Pub Date : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454238
Md. Ali Azam, Abir Hossen, Md Hafizur Rahman
Biologically inspired computing techniques are very effective and useful in many areas of research including data clustering. Ant clustering algorithm is a nature-inspired clustering technique which is extensively studied for over two decades. In this study, we extend the ant clustering algorithm (ACA) to a hybrid ant clustering algorithm (hACA). Specifically, we include a genetic algorithm in standard ACA to extend the hybrid algorithm for better performance. We also introduced novel pick up and drop off rules to speed up the clustering performance. We study the performance of the hACA algorithm and compare with standard ACA as a benchmark.
{"title":"Hybrid Ant Swarm-Based Data Clustering","authors":"Md. Ali Azam, Abir Hossen, Md Hafizur Rahman","doi":"10.1109/AIIoT52608.2021.9454238","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454238","url":null,"abstract":"Biologically inspired computing techniques are very effective and useful in many areas of research including data clustering. Ant clustering algorithm is a nature-inspired clustering technique which is extensively studied for over two decades. In this study, we extend the ant clustering algorithm (ACA) to a hybrid ant clustering algorithm (hACA). Specifically, we include a genetic algorithm in standard ACA to extend the hybrid algorithm for better performance. We also introduced novel pick up and drop off rules to speed up the clustering performance. We study the performance of the hACA algorithm and compare with standard ACA as a benchmark.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126824872","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454225
Vishnu Mohan, D. Czarkowski
A controller for multi-phase wireless power transfer system is designed using H-infinity mixed sensitivity approach to improve the dynamic performance of the system. Robustness analysis is conducted for the combined system to check if the system is stable for different coupling coefficients. Experimental verification of the wireless power transfer system with the controller is performed on a 1.5 kW setup.
{"title":"Controller Design and Robustness Analysis for Multi-phase Wireless Power Transfer System Using H-infinity Mixed Sensitivity Approach","authors":"Vishnu Mohan, D. Czarkowski","doi":"10.1109/AIIoT52608.2021.9454225","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454225","url":null,"abstract":"A controller for multi-phase wireless power transfer system is designed using H-infinity mixed sensitivity approach to improve the dynamic performance of the system. Robustness analysis is conducted for the combined system to check if the system is stable for different coupling coefficients. Experimental verification of the wireless power transfer system with the controller is performed on a 1.5 kW setup.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131450317","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454213
R. R. Maaliw
Graduate's success on licensure examinations has a significant impact on various facets of a higher educational institution. Using a comprehensive data mining process, this research compared the accuracy of multiple classification algorithms to determine predictors of students' professional certification performance. The Random Forest model achieved the best cross-validated accuracy score of 92.70% based on the evaluation data. A model inspection method of permutation feature importance was used to uncover information from 500 graduates of Southern Luzon State University's electronics engineering program from 2014 to 2019. Among the 33 variables examined, the verbal reasoning or reading comprehension ability of students unveils a clear attribution with their licensure test results along with ratings from different courses in mathematics, professional, and electrical circuits. Thus, the data-driven information can be used to develop programs, initiatives, and techniques to improve success on the electronics engineering licensure examinations.
