Pub Date : 2022-04-25DOI: 10.1109/syscon53536.2022.9773832
Jacqueline Heaton, S. Givigi
Emotion classification has many applications in human-computer interaction, and is a necessary mode of communication for many different tasks where humans and robots must work together or in close quarters. When working with people who have trouble using verbal communication, or when it is unrealistic to expect verbal communication, robots must still be capable of taking the person’s emotions into account, whether through facial cues, body language, or other signals. Electroencephalograms are capable of capturing the signals of the brain, which can be processed and classified using various artificial intelligence architectures. In this paper, a deep convolutional neural network is applied to an emotion classification task, where it successfully learns to identify six second windows as one of four emotions: boredom, relaxation, horror, and humour. The neural network is applied to 14 individuals and a high accuracy of nearly 100% is achieved when the test data is chosen randomly from the dataset. A study is performed to find what conditions in the data are necessary for high classification accuracy. The emotion data was collected from subjects as they played four games of different genres, designed to evoke one emotion out of boredom, relaxation, humour, or fear, as assessed by the professional game critic services.
{"title":"A Deep CNN System for Classification of Emotions Using EEG Signals","authors":"Jacqueline Heaton, S. Givigi","doi":"10.1109/syscon53536.2022.9773832","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773832","url":null,"abstract":"Emotion classification has many applications in human-computer interaction, and is a necessary mode of communication for many different tasks where humans and robots must work together or in close quarters. When working with people who have trouble using verbal communication, or when it is unrealistic to expect verbal communication, robots must still be capable of taking the person’s emotions into account, whether through facial cues, body language, or other signals. Electroencephalograms are capable of capturing the signals of the brain, which can be processed and classified using various artificial intelligence architectures. In this paper, a deep convolutional neural network is applied to an emotion classification task, where it successfully learns to identify six second windows as one of four emotions: boredom, relaxation, horror, and humour. The neural network is applied to 14 individuals and a high accuracy of nearly 100% is achieved when the test data is chosen randomly from the dataset. A study is performed to find what conditions in the data are necessary for high classification accuracy. The emotion data was collected from subjects as they played four games of different genres, designed to evoke one emotion out of boredom, relaxation, humour, or fear, as assessed by the professional game critic services.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"224 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120942617","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773853
Pavlo Vlastos, A. Hunter, R. Curry, Carlos Isaac Espinosa Ramirez, G. Elkaim
Autonomous vehicles for exploration purposes are often limited by energy and computation capacity. Usually they are tasked with the goal of efficiently and optimally exploring a given region of space. Tasks involving path planning and spatial estimation can require computation time with exponential growth versus the number of measurements taken. This creates a problem if the number of measurements is large. This paper outlines an experiment to compare a spatial estimation method, ordinary kriging with a proposed method, partitioned ordinary kriging (POK) using real environmental data measured by a remote-operated autonomous surface vehicle (ASV). The ASV collected depth measurements of a small body of water, mapped to its GPS location while under remote-control. The mean absolute error (MAE) and computation time were compared as the number of measurements increased. The POK method demonstrated favorable error and computation time compared to ordinary kriging.
