Pub Date : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284875
Jason Nguyen, Ritu Chaturvedi
With the advent of the COVID-19 pandemic, people have flocked to social media in order to stage their thoughts surrounding these unusual circumstances. This paper aims to uncover public sentiment regarding the novel coronavirus pandemic on the microblogging platform Twitter. This is done through a proposed algorithm that builds off of existing aspect-based sentiment analysis approaches and opts for a Naïve-Bayes route to classify existing Tweets that have been atomized into n-grams. This research concludes that overall sentiment regarding the COVID-19 outbreak over July 2020 is a combination of pessimism and dejection as our quarantine denizens take to their online platforms in airing their polemic opinions.
{"title":"Quarantine Quibbles: A Sentiment Analysis of COVID-19 Tweets","authors":"Jason Nguyen, Ritu Chaturvedi","doi":"10.1109/IEMCON51383.2020.9284875","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284875","url":null,"abstract":"With the advent of the COVID-19 pandemic, people have flocked to social media in order to stage their thoughts surrounding these unusual circumstances. This paper aims to uncover public sentiment regarding the novel coronavirus pandemic on the microblogging platform Twitter. This is done through a proposed algorithm that builds off of existing aspect-based sentiment analysis approaches and opts for a Naïve-Bayes route to classify existing Tweets that have been atomized into n-grams. This research concludes that overall sentiment regarding the COVID-19 outbreak over July 2020 is a combination of pessimism and dejection as our quarantine denizens take to their online platforms in airing their polemic opinions.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"56 1","pages":"0346-0350"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76923659","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284859
R. Sundararajan, M. Iqbal
This paper discusses the design of solar energy system in offline mode along with the implementation of Vehicle to Home (V2H) to meet the energy demand of a smart house for Newfoundland condition. A site was selected (13 Polina road) located in St. John's, Newfoundland, Canada. For the chosen site, an optimized system was designed to meet the energy demand of the house using BE opt and Homer software. Further, based on Nissan Leaf (Electric Vehicle), the concept of V2H is implemented with the help of smart current sensors installed in the house that also helps in transmitting the sensor's data to cloud with the help of IoT. Based on the information generated, the system operates either in V2H more or in PV power mode.
{"title":"Design of an IoT interface for a solar energy system with vehicle to home option for Newfoundland conditions","authors":"R. Sundararajan, M. Iqbal","doi":"10.1109/IEMCON51383.2020.9284859","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284859","url":null,"abstract":"This paper discusses the design of solar energy system in offline mode along with the implementation of Vehicle to Home (V2H) to meet the energy demand of a smart house for Newfoundland condition. A site was selected (13 Polina road) located in St. John's, Newfoundland, Canada. For the chosen site, an optimized system was designed to meet the energy demand of the house using BE opt and Homer software. Further, based on Nissan Leaf (Electric Vehicle), the concept of V2H is implemented with the help of smart current sensors installed in the house that also helps in transmitting the sensor's data to cloud with the help of IoT. Based on the information generated, the system operates either in V2H more or in PV power mode.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"98 1","pages":"0597-0601"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80946985","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284860
M. Jasim, N. Siasi, S. Malapaka, D. Oliveira, O. Ugweje
Fog computing is a promising edge network paradigm for delay-sensitive applications, attributed to the reduced link delays between nodes. However, the limited resources at the fog nodes impose restrictions to serve computation-intensive requests. Therefore, these requests are often relayed to the cloud nodes and thereby incurring link delays. This approach impedes serving requests that are both delay-sensitive and computation-intensive. Hence this paper proposes a novel single-tier fog architecture that processes this category of requests at the edge of the network. Moreover, a service function chain (SFC) provisioning scheme is implemented on the proposed architecture using Hooke Jeeves algorithm. Simulation results indicate that the proposed solution outperforms hybrid fog-cloud architectures for the same number of hosted requests (saturation level) in terms of delay and energy consumption.
