Pub Date : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00060
I. Dey, Vibhav Pratap
With the advent of machine learning and its numerous techniques, many real-world problems have been solved like credit card fraud detection, cancer susceptibility and survival prediction, identification of spam, and customer segmentation, to name a few. Machine learning works on huge loads of data to give the correct prediction and maximum accuracy. Now, accuracy of any machine learning model depends on the dataset been fed into that model, in the first place. And from here comes the concept of oversampling and under-sampling. Under-sampling is the process of shortening the majority class or deleting samples from the majority class in order to balance the dataset, and over-sampling is the process of adding additional synthetic samples to the minority class. So, this study is based on the three methods namely, SMOTE, Borderline-SMOTE, and ADASYN. This study includes the collation of the above-mentioned oversampling techniques based on their accuracy, precision, recall, F1-measure and ROC curve.
{"title":"A Comparative Study of SMOTE, Borderline-SMOTE, and ADASYN Oversampling Techniques using Different Classifiers","authors":"I. Dey, Vibhav Pratap","doi":"10.1109/ICSMDI57622.2023.00060","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00060","url":null,"abstract":"With the advent of machine learning and its numerous techniques, many real-world problems have been solved like credit card fraud detection, cancer susceptibility and survival prediction, identification of spam, and customer segmentation, to name a few. Machine learning works on huge loads of data to give the correct prediction and maximum accuracy. Now, accuracy of any machine learning model depends on the dataset been fed into that model, in the first place. And from here comes the concept of oversampling and under-sampling. Under-sampling is the process of shortening the majority class or deleting samples from the majority class in order to balance the dataset, and over-sampling is the process of adding additional synthetic samples to the minority class. So, this study is based on the three methods namely, SMOTE, Borderline-SMOTE, and ADASYN. This study includes the collation of the above-mentioned oversampling techniques based on their accuracy, precision, recall, F1-measure and ROC curve.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134137678","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00037
Yuejuan Wang
Among the mainstream methods of image aesthetic quality evaluation, it can be divided into traditional aesthetic evaluation methods based on the artificial design features and current popular aesthetic evaluation methods based on deep learning. Hence, this paper studies the intelligentization of art design system based on the multidimensional visual image reconstruction algorithm. In the proposed study, the modelling process contains the 2 essential aspects, namely, the multidimensional image analysis and image perception, respectively. The framework is modelled separately and then combined for the comprehensive analysis of image aesthetic quality evaluation for art design system. After testing on large sets of data, the performance is shown to be efficient.
{"title":"Intelligentization of Art Design System Based on Multidimensional Visual Image Reconstruction Algorithm","authors":"Yuejuan Wang","doi":"10.1109/ICSMDI57622.2023.00037","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00037","url":null,"abstract":"Among the mainstream methods of image aesthetic quality evaluation, it can be divided into traditional aesthetic evaluation methods based on the artificial design features and current popular aesthetic evaluation methods based on deep learning. Hence, this paper studies the intelligentization of art design system based on the multidimensional visual image reconstruction algorithm. In the proposed study, the modelling process contains the 2 essential aspects, namely, the multidimensional image analysis and image perception, respectively. The framework is modelled separately and then combined for the comprehensive analysis of image aesthetic quality evaluation for art design system. After testing on large sets of data, the performance is shown to be efficient.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125925669","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00101
A. Vijay, A. Mustafa, Wardah Afzal, Aatika Shehzad, M. Tariq, Kousalya K
Autonomous robot taxis are the future self-driven robot taxis that can be operated on their own without the help of any driver. They are superfast in speed and would not cause an accident. In short, they are ideal supercars. The motive behind their development is to provide such vehicles that could easily bring a revolution in the traffic system of the world. To achieve that much better performance, industrialists have started doing their work in this field by solving the issues that they are facing in the current cars to obtain their desired results. Full cooperation of such vehicle with other vehicles, road and the whole network is needed in successful launch of such vehicles. One of the major challenges in it is to convert the already formulated cars into such type of robot taxis. Then, for the smooth working of these robot taxis, the obstacles around it must be evaluated. Many other challenges along with solutions are presented in this paper. A brief analysis about the sensors used in the development of autonomous robot taxis are mentioned in the architecture section. Moreover, this article also deals with the architecture of such robot taxis. It is also predicted that these vehicles will be widely used all over the world in the near future.