{"title":"Applied Partitioned Ordinary Kriging for Online Updates for Autonomous Vehicles","authors":"Pavlo Vlastos, A. Hunter, R. Curry, Carlos Isaac Espinosa Ramirez, G. Elkaim","doi":"10.1109/syscon53536.2022.9773853","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773853","url":null,"abstract":"Autonomous vehicles for exploration purposes are often limited by energy and computation capacity. Usually they are tasked with the goal of efficiently and optimally exploring a given region of space. Tasks involving path planning and spatial estimation can require computation time with exponential growth versus the number of measurements taken. This creates a problem if the number of measurements is large. This paper outlines an experiment to compare a spatial estimation method, ordinary kriging with a proposed method, partitioned ordinary kriging (POK) using real environmental data measured by a remote-operated autonomous surface vehicle (ASV). The ASV collected depth measurements of a small body of water, mapped to its GPS location while under remote-control. The mean absolute error (MAE) and computation time were compared as the number of measurements increased. The POK method demonstrated favorable error and computation time compared to ordinary kriging.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"235 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122946090","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773875
Lirim Ashiku, Richard A. Threlkeld, C. Canfield, C. Dagli
The current organ placement process for transplantation is an evolving system of systems with emergent behavior. This highly integrated complex system consists of Organ Procurement Organizations (OPOs), Transplant Centers (TXC), patients, and their interactions. The number of waitlisted kidney candidates is nearly five times the available supply. Unfortunately, over twenty percent of donated deceased donor kidneys (supply) are discarded due to issues with kidney quality. While some of this discard is medically necessary, some represent a lost opportunity. One approach is to develop a decision support system to identify the right candidate for the right donor at the right time and then communicate that analysis to various stakeholders in different locations over time. This paper uses an incremental hierarchical systems engineering approach to capture the current kidney allocation systems architecture and identify opportunities for an Artificial Intelligence (AI) decision support system to reduce kidney discard. The incremental hierarchical (top to bottom) approach was combined with model-based system engineering (MBSE) to aid in eliciting stakeholders’ needs, behaviors, boundaries, and interactions. This approach led to a structured development process for the attractor “reducing kidney discard” and facilitated systematically documenting the opportunity space. Stakeholders reviewed proposed AI decision support systems, ensuring that decision points with more significant opportunities were addressed. Ultimately, the effectiveness of the systems engineering approach is justified with a data-driven deep learning TXC decision support system validated by transplant surgeons. Future work will include developing data-driven models for all stakeholders using current data incorporating the most recent kidney allocation policy changes.
{"title":"Identifying AI Opportunities in Donor Kidney Acceptance: Incremental Hierarchical Systems Engineering Approach","authors":"Lirim Ashiku, Richard A. Threlkeld, C. Canfield, C. Dagli","doi":"10.1109/syscon53536.2022.9773875","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773875","url":null,"abstract":"The current organ placement process for transplantation is an evolving system of systems with emergent behavior. This highly integrated complex system consists of Organ Procurement Organizations (OPOs), Transplant Centers (TXC), patients, and their interactions. The number of waitlisted kidney candidates is nearly five times the available supply. Unfortunately, over twenty percent of donated deceased donor kidneys (supply) are discarded due to issues with kidney quality. While some of this discard is medically necessary, some represent a lost opportunity. One approach is to develop a decision support system to identify the right candidate for the right donor at the right time and then communicate that analysis to various stakeholders in different locations over time. This paper uses an incremental hierarchical systems engineering approach to capture the current kidney allocation systems architecture and identify opportunities for an Artificial Intelligence (AI) decision support system to reduce kidney discard. The incremental hierarchical (top to bottom) approach was combined with model-based system engineering (MBSE) to aid in eliciting stakeholders’ needs, behaviors, boundaries, and interactions. This approach led to a structured development process for the attractor “reducing kidney discard” and facilitated systematically documenting the opportunity space. Stakeholders reviewed proposed AI decision support systems, ensuring that decision points with more significant opportunities were addressed. Ultimately, the effectiveness of the systems engineering approach is justified with a data-driven deep learning TXC decision support system validated by transplant surgeons. Future work will include developing data-driven models for all stakeholders using current data incorporating the most recent kidney allocation policy changes.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124714591","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773852
Johan Bergelin, A. Cicchetti, Emil Lundin
In general, trains are referred to as environment-friendly transportation means when compared e.g. to cars, busses, or aircraft, being modern trains electrified systems. Unfortunately, the costs due to creation and maintenance of railway infrastructures, notably the overhead lines to power the trains, impose boundaries to their expansion potentials. In this respect, the advances in battery technologies are disclosing new opportunities, like serving partially electrified tracks. In particular, on board batteries can be used as backup energy where overhead lines are not available. In such scenarios, analysing battery requirements and evaluating possible solutions is of critical importance.This paper proposes a model-based systems engineering methodology for evaluating the feasibility of heterogeneous battery systems in the railway domain. The methodology leverages separation of concerns to reduce the complexity of the problem and abstracts the different railway system components by means of corresponding simulation models. The methodology is illustrated through a study performed at an industrial partner; in particular, the paper discusses how simulation models have been conceived, refined, validated, and integrated to analyse the properties of various battery configurations for several passenger trains operating on commercial lines in France. Interestingly, the results demonstrate that heterogeneous battery systems provide a suitable trade-off alternative when compared to homogeneous batteries.