{"title":"A Single-Tier Fog Architecture for Delay-Sensitive and Computation-Intensive SFC Requests","authors":"M. Jasim, N. Siasi, S. Malapaka, D. Oliveira, O. Ugweje","doi":"10.1109/IEMCON51383.2020.9284860","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284860","url":null,"abstract":"Fog computing is a promising edge network paradigm for delay-sensitive applications, attributed to the reduced link delays between nodes. However, the limited resources at the fog nodes impose restrictions to serve computation-intensive requests. Therefore, these requests are often relayed to the cloud nodes and thereby incurring link delays. This approach impedes serving requests that are both delay-sensitive and computation-intensive. Hence this paper proposes a novel single-tier fog architecture that processes this category of requests at the edge of the network. Moreover, a service function chain (SFC) provisioning scheme is implemented on the proposed architecture using Hooke Jeeves algorithm. Simulation results indicate that the proposed solution outperforms hybrid fog-cloud architectures for the same number of hosted requests (saturation level) in terms of delay and energy consumption.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"1 1","pages":"0654-0660"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85990668","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284897
M. Maswood, Md Ashif Uddin, Uzzwal Kumar Dey, Md Mainul Islam Mamun, Moriom Akter, Shamima Sultana Sonia, Abdullah G. Alharbi
Chemical sensing is an important issue in food, water, environment, biomedical, and pharmaceutical field. Conventional methods used in laboratory for sensing the chemical are costly, time consuming, and sometimes wastes significant amount of sample. Photonic Crystal Fiber (PCF) offers high compactness and design flexibility and it can be used as biosensor, chemical sensor, liquid sensor, temperature sensor, mechanical sensor, gas sensor, and so on. In this work, we designed PCF to sense different concentrations of different liquids by one PCF structure. We designed different structure for silica cladding hexagonal PCF to sense different concentrations of benzene-toluene and ethanol-water mixer. Core diameter, air hole diameter, and air hole diameter to lattice pitch ratio are varied to get the optimal result as well to explore the effect of core size, air hole size and the pitch on liquid chemical sensing. Performance of the chemical sensors was examined based on confinement loss and sensitivity. The performance of the sensor varied a lot and basically it depends not only on refractive index of the liquid but also on sensing wavelengths. Our designed sensor can provide comparatively high sensitivity and low confinement loss.
{"title":"A Novel Sensor Design to Sense Liquid Chemical Mixtures using Photonic Crystal Fiber to Achieve High Sensitivity and Low Confinement Losses","authors":"M. Maswood, Md Ashif Uddin, Uzzwal Kumar Dey, Md Mainul Islam Mamun, Moriom Akter, Shamima Sultana Sonia, Abdullah G. Alharbi","doi":"10.1109/IEMCON51383.2020.9284897","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284897","url":null,"abstract":"Chemical sensing is an important issue in food, water, environment, biomedical, and pharmaceutical field. Conventional methods used in laboratory for sensing the chemical are costly, time consuming, and sometimes wastes significant amount of sample. Photonic Crystal Fiber (PCF) offers high compactness and design flexibility and it can be used as biosensor, chemical sensor, liquid sensor, temperature sensor, mechanical sensor, gas sensor, and so on. In this work, we designed PCF to sense different concentrations of different liquids by one PCF structure. We designed different structure for silica cladding hexagonal PCF to sense different concentrations of benzene-toluene and ethanol-water mixer. Core diameter, air hole diameter, and air hole diameter to lattice pitch ratio are varied to get the optimal result as well to explore the effect of core size, air hole size and the pitch on liquid chemical sensing. Performance of the chemical sensors was examined based on confinement loss and sensitivity. The performance of the sensor varied a lot and basically it depends not only on refractive index of the liquid but also on sensing wavelengths. Our designed sensor can provide comparatively high sensitivity and low confinement loss.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"44 1","pages":"0686-0691"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90491917","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}
In the present world, the commotion centering Big Data is somewhat obscuring the craft of mining information from smaller samples. Populations with limited examples but huge dimensionality are a common phenomenon, otherwise known as the curse of dimensionality-especially in the health sector-thanks to the recently-discovered potential of data mining and the enthusiasm for feature engineering. Correlated, noisy, redundant features are byproducts of this tendency, which makes learning algorithms converge with greater efforts. This paper proposes a novel feature-pruning technique relying on computational graph theory. Restoring the appeal of pre-AI conventional computing, the paper applies Disjoint Set Union (DSU) on unidirectional graphs prepared basis thresholded Spearman's rank correlation coefficient, r. Gradual withdrawal of leniency on Spearman's r caused a greater tendency in features to form clusters, causing the dimensionality to shrink. The results-extracting out finer, more representative roots as features-have been $k$-fold cross-validated on a case study examining subjects for Parkinson's. Qualitatively, the method overcomes Principal Component Analysis's (PCA) limitation of inexplicit merging of features and Linear Discriminant Analysis's (LDA) limitation of inextendibility to multiple classes. Statistical inference verified a significant rise in performance, establishing an example of conventional hard computing reinforcing modern soft computing.