{"title":"Architecture and Challenges of IoT in developing an Infrastructure for Robot Taxi","authors":"A. Vijay, A. Mustafa, Wardah Afzal, Aatika Shehzad, M. Tariq, Kousalya K","doi":"10.1109/ICSMDI57622.2023.00101","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00101","url":null,"abstract":"Autonomous robot taxis are the future self-driven robot taxis that can be operated on their own without the help of any driver. They are superfast in speed and would not cause an accident. In short, they are ideal supercars. The motive behind their development is to provide such vehicles that could easily bring a revolution in the traffic system of the world. To achieve that much better performance, industrialists have started doing their work in this field by solving the issues that they are facing in the current cars to obtain their desired results. Full cooperation of such vehicle with other vehicles, road and the whole network is needed in successful launch of such vehicles. One of the major challenges in it is to convert the already formulated cars into such type of robot taxis. Then, for the smooth working of these robot taxis, the obstacles around it must be evaluated. Many other challenges along with solutions are presented in this paper. A brief analysis about the sensors used in the development of autonomous robot taxis are mentioned in the architecture section. Moreover, this article also deals with the architecture of such robot taxis. It is also predicted that these vehicles will be widely used all over the world in the near future.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"329 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130180853","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 study of speech is a subfield of the measurement of uniquely identifying and measurable patterns in human activity. It communicates details about a person's characteristics. As an illustration, consider gender, age, emotion, and health status. Voice Identification is the process of pointing person purely with the help of vocal pulses. Using voice assistants and smart gadgets as a base, several academics are exploring in this field. This article offers detailed information on speaker recognition. Additionally, a different adversarial technique is described in advance of the well-known machine learning methods. This study provides an overview of the speaker recognition topic by covering its system, modelling approaches, applications, and some underlying theories. The strengths of speaker recognition technologies will be discussed.
{"title":"Comprehensive Research on Speaker Recognition and its Challenges","authors":"Venkata Syama Sowmya Sri Hari, Arun Kumar Annavarapu, Vamsi Shesamsetti, Sathwik Nalla","doi":"10.1109/ICSMDI57622.2023.00034","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00034","url":null,"abstract":"The study of speech is a subfield of the measurement of uniquely identifying and measurable patterns in human activity. It communicates details about a person's characteristics. As an illustration, consider gender, age, emotion, and health status. Voice Identification is the process of pointing person purely with the help of vocal pulses. Using voice assistants and smart gadgets as a base, several academics are exploring in this field. This article offers detailed information on speaker recognition. Additionally, a different adversarial technique is described in advance of the well-known machine learning methods. This study provides an overview of the speaker recognition topic by covering its system, modelling approaches, applications, and some underlying theories. The strengths of speaker recognition technologies will be discussed.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"695 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128035535","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00024
M. Srikanth, R. Mohan, M. Naik
Agriculture plays a major role in a country's economy and GDP. Most of the farmers still follow old and conventional farming practices which may lead to crop failure. They depend on brokers/middlemen. Create a huge loss for the farmers due to which suicidal rates have increased. Provides accurate results as it considers various factors like soil and weather conditions for determining the best crop and for predicting the yield. The government gets an estimation of the total yield per area. Provides a Peer- To-Peer environment between farmers and buyers removing the need for brokerage, which enables farmers to get profit directly from buyers. Crop failures result from planting a crop without adequate knowledge of climatic and soil conditions, and selling goods to a broker result in even more losses. The goal is to develop a system that allows farmers to receive crop proposals and yield forecasts, as well as sell their harvest directly to the government without the involvement of middlemen. Crop yield is determined by soil and meteorological factors such as pH, NPK levels, temperature, rainfall, and humidity. Based on this, the system advises farmers on which crop is the most suitable and profitable, as well as the potential yield. To tackle the Small Holders' Crop Classification using Optimal Points, Values, Crop Prediction Regression Using Multiple Linear Regression, and Logistic Regression are supervised machine learning models. It eliminates the need for an intermediary to sell to buyers, allowing farmers to earn directly