{"title":"Early validation of heterogeneous battery systems in the railway domain","authors":"Johan Bergelin, A. Cicchetti, Emil Lundin","doi":"10.1109/syscon53536.2022.9773852","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773852","url":null,"abstract":"In general, trains are referred to as environment-friendly transportation means when compared e.g. to cars, busses, or aircraft, being modern trains electrified systems. Unfortunately, the costs due to creation and maintenance of railway infrastructures, notably the overhead lines to power the trains, impose boundaries to their expansion potentials. In this respect, the advances in battery technologies are disclosing new opportunities, like serving partially electrified tracks. In particular, on board batteries can be used as backup energy where overhead lines are not available. In such scenarios, analysing battery requirements and evaluating possible solutions is of critical importance.This paper proposes a model-based systems engineering methodology for evaluating the feasibility of heterogeneous battery systems in the railway domain. The methodology leverages separation of concerns to reduce the complexity of the problem and abstracts the different railway system components by means of corresponding simulation models. The methodology is illustrated through a study performed at an industrial partner; in particular, the paper discusses how simulation models have been conceived, refined, validated, and integrated to analyse the properties of various battery configurations for several passenger trains operating on commercial lines in France. Interestingly, the results demonstrate that heterogeneous battery systems provide a suitable trade-off alternative when compared to homogeneous batteries.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126986171","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773925
Dina Nawara, R. Kashef
Recommendation Systems (RSs) have proved a compelling performance to overcome the data overload problem. Context-aware recommenders guide users/clients to more personalized recommendations. Incorporating contextual features in recommendation systems improves the systems’ accuracy; however, they still suffer from sparsity and scalability problems which impact the quality of recommendations. In this paper, to overcome these limitations, we propose a context-aware recommendation system using the notion of consensus clustering, named CARS-CC. The proposed recommendation system is experimentally evaluated using contextual Pre-filtering and Post-filtering approaches. Experimental results show that the concept of consensus learning using clustering analysis can significantly improve the recommender systems’ accuracy. The proposed method surpasses the other recommendation algorithms in terms of accuracy, precision and recall, particularly using the Hybrid Bipartite Graph Formulation (HBGF) method. In addition, CARS-CC(hgpa) has outperformed all other clustering techniques in terms of MAE and RMSE with 23.73% and 7.54%, respectively. The MAE and RMSE results show that consensus clustering leads to better accuracy measures and a more stable resilient recommendation system. The response time taken to generate recommendations using post-filtering is less than that of the pre-filtering approach. The CARS-CC(HGPA) in the post-filtering approach; generates recommendations 58.4% faster than pre-filtering, which speeds up the recommendation process and facilitates real-time response.