在当今世界，以大数据为中心的骚动在某种程度上模糊了从较小样本中挖掘信息的工艺。由于最近发现的数据挖掘的潜力和对特征工程的热情，人口数量有限但维度巨大是一种常见的现象，或者被称为维度的诅咒——特别是在卫生部门。相关的、有噪声的、冗余的特征是这种趋势的副产品，这使得学习算法需要更大的努力才能收敛。本文提出了一种基于计算图论的特征剪枝技术。为了恢复前ai传统计算的吸引力，本文在基于阈值的Spearman秩相关系数r制备的单向图上应用Disjoint Set Union (DSU)。逐渐撤销对Spearman秩相关系数r的宽容度导致特征更倾向于形成聚类，从而导致维数缩小。结果——提取出更精细、更有代表性的根作为特征——已经在一个帕金森患者的案例研究中得到了k倍的交叉验证。在定性上，该方法克服了主成分分析(PCA)特征不明确合并的局限性和线性判别分析(LDA)不可扩展到多类的局限性。统计推断验证了性能的显著提高，建立了传统硬计算增强现代软计算的例子。
{"title":"Graph Theory for Dimensionality Reduction: A Case Study to Prognosticate Parkinson's","authors":"Shithi Maitra, Tonmoy Hossain, Khan Md Hasib, Fairuz Shadmani Shishir","doi":"10.1109/IEMCON51383.2020.9284926","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284926","url":null,"abstract":"In the present world, the commotion centering Big Data is somewhat obscuring the craft of mining information from smaller samples. Populations with limited examples but huge dimensionality are a common phenomenon, otherwise known as the curse of dimensionality-especially in the health sector-thanks to the recently-discovered potential of data mining and the enthusiasm for feature engineering. Correlated, noisy, redundant features are byproducts of this tendency, which makes learning algorithms converge with greater efforts. This paper proposes a novel feature-pruning technique relying on computational graph theory. Restoring the appeal of pre-AI conventional computing, the paper applies Disjoint Set Union (DSU) on unidirectional graphs prepared basis thresholded Spearman's rank correlation coefficient, r. Gradual withdrawal of leniency on Spearman's r caused a greater tendency in features to form clusters, causing the dimensionality to shrink. The results-extracting out finer, more representative roots as features-have been $k$-fold cross-validated on a case study examining subjects for Parkinson's. Qualitatively, the method overcomes Principal Component Analysis's (PCA) limitation of inexplicit merging of features and Linear Discriminant Analysis's (LDA) limitation of inextendibility to multiple classes. Statistical inference verified a significant rise in performance, establishing an example of conventional hard computing reinforcing modern soft computing.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"36 1","pages":"0134-0140"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88565664","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284828
A. Pundir, L. Ganapathy, Pratik Maheshwari, Shashikant Thakur
The purpose of this paper is to identify, analyze the enablers for blockchain technology in supply chain. We have proposed Interpretive Structural Modeling (ISM) to analyze the different types of enablers like autonomous enablers, dependent enablers, linkage enablers and driver enablers from SSIM (Structural Self-Interaction Matrix) in MATLAB software by using partition level and iterations for the prioritization of enablers. We have prioritized the different enablers and proposed cluster diagram of enablers for blockchain technology in supply chain. On the basis of our analysis, we have formed five clusters and found that traceability transparency, seamless connectivity, verifiability of transaction enablers are highly driven and dependent on the other input variables included in the supply chain system. The present work suggested the platform for both academicians and researchers to understand the relationship between enablers of blockchain technology in supply chain. This paper also provides the future direction to the practitioners for optimally assign the efforts and available resources to increase the current performance of supply chain system. This article prioritizes the enablers of block chain in clusters according to their level of impact.