{"title":"Tackle Outliers for Predictive Small Holder Farming Analysis","authors":"M. Srikanth, R. Mohan, M. Naik","doi":"10.1109/ICSMDI57622.2023.00024","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00024","url":null,"abstract":"Agriculture plays a major role in a country's economy and GDP. Most of the farmers still follow old and conventional farming practices which may lead to crop failure. They depend on brokers/middlemen. Create a huge loss for the farmers due to which suicidal rates have increased. Provides accurate results as it considers various factors like soil and weather conditions for determining the best crop and for predicting the yield. The government gets an estimation of the total yield per area. Provides a Peer- To-Peer environment between farmers and buyers removing the need for brokerage, which enables farmers to get profit directly from buyers. Crop failures result from planting a crop without adequate knowledge of climatic and soil conditions, and selling goods to a broker result in even more losses. The goal is to develop a system that allows farmers to receive crop proposals and yield forecasts, as well as sell their harvest directly to the government without the involvement of middlemen. Crop yield is determined by soil and meteorological factors such as pH, NPK levels, temperature, rainfall, and humidity. Based on this, the system advises farmers on which crop is the most suitable and profitable, as well as the potential yield. To tackle the Small Holders' Crop Classification using Optimal Points, Values, Crop Prediction Regression Using Multiple Linear Regression, and Logistic Regression are supervised machine learning models. It eliminates the need for an intermediary to sell to buyers, allowing farmers to earn directly","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"129 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124231034","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00077
Kajal Saini, Ruchi Jain
Since the growth of the internet, there has been an increase in the circulation offalse information. The very network that keeps us informed about what's going on in the world also provides the ideal environment for the spread of bad content and fake news. Fighting against this fake news is vital since information is what shapes people's perspectives around the world. People don't just establish their own beliefs, but also make significant judgments based on the information that they gather. Should this information turn out to be wrong, the repercussions might be catastrophic. It is entirely impossible for a person to verify each and every piece of news individually. This article has proposed a hybrid deep learning model based on LS TM and BERT with Glove followed by a feature extraction method using TFIDF vectorizer, implement machine learning methods like naive Bayes, ensemble learning, and XG-boost, and evaluate the performance using accuracy and loss, the BERT model outperform with accuracy 99% and 3% loss.
自互联网发展以来，虚假信息的流通有所增加。让我们了解世界上正在发生的事情的网络，也为不良内容和假新闻的传播提供了理想的环境。打击这种假新闻至关重要，因为信息塑造了世界各地人们的观点。人们不仅会建立自己的信念，还会根据他们收集到的信息做出重要的判断。如果这些信息被证明是错误的，后果可能是灾难性的。一个人完全不可能逐一核实每一条新闻。本文提出了一种基于LS TM和BERT with Glove的混合深度学习模型，然后采用TFIDF矢量器进行特征提取方法，实现了朴素贝叶斯、集成学习和XG-boost等机器学习方法，并使用准确率和损失进行了性能评估，BERT模型的准确率为99%，损失为3%。
{"title":"A Hybrid LSTM-BERT and Glove-based Deep Learning Approach for the Detection of Fake News","authors":"Kajal Saini, Ruchi Jain","doi":"10.1109/ICSMDI57622.2023.00077","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00077","url":null,"abstract":"Since the growth of the internet, there has been an increase in the circulation offalse information. The very network that keeps us informed about what's going on in the world also provides the ideal environment for the spread of bad content and fake news. Fighting against this fake news is vital since information is what shapes people's perspectives around the world. People don't just establish their own beliefs, but also make significant judgments based on the information that they gather. Should this information turn out to be wrong, the repercussions might be catastrophic. It is entirely impossible for a person to verify each and every piece of news individually. This article has proposed a hybrid deep learning model based on LS TM and BERT with Glove followed by a feature extraction method using TFIDF vectorizer, implement machine learning methods like naive Bayes, ensemble learning, and XG-boost, and evaluate the performance using accuracy and loss, the BERT model outperform with accuracy 99% and 3% loss.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"168 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120865897","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00093
Tejasri Velugoti, L. Kumar, Koneru Vinay, M. Vanitha
In the ultramodern era of fast moving technology, we are able to do actions that were never before possible. However, in order to complete and manage these studies, a platform that can easily and comfortably automate all of our chores is required Therefore, we must create a Special Assistant with excellent deductive skills and the capacity to communicate with the outside world only via one of the materialistic forms of human contact, i.e. HUMANVOICE.[1]There aredifferent ways to develop voice user interfaces. People without programming skills can use this Voice flow to develop their first voice project. Voice flow is suitable for simple and more complicated voice projects. It can beused for Alexia Skills and Google Actions as well.