{"title":"Context-Aware Recommendation Systems Using Consensus-Clustering","authors":"Dina Nawara, R. Kashef","doi":"10.1109/syscon53536.2022.9773925","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773925","url":null,"abstract":"Recommendation Systems (RSs) have proved a compelling performance to overcome the data overload problem. Context-aware recommenders guide users/clients to more personalized recommendations. Incorporating contextual features in recommendation systems improves the systems’ accuracy; however, they still suffer from sparsity and scalability problems which impact the quality of recommendations. In this paper, to overcome these limitations, we propose a context-aware recommendation system using the notion of consensus clustering, named CARS-CC. The proposed recommendation system is experimentally evaluated using contextual Pre-filtering and Post-filtering approaches. Experimental results show that the concept of consensus learning using clustering analysis can significantly improve the recommender systems’ accuracy. The proposed method surpasses the other recommendation algorithms in terms of accuracy, precision and recall, particularly using the Hybrid Bipartite Graph Formulation (HBGF) method. In addition, CARS-CC(hgpa) has outperformed all other clustering techniques in terms of MAE and RMSE with 23.73% and 7.54%, respectively. The MAE and RMSE results show that consensus clustering leads to better accuracy measures and a more stable resilient recommendation system. The response time taken to generate recommendations using post-filtering is less than that of the pre-filtering approach. The CARS-CC(HGPA) in the post-filtering approach; generates recommendations 58.4% faster than pre-filtering, which speeds up the recommendation process and facilitates real-time response.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121552878","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773931
Jayanta Debnath, Derock Xie
Common Vulnerability Scoring System (CVSS) is intended to capture the key characteristics of a vulnerability and correspondingly produce a numerical score to indicate the severity. Important efforts are conducted for building a CVSS stochastic model in order to provide a high-level risk assessment to better support cybersecurity decision-making. However, these efforts consider nothing regarding HPC (High-Performance Computing) networks using a Science Demilitary Zone (DMZ) architecture that has special design principles to facilitate data transition, analysis, and store through in a broadband backbone. In this paper, an HPCvul (CVSS-based vulnerability and risk assessment) approach is proposed for HPC networks in order to provide an understanding of the ongoing awareness of the HPC security situation under a dynamic cybersecurity environment. For such a purpose, HPCvul advocates the standardization of the collected security-related data from the network to achieve data portability. HPCvul adopts an attack graph to model the likelihood of successful exploitation of a vulnerability. It is able to merge multiple attack graphs from different HPC subnets to yield a full picture of a large HPC network. Substantial results are presented in this work to demonstrate HPCvul design and its performance.
{"title":"CVSS-based Vulnerability and Risk Assessment for High Performance Computing Networks","authors":"Jayanta Debnath, Derock Xie","doi":"10.1109/syscon53536.2022.9773931","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773931","url":null,"abstract":"Common Vulnerability Scoring System (CVSS) is intended to capture the key characteristics of a vulnerability and correspondingly produce a numerical score to indicate the severity. Important efforts are conducted for building a CVSS stochastic model in order to provide a high-level risk assessment to better support cybersecurity decision-making. However, these efforts consider nothing regarding HPC (High-Performance Computing) networks using a Science Demilitary Zone (DMZ) architecture that has special design principles to facilitate data transition, analysis, and store through in a broadband backbone. In this paper, an HPCvul (CVSS-based vulnerability and risk assessment) approach is proposed for HPC networks in order to provide an understanding of the ongoing awareness of the HPC security situation under a dynamic cybersecurity environment. For such a purpose, HPCvul advocates the standardization of the collected security-related data from the network to achieve data portability. HPCvul adopts an attack graph to model the likelihood of successful exploitation of a vulnerability. It is able to merge multiple attack graphs from different HPC subnets to yield a full picture of a large HPC network. Substantial results are presented in this work to demonstrate HPCvul design and its performance.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132328320","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773806
A. Hayward, Maximilian Rappl, A. Fay
Cyber-physical systems (CPSs) are able to collaborate with other CPSs in their environment. Such collaboration, which contains a suitable combination and aggregation of the individual functions of different CPSs, makes it possible that goals can be jointly achieved that the individual CPS would not have been able to achieve on its own. As part of the collaboration, the collaborative CPSs, which may come from different manufacturers, form temporary system groups in which they assume different roles and associated responsibilities. This paper presents a domain-independent function-centered approach that enables the modeling of such system groups at the design time of the single collaborative CPS and thus serves as a basis for cross-manufacturer collaboration planning. The approach describes in 6 steps how the constitution and the behavior of the system group with its participants can be modeled with the help of functions and independent of specific components. The modeling is based on the Systems Modeling Language (SysML), which has been extended to be able to express aspects of the system group.