{"title":"Early Prediction of Electronics Engineering Licensure Examination Performance using Random Forest","authors":"R. R. Maaliw","doi":"10.1109/AIIoT52608.2021.9454213","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454213","url":null,"abstract":"Graduate's success on licensure examinations has a significant impact on various facets of a higher educational institution. Using a comprehensive data mining process, this research compared the accuracy of multiple classification algorithms to determine predictors of students' professional certification performance. The Random Forest model achieved the best cross-validated accuracy score of 92.70% based on the evaluation data. A model inspection method of permutation feature importance was used to uncover information from 500 graduates of Southern Luzon State University's electronics engineering program from 2014 to 2019. Among the 33 variables examined, the verbal reasoning or reading comprehension ability of students unveils a clear attribution with their licensure test results along with ratings from different courses in mathematics, professional, and electrical circuits. Thus, the data-driven information can be used to develop programs, initiatives, and techniques to improve success on the electronics engineering licensure examinations.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121309441","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454186
Narayana Darapaneni, Shyamal Dhua, Nikita Khare, K. Ayush, K. N., Supriya Ghodke, Abhishek Rajput, Saswat P Beurik, A. Paduri
This paper aims to study the COVID-19 vaccination drive in India to forecast the time, it will take vaccinate the minimum number of population for achieving herd immunity. As per the government data on 25th March, 2021, a total of 5,55,04,440 doses have been administered as first dose and 85,02,968 as the second dose, which is just a mere fraction of the total population of India which stands at 1.3 billion. As the number of cases are rising, considering the situation, it is important to expedite the drive and follow strict restrictions to achieve herd immunity. A simulation of the SIR model has been created to identify the effective reproduction number (Re), and then through time series analysis using Prophet model, the conclusion has been drawn for the number of days it will take to vaccinate enough population to achieve herd immunity. As an initial step, we will be fitting the data available for COVID-19 for India in the SIR model which is a set of three Ordinary Differential Equations (ODE). The results from the ODEs will be used to determining the initial Re which will be matched with the data set. Once confirming the Re value present in data set, the same will be passed to the data-driven forecasting time series model to get insights and draw conclusions which will help authorities to help in planning the drive and implement necessary actions to avoid further growth of COVID-19 cases.
{"title":"Forecasting Vaccination Drive In India for Herd Immunity using SIR and Prophet Model","authors":"Narayana Darapaneni, Shyamal Dhua, Nikita Khare, K. Ayush, K. N., Supriya Ghodke, Abhishek Rajput, Saswat P Beurik, A. Paduri","doi":"10.1109/AIIoT52608.2021.9454186","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454186","url":null,"abstract":"This paper aims to study the COVID-19 vaccination drive in India to forecast the time, it will take vaccinate the minimum number of population for achieving herd immunity. As per the government data on 25th March, 2021, a total of 5,55,04,440 doses have been administered as first dose and 85,02,968 as the second dose, which is just a mere fraction of the total population of India which stands at 1.3 billion. As the number of cases are rising, considering the situation, it is important to expedite the drive and follow strict restrictions to achieve herd immunity. A simulation of the SIR model has been created to identify the effective reproduction number (Re), and then through time series analysis using Prophet model, the conclusion has been drawn for the number of days it will take to vaccinate enough population to achieve herd immunity. As an initial step, we will be fitting the data available for COVID-19 for India in the SIR model which is a set of three Ordinary Differential Equations (ODE). The results from the ODEs will be used to determining the initial Re which will be matched with the data set. Once confirming the Re value present in data set, the same will be passed to the data-driven forecasting time series model to get insights and draw conclusions which will help authorities to help in planning the drive and implement necessary actions to avoid further growth of COVID-19 cases.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130778734","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454223
Shihui Fu, Guiwen Luo, Guang Gong
Membership proof is a very useful building block for checking if an entity is in a list. This tool is widely used in many scenarios. For instance in blockchain where checking membership of an unspent coin in a huge set is necessary, or in the scenario where certain privacy-preserving property on the list or on the entity is required. When it comes to multi-user applications, the naive way that verifies the membership relations one by one is very inefficient. In this work, we utilize subvector commitment schemes and non-interactive proofs of knowledge of elliptic curve discrete logarithms to present two batched membership proofs for multiple users, i.e., batched non-anonymous membership proofs and batched anonymous membership proofs, which offer plausible anonymity assurance respectively on the organization group list and on the users when combined within the blockchain applications. The non-anonymous membership proof scheme requires a trusted setup, but its proof size is only one bilinear group element and is independent of both the size of list and the number of users. The anonymous membership proof scheme requires no trusted setup, and its proof size is linear in the size of organization group and is independent of the number of users. Their security relies respectively on the CubeDH and the discrete logarithm assumptions. Finally, as a use-case application scenario, we extend Mesh which is a blockchain based supply chain management solution to Mesh+ which supports batched anonymous membership proofs.