{"title":"Voice Flow Control using Artificial Intelligence","authors":"Tejasri Velugoti, L. Kumar, Koneru Vinay, M. Vanitha","doi":"10.1109/ICSMDI57622.2023.00093","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00093","url":null,"abstract":"In the ultramodern era of fast moving technology, we are able to do actions that were never before possible. However, in order to complete and manage these studies, a platform that can easily and comfortably automate all of our chores is required Therefore, we must create a Special Assistant with excellent deductive skills and the capacity to communicate with the outside world only via one of the materialistic forms of human contact, i.e. HUMANVOICE.[1]There aredifferent ways to develop voice user interfaces. People without programming skills can use this Voice flow to develop their first voice project. Voice flow is suitable for simple and more complicated voice projects. It can beused for Alexia Skills and Google Actions as well.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115448317","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00069
Y. Kumar, K. Shirisha, N. Niveditha, M. Swapna, Pavitra Sagar, I. Prashanth
Agriculture relies greatly on rainfall. Recent years witnessed a substantial improvement in the complexity of rainfall prediction. Rainfall forecast will provide valuable predictions to farmers to help them take the appropriate precautions to protect their crops from various weather conditions. Several techniques are available to predict rainfall. Machine Learning (ML) algorithms are particularly useful for predicting rainfall. As machine learning is a type of Artificial Intelligence (AI), it is essential for anticipating rainfall as it enables computer algorithms to make predictions more correctly without explicit guidance. Machine Learning (ML) uses previous data as input to predict the new output values. Meteorologists have attempted to predict future rainfall patterns via previous data. This method is referred to as rainfall forecasting. The primary objective of this research is to identify the best algorithm for predicting rainfall. In this work, SVR (Support Vector Regression) and linear regression strategies were used.
{"title":"Utilizing Machine Learning Algorithms for Rainfall Analysis","authors":"Y. Kumar, K. Shirisha, N. Niveditha, M. Swapna, Pavitra Sagar, I. Prashanth","doi":"10.1109/ICSMDI57622.2023.00069","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00069","url":null,"abstract":"Agriculture relies greatly on rainfall. Recent years witnessed a substantial improvement in the complexity of rainfall prediction. Rainfall forecast will provide valuable predictions to farmers to help them take the appropriate precautions to protect their crops from various weather conditions. Several techniques are available to predict rainfall. Machine Learning (ML) algorithms are particularly useful for predicting rainfall. As machine learning is a type of Artificial Intelligence (AI), it is essential for anticipating rainfall as it enables computer algorithms to make predictions more correctly without explicit guidance. Machine Learning (ML) uses previous data as input to predict the new output values. Meteorologists have attempted to predict future rainfall patterns via previous data. This method is referred to as rainfall forecasting. The primary objective of this research is to identify the best algorithm for predicting rainfall. In this work, SVR (Support Vector Regression) and linear regression strategies were used.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"313 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115867764","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00074
Kakarla Sai Bharath, Anudeep Sanakkayala, Abhishek Kadiyam, Gudapati Pradeep Chandra, Iwin Thanakumar Joseph S, K. B. Brahma Rao
Heart is a vital organ in the human body. It is one of the superior organs, which receives more attention from the internal organs. According to various research studies, heart diseases were considered as the leading cause of death worldwide. The heart disease prediction techniques require more accuracy and precision to identify and forecast various disorders. Any improper disease diagnosis can cause death. Many researchers have been experimenting to develop a software system to predict heart disease using machine learning. The primary goal of this research work is to predict cardiac disease in humans using a machine learning algorithm. This study has reviewed some data mining and machine learning methodologies to perform heart disease prediction.