{"title":"A SysML-based Function-Centered Approach for the Modeling of System Groups for Collaborative Cyber-Physical Systems","authors":"A. Hayward, Maximilian Rappl, A. Fay","doi":"10.1109/syscon53536.2022.9773806","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773806","url":null,"abstract":"Cyber-physical systems (CPSs) are able to collaborate with other CPSs in their environment. Such collaboration, which contains a suitable combination and aggregation of the individual functions of different CPSs, makes it possible that goals can be jointly achieved that the individual CPS would not have been able to achieve on its own. As part of the collaboration, the collaborative CPSs, which may come from different manufacturers, form temporary system groups in which they assume different roles and associated responsibilities. This paper presents a domain-independent function-centered approach that enables the modeling of such system groups at the design time of the single collaborative CPS and thus serves as a basis for cross-manufacturer collaboration planning. The approach describes in 6 steps how the constitution and the behavior of the system group with its participants can be modeled with the help of functions and independent of specific components. The modeling is based on the Systems Modeling Language (SysML), which has been extended to be able to express aspects of the system group.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127781946","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773823
Andrew E. Hong, Lauren A. Rayson, William S. Bland, Jennifer A. Richkus, S. Rosen
This work examines the regional effects of COVID-19 supply chain shocks and potential inventory strategies to sustaining overall economic productivity through lockdowns. We introduce a multi-region extension to the economic production model proposed by Pichler, et al. [15] that was used to assess the extent of Covid-related shocks to gross output through modeling the interdependency between regional and national economies at the industry level. Our extended modeling approach aims to optimize, through genetic search, the degree to which the increased inventory supply targets allow for improved economic productivity and the ideal allocation per industry which most efficiently achieves this mitigation. The approach also integrates a new data regionalization procedure which we apply to a case study of the Alabama state economy. This application is shown to identify a set of major manufacturing and service sectors, where additional inventories enable greater sustained productivity across the Alabama region. This regional analysis of the Alabama economy highlighted the importance of sectors such as chemical, petroleum, food and beverage, and vehicle manufacturing and public administration, construction, management, transportation, and healthcare towards maintaining economic productivity. The ability to quantity regional production impacts from inventory allocations is leading to starting points for determining local government policies that target their most sensitive industries.
{"title":"Regionalized Modeling of Supply Chain Resiliency for Analyzing Incentive Options","authors":"Andrew E. Hong, Lauren A. Rayson, William S. Bland, Jennifer A. Richkus, S. Rosen","doi":"10.1109/syscon53536.2022.9773823","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773823","url":null,"abstract":"This work examines the regional effects of COVID-19 supply chain shocks and potential inventory strategies to sustaining overall economic productivity through lockdowns. We introduce a multi-region extension to the economic production model proposed by Pichler, et al. [15] that was used to assess the extent of Covid-related shocks to gross output through modeling the interdependency between regional and national economies at the industry level. Our extended modeling approach aims to optimize, through genetic search, the degree to which the increased inventory supply targets allow for improved economic productivity and the ideal allocation per industry which most efficiently achieves this mitigation. The approach also integrates a new data regionalization procedure which we apply to a case study of the Alabama state economy. This application is shown to identify a set of major manufacturing and service sectors, where additional inventories enable greater sustained productivity across the Alabama region. This regional analysis of the Alabama economy highlighted the importance of sectors such as chemical, petroleum, food and beverage, and vehicle manufacturing and public administration, construction, management, transportation, and healthcare towards maintaining economic productivity. The ability to quantity regional production impacts from inventory allocations is leading to starting points for determining local government policies that target their most sensitive industries.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133797050","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773838
N. C. Will
The use of IoT devices is increasingly present in our daily lives, as they offer many possibilities for developers and the industry to develop applications, taking advantage of their connectivity capabilities, low cost, and, often, small size. As the use of these applications is continuously increasing, the concerns about the privacy and confidentiality of the data generated by these devices also increase, since many applications share the collected data with fog and cloud servers, due to the computational constraints of the edge devices. Fog and cloud environments are used to aggregate and analyze data collected by multiple devices, allowing to summarize these data and to offer personalized services to the users. As IoT devices can collect sensitive data from users, such as personal and behavioral information, it is crucial to handle such data ensuring the privacy of their owners. Privacy-preserving data aggregation schemes are proposed in the literature, but many of them are limited to specific functions and homogeneous data or to specific contexts, such as smart metering and e-health, and there is no publicly available tool to handle heterogeneous data. This paper describes ongoing research that aims to build a generic data aggregation scheme, taking advantage of Trusted Execution Environments (TEE) to ensure data and user privacy and allowing to process heterogeneous data and perform complex computations, including the use of machine learning algorithms. We describe the system architecture, our preliminary findings, and the next steps to implement and validate our proposal.