{"title":"Interpretive Structural Modelling to assess the enablers of blockchain technology in supply chain","authors":"A. Pundir, L. Ganapathy, Pratik Maheshwari, Shashikant Thakur","doi":"10.1109/IEMCON51383.2020.9284828","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284828","url":null,"abstract":"The purpose of this paper is to identify, analyze the enablers for blockchain technology in supply chain. We have proposed Interpretive Structural Modeling (ISM) to analyze the different types of enablers like autonomous enablers, dependent enablers, linkage enablers and driver enablers from SSIM (Structural Self-Interaction Matrix) in MATLAB software by using partition level and iterations for the prioritization of enablers. We have prioritized the different enablers and proposed cluster diagram of enablers for blockchain technology in supply chain. On the basis of our analysis, we have formed five clusters and found that traceability transparency, seamless connectivity, verifiability of transaction enablers are highly driven and dependent on the other input variables included in the supply chain system. The present work suggested the platform for both academicians and researchers to understand the relationship between enablers of blockchain technology in supply chain. This paper also provides the future direction to the practitioners for optimally assign the efforts and available resources to increase the current performance of supply chain system. This article prioritizes the enablers of block chain in clusters according to their level of impact.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"19 1","pages":"0223-0229"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77687026","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284872
Anirban Dutta, G. Prabhakar, C. V. R. Rao
Spectro-temporal features have recently showed its usefulness for various speech recognition tasks. Two Dimensional (2-D) Gabor filters are characterized by local patches of spectro-temporal fields which allows them to extract the spectro-temporal information from speech samples. The state of art spectro-temporal features uses the real part of the Gabor filters without any phase offset value. In this work, we analyzed to see whether the Gabor phase have any relevance in the context of spectro-temporal feature extraction in building the acoustic module of a hybrid Automatic Speech Recognition (ASR) system. Different phase offset values are investigated to see if it carries equivalent or complementary information. The experiments are carried out using TIMIT dataset corrupted with different noises at various SNR values. It is found that a Gabor offset phase of 0 degree and 90 degree is equally important in the Gabor filter design for building a robust ASR system.
{"title":"On the Impact of Gabor Phase for Spectro-Temporal Feature Extraction in Building an ASR System","authors":"Anirban Dutta, G. Prabhakar, C. V. R. Rao","doi":"10.1109/IEMCON51383.2020.9284872","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284872","url":null,"abstract":"Spectro-temporal features have recently showed its usefulness for various speech recognition tasks. Two Dimensional (2-D) Gabor filters are characterized by local patches of spectro-temporal fields which allows them to extract the spectro-temporal information from speech samples. The state of art spectro-temporal features uses the real part of the Gabor filters without any phase offset value. In this work, we analyzed to see whether the Gabor phase have any relevance in the context of spectro-temporal feature extraction in building the acoustic module of a hybrid Automatic Speech Recognition (ASR) system. Different phase offset values are investigated to see if it carries equivalent or complementary information. The experiments are carried out using TIMIT dataset corrupted with different noises at various SNR values. It is found that a Gabor offset phase of 0 degree and 90 degree is equally important in the Gabor filter design for building a robust ASR system.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"25 1","pages":"0341-0345"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82683926","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}
Joint activity recognition and localization plays an important role in many fields such as smart healthcare system, smart home, human-computer interaction, and robotics. Ultrawideband (UWB) is considered as a promising technology for high-precision indoor positioning system. But few studies have been done to simultaneously recognize and localize human activities based on the UWB indoor positioning system. In this paper, the possibility of simultaneously recognizing and localizing human activities with a self-developed UWB indoor positioning system is investigated. First, a few signal processing and machine learning techniques are applied to improve the positioning accuracy of the UWB indoor positioning system. Three machine learning methods based on support vector machine, artificial neural network, and hidden Markov model are then used to recognize five types of human activities based on the range measurements from the UWB indoor positioning system. Experimental results show that our approach achieves satisfactory performances in the joint activity recognition and localization task.
{"title":"Activity Recognition and Localization based on UWB Indoor Positioning System and Machine Learning","authors":"Long Cheng, Anguo Zhao, Kexin Wang, Hengguang Li, Yifan Wang, Ruofei Chang","doi":"10.1109/IEMCON51383.2020.9284937","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284937","url":null,"abstract":"Joint activity recognition and localization plays an important role in many fields such as smart healthcare system, smart home, human-computer interaction, and robotics. Ultrawideband (UWB) is considered as a promising technology for high-precision indoor positioning system. But few studies have been done to simultaneously recognize and localize human activities based on the UWB indoor positioning system. In this paper, the possibility of simultaneously recognizing and localizing human activities with a self-developed UWB indoor positioning system is investigated. First, a few signal processing and machine learning techniques are applied to improve the positioning accuracy of the UWB indoor positioning system. Three machine learning methods based on support vector machine, artificial neural network, and hidden Markov model are then used to recognize five types of human activities based on the range measurements from the UWB indoor positioning system. Experimental results show that our approach achieves satisfactory performances in the joint activity recognition and localization task.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"26 1","pages":"0528-0533"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84096516","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284891
Delica S. Leboe-McGowan, Md Momin Al Aziz, N. Mohammed
State-of-the-art frameworks for privacy-preserving artificial neural networks often rely on secret sharing to protect sensitive data. Unfortunately, operating on secret shared data complicates a number of non-linear functions that are central to deep learning, such as batch normalization and rectified linear units (ReLUs). We offer simple procedures for approximating these non-linear operations. The approximations we propose significantly reduce the training runtime of a privacy-preserving convolutional neural network (CNN) that we designed to diagnose breast cancer from secret shared gene expression profiles. In just over five minutes of training, our approximation-based privacy-preserving CNN achieves an average test accuracy of 96%. When we apply an exact garbled circuit solution for the ReLU function, we find that the privacy-preserving model requires days of computation to achieve the same level of accuracy. The dramatic improvement in training runtime yielded by our ReLU approximation may prove useful for other medical applications of privacy-preserving neural networks.