{"title":"A Performance Comparison on Machine Learning for Forecasting Heart Disease","authors":"Kakarla Sai Bharath, Anudeep Sanakkayala, Abhishek Kadiyam, Gudapati Pradeep Chandra, Iwin Thanakumar Joseph S, K. B. Brahma Rao","doi":"10.1109/ICSMDI57622.2023.00074","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00074","url":null,"abstract":"Heart is a vital organ in the human body. It is one of the superior organs, which receives more attention from the internal organs. According to various research studies, heart diseases were considered as the leading cause of death worldwide. The heart disease prediction techniques require more accuracy and precision to identify and forecast various disorders. Any improper disease diagnosis can cause death. Many researchers have been experimenting to develop a software system to predict heart disease using machine learning. The primary goal of this research work is to predict cardiac disease in humans using a machine learning algorithm. This study has reviewed some data mining and machine learning methodologies to perform heart disease prediction.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126769134","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 : 2023-03-01DOI: 10.1109/ICSMDI57622.2023.00045
Ashwini Bhoware, K. Jajulwar, S. Ghodmare, K. Dabhekar, Vaibhav Bartakke
To mine a blockchain on IP Networks, one must do several tasks related to chain management, rule optimization, verification, and hash generation design. Various consensus model subsets may benefit from the various blockchain mining techniques proposed by researchers. Most of these techniques, however, are rather complicated, which slows down the mining process for large-scale blockchains. Overly simplistic models that include unnecessary redundancies are inefficient and have little practical use. To solve these issues and boost blockchain mining efficiency in large-scale deployments, the authors of this paper propose creating a novel hybrid bioinspired approach. The proposed IP Network model is adaptable to almost all consensus procedures and may be easily combined with dynamic consensus models with few alterations. After collecting performance and context-specific data from the underlying blockchains, the technique uses Genetic Algorithm (GA) that distributes these range sets among miner nodes that support trust, allowing for high-performance mining while maintaining a high degree of trust under actual application situations. The model was tested against Proof-of-Stake (PoS), Proof-of- Work (PoW), Proof-of- Trust (PoT), and Practical Byzantine Fault Tolerance (PBFT) based consensus algorithms to ensure its effectiveness in real-world scenarios. Mining latency, energy consumption, and computational complexity were used as metrics against which this performance was measured. This analysis revealed that the proposed model has the potential to decrease mining latency by 4.5%, energy usage by 3.9%, and compute complexity by 4.1% across a variety of consensus mechanisms, making it suitable for a number of real-time applications.
{"title":"Performance Analysis of Network Management System using Bioinspired -Blockchain Techniquefor IP Networks","authors":"Ashwini Bhoware, K. Jajulwar, S. Ghodmare, K. Dabhekar, Vaibhav Bartakke","doi":"10.1109/ICSMDI57622.2023.00045","DOIUrl":"https://doi.org/10.1109/ICSMDI57622.2023.00045","url":null,"abstract":"To mine a blockchain on IP Networks, one must do several tasks related to chain management, rule optimization, verification, and hash generation design. Various consensus model subsets may benefit from the various blockchain mining techniques proposed by researchers. Most of these techniques, however, are rather complicated, which slows down the mining process for large-scale blockchains. Overly simplistic models that include unnecessary redundancies are inefficient and have little practical use. To solve these issues and boost blockchain mining efficiency in large-scale deployments, the authors of this paper propose creating a novel hybrid bioinspired approach. The proposed IP Network model is adaptable to almost all consensus procedures and may be easily combined with dynamic consensus models with few alterations. After collecting performance and context-specific data from the underlying blockchains, the technique uses Genetic Algorithm (GA) that distributes these range sets among miner nodes that support trust, allowing for high-performance mining while maintaining a high degree of trust under actual application situations. The model was tested against Proof-of-Stake (PoS), Proof-of- Work (PoW), Proof-of- Trust (PoT), and Practical Byzantine Fault Tolerance (PBFT) based consensus algorithms to ensure its effectiveness in real-world scenarios. Mining latency, energy consumption, and computational complexity were used as metrics against which this performance was measured. This analysis revealed that the proposed model has the potential to decrease mining latency by 4.5%, energy usage by 3.9%, and compute complexity by 4.1% across a variety of consensus mechanisms, making it suitable for a number of real-time applications.","PeriodicalId":373017,"journal":{"name":"2023 3rd International Conference on Smart Data Intelligence (ICSMDI)","volume":"348 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124289417","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：