{"title":"Batching Anonymous and Non-Anonymous Membership Proofs for Blockchain Applications","authors":"Shihui Fu, Guiwen Luo, Guang Gong","doi":"10.1109/AIIoT52608.2021.9454223","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454223","url":null,"abstract":"Membership proof is a very useful building block for checking if an entity is in a list. This tool is widely used in many scenarios. For instance in blockchain where checking membership of an unspent coin in a huge set is necessary, or in the scenario where certain privacy-preserving property on the list or on the entity is required. When it comes to multi-user applications, the naive way that verifies the membership relations one by one is very inefficient. In this work, we utilize subvector commitment schemes and non-interactive proofs of knowledge of elliptic curve discrete logarithms to present two batched membership proofs for multiple users, i.e., batched non-anonymous membership proofs and batched anonymous membership proofs, which offer plausible anonymity assurance respectively on the organization group list and on the users when combined within the blockchain applications. The non-anonymous membership proof scheme requires a trusted setup, but its proof size is only one bilinear group element and is independent of both the size of list and the number of users. The anonymous membership proof scheme requires no trusted setup, and its proof size is linear in the size of organization group and is independent of the number of users. Their security relies respectively on the CubeDH and the discrete logarithm assumptions. Finally, as a use-case application scenario, we extend Mesh which is a blockchain based supply chain management solution to Mesh+ which supports batched anonymous membership proofs.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131002119","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454249
Fu-Shiung Hsieh
The widespread use of IoT endows Cyber-Physical Systems (CPS) with the capability to supervise, monitor and control of manufacturing systems. However, study on the impact of unexpected events such as resource failures on the operations of CPS is less explored. The goal of this study is to propose a systematic framework to analyze CPS and pave the way for the development of an effective strategy to handle unexpected events. In this paper, we will focus on analysis of a class of resource failures events in CPS. To analyze the effects of resource failures on CPS, a network model is constructed based on the timed Petri net models to represent the cyber-world model of CPS. An optimization problem is formulated to check the feasibility to meet the original deadline. The influence of resource failures on CPS is analyzed by solving the optimization problem based on a network constructed based on the cyber-world model of CPS. The proposed method is illustrated by an example.
{"title":"Robustness Analysis of Cyber-Physical systems based on Discrete Timed Cyber-Physical Models","authors":"Fu-Shiung Hsieh","doi":"10.1109/AIIoT52608.2021.9454249","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454249","url":null,"abstract":"The widespread use of IoT endows Cyber-Physical Systems (CPS) with the capability to supervise, monitor and control of manufacturing systems. However, study on the impact of unexpected events such as resource failures on the operations of CPS is less explored. The goal of this study is to propose a systematic framework to analyze CPS and pave the way for the development of an effective strategy to handle unexpected events. In this paper, we will focus on analysis of a class of resource failures events in CPS. To analyze the effects of resource failures on CPS, a network model is constructed based on the timed Petri net models to represent the cyber-world model of CPS. An optimization problem is formulated to check the feasibility to meet the original deadline. The influence of resource failures on CPS is analyzed by solving the optimization problem based on a network constructed based on the cyber-world model of CPS. The proposed method is illustrated by an example.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133340271","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454244
Md. Sajjad Mahmud Khan, Mahiuddin Ahmed, Raseduz Zaman Rasel, Mohammad Monirujjaman Khan
Cataract is one of the prevalent causes of visual impairment and blindness worldwide. There is around 50% of overall blindness. Therefore, an early detection and prevention of cataract may reduce the visual impairment and the blindness. The advancement of Artificial Intelligence (AI) in the field of ophthalmology such as glaucoma, macular degeneration, diabetic retinopathy, corneal conditions, age related eye diseases is quite fruitful unlike cataract. Most of the existing approaches on cataract detection are based on traditional machine learning methods. On the other hand, the manual extraction of retinal features is a time-consuming process and requires an expert ophthalmologist. So, we proposed a model VGG19 which is a convolutional neural network model to detect the cataract by using color fundus images.
