automate the process of grading subjective questions and requires a lot of manual work hours. This study intends to automate the process of grading subjective questions using Machine Learning (ML) and Natural Language Processing (NLP). The study has compared the subjective answer with an ideal answer that is provided by the authority that creates the question. Based on the similarity between the two answers, a score is generated which can be mapped to an appropriate grade. The authors have provided a web application made using the Django framework for people to give online examinations and be automatically graded in near real-time. No machine learning model can be 100% accurate, so there is a functionality for admins to edit the grades.
{"title":"Automatic Subjective Answer Grading Software Using Machine Learning","authors":"Rishabh Kothari, B. Rangwala, Kush Patel","doi":"10.1109/ICOEI56765.2023.10125786","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125786","url":null,"abstract":"One of the major challenges during online examinations is the assessment of answers, particularly of the subjective type. Subjective answers test a student's ability to retain information and express it in natural language. While objective questions have a correct fixed answer, subjective questions can have multiple correct answers. These answers can convey the same information while using a completely different language and grammatical syntax. This makes it difficult to automate the process of grading subjective questions and requires a lot of manual work hours. This study intends to automate the process of grading subjective questions using Machine Learning (ML) and Natural Language Processing (NLP). The study has compared the subjective answer with an ideal answer that is provided by the authority that creates the question. Based on the similarity between the two answers, a score is generated which can be mapped to an appropriate grade. The authors have provided a web application made using the Django framework for people to give online examinations and be automatically graded in near real-time. No machine learning model can be 100% accurate, so there is a functionality for admins to edit the grades.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125798625","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}
This study proposes a unique segmentation-based clustering algorithm that utilises k-means for segmentation, further uses a local feature selection technique to obtain the top bands for each cluster and deploys clustering on segmented hyperspectral imagery. The suggested methodology is a framework with several stages. k-means is initially utilized for image segmentation. From the obtained segments, significant segments are identified using Gini impurity. Finally, the cluster map is obtained by merging insignificant clusters with significant clusters. This step also makes use of novel local feature selection strategy. Three sets of hyperspectral images are used in investigations to evaluate the efficiency of the proposed methodology. For assessment, the criteria Normalized Mutual Information and Purity score are utilised. The investigation findings demonstrate that the proposed methodology outperforms the other segmentation methodologies that were compared. According to the results, using band selection and redundancy strategies significantly improves accuracy.
{"title":"Hyper Spectral Image Clustering and Local Feature Selection using Gini Impurity","authors":"Prashant Kumar Mali, Hitenkumar Motiyani, Quazi Sameed, Anand Mehta","doi":"10.1109/ICOEI56765.2023.10125605","DOIUrl":"https://doi.org/10.1109/ICOEI56765.2023.10125605","url":null,"abstract":"This study proposes a unique segmentation-based clustering algorithm that utilises k-means for segmentation, further uses a local feature selection technique to obtain the top bands for each cluster and deploys clustering on segmented hyperspectral imagery. The suggested methodology is a framework with several stages. k-means is initially utilized for image segmentation. From the obtained segments, significant segments are identified using Gini impurity. Finally, the cluster map is obtained by merging insignificant clusters with significant clusters. This step also makes use of novel local feature selection strategy. Three sets of hyperspectral images are used in investigations to evaluate the efficiency of the proposed methodology. For assessment, the criteria Normalized Mutual Information and Purity score are utilised. The investigation findings demonstrate that the proposed methodology outperforms the other segmentation methodologies that were compared. According to the results, using band selection and redundancy strategies significantly improves accuracy.","PeriodicalId":168942,"journal":{"name":"2023 7th International Conference on Trends in Electronics and Informatics (ICOEI)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129776219","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：