{"title":"Simple Approximations for Fast and Secure Deep Learning on Genomic Data","authors":"Delica S. Leboe-McGowan, Md Momin Al Aziz, N. Mohammed","doi":"10.1109/IEMCON51383.2020.9284891","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284891","url":null,"abstract":"State-of-the-art frameworks for privacy-preserving artificial neural networks often rely on secret sharing to protect sensitive data. Unfortunately, operating on secret shared data complicates a number of non-linear functions that are central to deep learning, such as batch normalization and rectified linear units (ReLUs). We offer simple procedures for approximating these non-linear operations. The approximations we propose significantly reduce the training runtime of a privacy-preserving convolutional neural network (CNN) that we designed to diagnose breast cancer from secret shared gene expression profiles. In just over five minutes of training, our approximation-based privacy-preserving CNN achieves an average test accuracy of 96%. When we apply an exact garbled circuit solution for the ReLU function, we find that the privacy-preserving model requires days of computation to achieve the same level of accuracy. The dramatic improvement in training runtime yielded by our ReLU approximation may prove useful for other medical applications of privacy-preserving neural networks.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"41 1","pages":"0860-0866"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81677778","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 : 2020-11-04DOI: 10.1109/IEMCON51383.2020.9284847
Charan Gudla, A. Sung
Software-Defined Networking (SDN) dissociates the control plane from the data plane, creating a central point facilitating managed services and network virtualization. SDN allows flexibility by dynamic programmability features. On the other hand, Moving Target Defense (MTD) increases complexity in the network to prevent or delay attacks by continuously creating and adapting to the dynamic environment. In this paper, we implement a Moving Target Defense technique in Software-Defined Networking and analyze the TCP and UDP traffic generated in the network. MTD implementation in SDN has been well studied, but there is little work to analyze and evaluate the impact of this dynamic environment on SDN performance. The network's dynamic nature creates considerable overhead on the controller, resulting in poor performance, latency, jitter, and packet loss. This paper analyzes MTD traffic implemented in a software-defined network and compares the results with the traditional (without MTD) software-defined network topology.
SDN (software defined Networking)将控制平面和数据平面分离开来，形成一个中心点，便于管理业务和网络虚拟化。SDN通过动态可编程特性实现灵活性。另一方面，移动目标防御(MTD)增加了网络的复杂性，通过不断创造和适应动态环境来防止或延迟攻击。本文在软件定义网络中实现了一种移动目标防御技术，并对网络中产生的TCP和UDP流量进行了分析。MTD在SDN中的实现已经得到了很好的研究，但是分析和评估这种动态环境对SDN性能的影响的工作很少。网络的动态特性会给控制器带来相当大的开销，从而导致性能差、延迟、抖动和丢包。本文分析了在软件定义网络中实现的MTD流量，并将结果与传统的(没有MTD的)软件定义网络拓扑进行了比较。
{"title":"Moving Target Defense Application and Analysis in Software-Defined Networking","authors":"Charan Gudla, A. Sung","doi":"10.1109/IEMCON51383.2020.9284847","DOIUrl":"https://doi.org/10.1109/IEMCON51383.2020.9284847","url":null,"abstract":"Software-Defined Networking (SDN) dissociates the control plane from the data plane, creating a central point facilitating managed services and network virtualization. SDN allows flexibility by dynamic programmability features. On the other hand, Moving Target Defense (MTD) increases complexity in the network to prevent or delay attacks by continuously creating and adapting to the dynamic environment. In this paper, we implement a Moving Target Defense technique in Software-Defined Networking and analyze the TCP and UDP traffic generated in the network. MTD implementation in SDN has been well studied, but there is little work to analyze and evaluate the impact of this dynamic environment on SDN performance. The network's dynamic nature creates considerable overhead on the controller, resulting in poor performance, latency, jitter, and packet loss. This paper analyzes MTD traffic implemented in a software-defined network and compares the results with the traditional (without MTD) software-defined network topology.","PeriodicalId":6871,"journal":{"name":"2020 11th IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)","volume":"32 1","pages":"0641-0646"},"PeriodicalIF":0.0,"publicationDate":"2020-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88773974","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：