{"title":"Cataract Detection Using Convolutional Neural Network with VGG-19 Model","authors":"Md. Sajjad Mahmud Khan, Mahiuddin Ahmed, Raseduz Zaman Rasel, Mohammad Monirujjaman Khan","doi":"10.1109/AIIoT52608.2021.9454244","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454244","url":null,"abstract":"Cataract is one of the prevalent causes of visual impairment and blindness worldwide. There is around 50% of overall blindness. Therefore, an early detection and prevention of cataract may reduce the visual impairment and the blindness. The advancement of Artificial Intelligence (AI) in the field of ophthalmology such as glaucoma, macular degeneration, diabetic retinopathy, corneal conditions, age related eye diseases is quite fruitful unlike cataract. Most of the existing approaches on cataract detection are based on traditional machine learning methods. On the other hand, the manual extraction of retinal features is a time-consuming process and requires an expert ophthalmologist. So, we proposed a model VGG19 which is a convolutional neural network model to detect the cataract by using color fundus images.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133484922","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454197
Z. Abbood, M. Ilyas, Ç. Aydin, Mahmoud Shuker Mahmoud, Nida Abdulredha
The presence of Roadside Units (RSUs) helps network loads to be expanded to the other nodes that have already been far away from frequent node exposure. We proposed in this work utilizing the mobile node to operate as a roadside unit and operate data packet routing such as roadside units does. The main problem of utilizing the huge number of roadside units is the spending of huge time for data provision by a reduction in performance. Also, in this paper, we attempt using the various number of mobile nodes such as roadside units that are different from traditional roadside units such as the past is fixed and the second is active as a movement previous. The proposed method was executing by utilizing ad hoc on demand distance vector (AODV) routing is a path protocol for mobile ad-hoc networks (MANETs) and other wireless ad-hoc networks, this protocol designed for usage of ad-hoc mobile networks. Also, it's an active protocol, the routes are made only when they are needed, common routing tables, one entry single destination, and supplement numbers in conformity with determine whether routing information is up-to-date and to forestall routing loops. In this paper, mobile vehicles move randomly on highways, so in the event of a collision is too high, it is assumed that the vehicle will stop, and the collision site will be subject to accommodate more than one vehicle. Where vehicles are driven at high speed. Due to the driver's ignorance of the accident area, they can enter it, and thus the problem is magnified. The results achieved shows that numerous mobile nodes as a roadside unit may enhance the communication according to the computation of the average time delay and the link duration of the connection and reconnecting each node. Therefore, the results may reduce the delay time and maintain the connection for a longer period, as shown in the fourth simulation model.
{"title":"Automatic Detection of Vehicle Congestion by Using Roadside Unit","authors":"Z. Abbood, M. Ilyas, Ç. Aydin, Mahmoud Shuker Mahmoud, Nida Abdulredha","doi":"10.1109/AIIoT52608.2021.9454197","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454197","url":null,"abstract":"The presence of Roadside Units (RSUs) helps network loads to be expanded to the other nodes that have already been far away from frequent node exposure. We proposed in this work utilizing the mobile node to operate as a roadside unit and operate data packet routing such as roadside units does. The main problem of utilizing the huge number of roadside units is the spending of huge time for data provision by a reduction in performance. Also, in this paper, we attempt using the various number of mobile nodes such as roadside units that are different from traditional roadside units such as the past is fixed and the second is active as a movement previous. The proposed method was executing by utilizing ad hoc on demand distance vector (AODV) routing is a path protocol for mobile ad-hoc networks (MANETs) and other wireless ad-hoc networks, this protocol designed for usage of ad-hoc mobile networks. Also, it's an active protocol, the routes are made only when they are needed, common routing tables, one entry single destination, and supplement numbers in conformity with determine whether routing information is up-to-date and to forestall routing loops. In this paper, mobile vehicles move randomly on highways, so in the event of a collision is too high, it is assumed that the vehicle will stop, and the collision site will be subject to accommodate more than one vehicle. Where vehicles are driven at high speed. Due to the driver's ignorance of the accident area, they can enter it, and thus the problem is magnified. The results achieved shows that numerous mobile nodes as a roadside unit may enhance the communication according to the computation of the average time delay and the link duration of the connection and reconnecting each node. Therefore, the results may reduce the delay time and maintain the connection for a longer period, as shown in the fourth simulation model.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114536279","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454215
Tarun Ganesh Palla, Shahab Tayeb
The advancement in recent IoT devices has led to catastrophic attacks on the devices by breaching user's privacy and exhausting the resources in organizations, which costs users and organizations time and money. One such malware which has been extremely harmful is Mirai, which has created worldwide recognition by impacting the digital world. There are several ways to mitigate Mirai, but Machine Learning-based approach has proved to be accurate and reliable in averting the malware. In this paper, a novel approach to detecting Mirai using Machine Learning Algorithm is proposed and implemented in Matlab 2018b. To evaluate the proposed approach, Mirai and Benign datasets are considered and training is performed on the dataset using Artificial Neural Network, which provides consistent results of Accuracy, Precision, Recall and F-1 score which are found to be considered accurate and reliable as the best performance was achieved with an accuracy of 92.9% and False Negative rate of 0.3, which is efficient in detecting the Mirai and is similar to the Anomaly-based Malware Detection in terms of Metrics.