{"title":"A Privacy-Preserving Data Aggregation Scheme for Fog/Cloud-Enhanced IoT Applications Using a Trusted Execution Environment","authors":"N. C. Will","doi":"10.1109/syscon53536.2022.9773838","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773838","url":null,"abstract":"The use of IoT devices is increasingly present in our daily lives, as they offer many possibilities for developers and the industry to develop applications, taking advantage of their connectivity capabilities, low cost, and, often, small size. As the use of these applications is continuously increasing, the concerns about the privacy and confidentiality of the data generated by these devices also increase, since many applications share the collected data with fog and cloud servers, due to the computational constraints of the edge devices. Fog and cloud environments are used to aggregate and analyze data collected by multiple devices, allowing to summarize these data and to offer personalized services to the users. As IoT devices can collect sensitive data from users, such as personal and behavioral information, it is crucial to handle such data ensuring the privacy of their owners. Privacy-preserving data aggregation schemes are proposed in the literature, but many of them are limited to specific functions and homogeneous data or to specific contexts, such as smart metering and e-health, and there is no publicly available tool to handle heterogeneous data. This paper describes ongoing research that aims to build a generic data aggregation scheme, taking advantage of Trusted Execution Environments (TEE) to ensure data and user privacy and allowing to process heterogeneous data and perform complex computations, including the use of machine learning algorithms. We describe the system architecture, our preliminary findings, and the next steps to implement and validate our proposal.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121975424","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 : 2022-04-25DOI: 10.1109/syscon53536.2022.9773835
Albert Wong, Puja Unni, Andréia Henrique, T. Nguyen, C. Chiu, Y. Khmelevsky, Joe Mahony
Traditional time-series techniques produce forecasts on future values based on the trend or seasonality of past values. It is not easy for these techniques to consider the impact of other exogenous and calendar-related variables. This paper uses the electricity usage data from Harris SmartWorks to demonstrate an approach to building and training machine learning models to overcome this problem. It is shown that Machine learning models produce accurate daily forecasts for hourly usage. The performance of these models could be evaluated by one conventional metric, and one explicitly built for articulating the model’s forecasting accuracy for peak periods.
{"title":"Machine Learning Models Application in Daily Forecasting of Hourly Electricity Usage","authors":"Albert Wong, Puja Unni, Andréia Henrique, T. Nguyen, C. Chiu, Y. Khmelevsky, Joe Mahony","doi":"10.1109/syscon53536.2022.9773835","DOIUrl":"https://doi.org/10.1109/syscon53536.2022.9773835","url":null,"abstract":"Traditional time-series techniques produce forecasts on future values based on the trend or seasonality of past values. It is not easy for these techniques to consider the impact of other exogenous and calendar-related variables. This paper uses the electricity usage data from Harris SmartWorks to demonstrate an approach to building and training machine learning models to overcome this problem. It is shown that Machine learning models produce accurate daily forecasts for hourly usage. The performance of these models could be evaluated by one conventional metric, and one explicitly built for articulating the model’s forecasting accuracy for peak periods.","PeriodicalId":437743,"journal":{"name":"2022 IEEE International Systems Conference (SysCon)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129440207","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：