{"title":"Intelligent Mirai Malware Detection in IoT Devices","authors":"Tarun Ganesh Palla, Shahab Tayeb","doi":"10.1109/AIIoT52608.2021.9454215","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454215","url":null,"abstract":"The advancement in recent IoT devices has led to catastrophic attacks on the devices by breaching user's privacy and exhausting the resources in organizations, which costs users and organizations time and money. One such malware which has been extremely harmful is Mirai, which has created worldwide recognition by impacting the digital world. There are several ways to mitigate Mirai, but Machine Learning-based approach has proved to be accurate and reliable in averting the malware. In this paper, a novel approach to detecting Mirai using Machine Learning Algorithm is proposed and implemented in Matlab 2018b. To evaluate the proposed approach, Mirai and Benign datasets are considered and training is performed on the dataset using Artificial Neural Network, which provides consistent results of Accuracy, Precision, Recall and F-1 score which are found to be considered accurate and reliable as the best performance was achieved with an accuracy of 92.9% and False Negative rate of 0.3, which is efficient in detecting the Mirai and is similar to the Anomaly-based Malware Detection in terms of Metrics.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"71 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116287289","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}
Streaming Big Data pipelines frequently use multiple platforms connected to each other for performing analytics with different communication models. Existing techniques like Access Control Lists (ACLs) or Role-Based Access Control (RBAC) are unable to address access control at the granularity of an individual tuple. Moreover, ACLs and RBAC fail to impose uniform access control over heterogeneous streaming platforms. In this paper, we present a unified mechanism to insert access control policies into data streams at the point of ingestion and enforce it across multiple platforms that use different communication models like publish-subscribe and point to point. We exemplify our solution through implementation with Apache Kafka and Apache Storm. We further illustrate the enforcement of access control in join queries involving streams with different access control rules.
{"title":"HACS: Access Control for Streaming Data Across Heterogeneous Communication Models","authors":"Atul Anand Gopalakrishnan, Ashish Christopher Victor, Deepika Karanji, Umashankar Sivakumar, Seema Nambiar, Subramaniam Kalambur","doi":"10.1109/AIIoT52608.2021.9454185","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454185","url":null,"abstract":"Streaming Big Data pipelines frequently use multiple platforms connected to each other for performing analytics with different communication models. Existing techniques like Access Control Lists (ACLs) or Role-Based Access Control (RBAC) are unable to address access control at the granularity of an individual tuple. Moreover, ACLs and RBAC fail to impose uniform access control over heterogeneous streaming platforms. In this paper, we present a unified mechanism to insert access control policies into data streams at the point of ingestion and enforce it across multiple platforms that use different communication models like publish-subscribe and point to point. We exemplify our solution through implementation with Apache Kafka and Apache Storm. We further illustrate the enforcement of access control in join queries involving streams with different access control rules.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134370357","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：