Pub Date : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976089
Chen Zhang, Lifeng Wu
Accurately predicting the battery cycle life of lithium-ion batteries in the early-cycle stage can provide a basis for long-term planning, bring economic benefits and avoid safety risks. However, it is very difficult to accurately predict the cycle life due to the weak degradation of battery performance in the early cycle stage. In this paper, an early stage prediction model of lithium-ion battery based on convolutional long short-term memory (ConvLSTM) with attention mechanism is proposed, which is called ConvLSTM-Attention model. ConvLSTM can not only extract the characteristics of single cycle information, but also mine the temporal relationship among each cycle data. For the features extracted by ConvLSTM, the attention mechanism is added, so that the model can pay attention to the important features and thus improve the prediction accuracy of the model. Experiments show that the model can predict the battery cycle life only by using the information of the first 10 cycles of the battery, and the model can predict whether the battery belongs to high-lifetime or low-lifetime only by using the information of the first 5 cycles of the battery. Comparison with other early prediction models show that the proposed model can achieve better prediction results by using less cycle data.
准确预测锂离子电池循环初期的电池循环寿命，可以为长期规划提供依据，带来经济效益，避免安全风险。然而，由于电池在循环初期性能下降较弱，因此很难准确预测电池的循环寿命。本文提出了一种基于卷积长短期记忆(convolutional long - short- memory, ConvLSTM)和注意机制的锂离子电池早期预测模型，称为ConvLSTM- attention模型。ConvLSTM不仅可以提取单周期信息的特征，还可以挖掘各周期数据之间的时间关系。对于ConvLSTM提取的特征，加入了注意机制，使模型能够注意到重要的特征，从而提高模型的预测精度。实验表明，该模型仅能利用电池前10次循环的信息来预测电池的循环寿命，仅能利用电池前5次循环的信息来预测电池是属于高寿命还是低寿命。与其他早期预测模型的比较表明，该模型使用更少的周期数据可以获得更好的预测结果。
{"title":"Life prediction model of lithium-ion batteries in the early-cycle stage based on convolutional long short-term memory with attention mechanism","authors":"Chen Zhang, Lifeng Wu","doi":"10.1109/INDIN51773.2022.9976089","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976089","url":null,"abstract":"Accurately predicting the battery cycle life of lithium-ion batteries in the early-cycle stage can provide a basis for long-term planning, bring economic benefits and avoid safety risks. However, it is very difficult to accurately predict the cycle life due to the weak degradation of battery performance in the early cycle stage. In this paper, an early stage prediction model of lithium-ion battery based on convolutional long short-term memory (ConvLSTM) with attention mechanism is proposed, which is called ConvLSTM-Attention model. ConvLSTM can not only extract the characteristics of single cycle information, but also mine the temporal relationship among each cycle data. For the features extracted by ConvLSTM, the attention mechanism is added, so that the model can pay attention to the important features and thus improve the prediction accuracy of the model. Experiments show that the model can predict the battery cycle life only by using the information of the first 10 cycles of the battery, and the model can predict whether the battery belongs to high-lifetime or low-lifetime only by using the information of the first 5 cycles of the battery. Comparison with other early prediction models show that the proposed model can achieve better prediction results by using less cycle data.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128774599","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976165
Ana P. Lopes, Daniel F. Silva, S. Lopes, J. H. Correia, Carlos S. Lima, Carlos Alberto Silva
A graphical integrated development environment (IDE) for computer vision applications allows developing solutions by composing graphical widgets that represent operators of a computer vision library. A challenge in developing such IDE is the development of a graphical interface for each operator in the library, which is a slow and repetitive task. In this paper, we propose to generate a specific graphical widget editor for the input parameters of each operator, based directly on the library documentation. Our approach allows reducing significantly the development time of an IDE. The only assumption of the proposed approach is that the documentation has a structured format. We validated our approach by integrating the computer vision library Halcon in an IDE, using only its HTML documentation.
{"title":"Documentation-driven GUI development for integration of image processing libraries","authors":"Ana P. Lopes, Daniel F. Silva, S. Lopes, J. H. Correia, Carlos S. Lima, Carlos Alberto Silva","doi":"10.1109/INDIN51773.2022.9976165","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976165","url":null,"abstract":"A graphical integrated development environment (IDE) for computer vision applications allows developing solutions by composing graphical widgets that represent operators of a computer vision library. A challenge in developing such IDE is the development of a graphical interface for each operator in the library, which is a slow and repetitive task. In this paper, we propose to generate a specific graphical widget editor for the input parameters of each operator, based directly on the library documentation. Our approach allows reducing significantly the development time of an IDE. The only assumption of the proposed approach is that the documentation has a structured format. We validated our approach by integrating the computer vision library Halcon in an IDE, using only its HTML documentation.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128967944","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976116
Yinghe Qing, Jifeng Sun, Ying Kong, Jianwu Lin
This paper investigates how to use deep learning methods to combine with traditional multi-factor models and construct a quantitative trading model based on an AutoEncoder algorithm (AE) to classify cryptocurrencies since 2009, so as to screen out ones with investment value and then construct an effective investment portfolio. The AE algorithm is capable of handling high-dimensional data and mining interfactor non-linearities. Our empirical results on cryptocurrencies show that the model outperforms single-type factors and benchmark in terms of Cumulative Returns and the Sharpe Ratio.
{"title":"Fundamental Multi-factor Deep-learning Strategy For Cryptocurrency Trading","authors":"Yinghe Qing, Jifeng Sun, Ying Kong, Jianwu Lin","doi":"10.1109/INDIN51773.2022.9976116","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976116","url":null,"abstract":"This paper investigates how to use deep learning methods to combine with traditional multi-factor models and construct a quantitative trading model based on an AutoEncoder algorithm (AE) to classify cryptocurrencies since 2009, so as to screen out ones with investment value and then construct an effective investment portfolio. The AE algorithm is capable of handling high-dimensional data and mining interfactor non-linearities. Our empirical results on cryptocurrencies show that the model outperforms single-type factors and benchmark in terms of Cumulative Returns and the Sharpe Ratio.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130367712","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976107
S. Rädler, E. Rigger, Juergen Mangler, S. Rinderle-Ma
In order to allow Systems Engineers to utilize data produced in cyber-physical systems (CPS), they have to cooperate with data-scientists for custom data-extraction, data-preparation, and/or data-transformation mechanisms. While interfaces in CPS systems might be generic, the data that is produced for custom application needs has to be transformed and merged in very specific ways, to allow systems engineers proper interpretation and insight-extraction. In order to enable efficient cooperation between systems engineers and data scientists, the systems engineers have to provide a fine-grained specification that (a) describes all parts of the CPS, (b) how they might interact, (c) what data is exchanged between them, and (d) how the data inter-relates. A data scientists can then iteratively (including further refinements of the specification) prepare the necessary custom machine-learning models and components. Therefore, this work introduces a method supporting the collaborative definition of machine learning tasks by leveraging model-based systems engineering in the formalization of the systems modeling language SysML. The method supports the identification and integration of various data sources, the required definition of semantic connections between data attributes and the definition of the data processing steps within the machine learning support. Integrating machine learning-specific properties in systems engineering techniques allows non-data scientists to define a machine learning problem, document knowledge on the data, and further supports data scientists to use the formalized knowledge as input for an implementation.
{"title":"Integration of Machine Learning Task Definition in Model-Based Systems Engineering using SysML","authors":"S. Rädler, E. Rigger, Juergen Mangler, S. Rinderle-Ma","doi":"10.1109/INDIN51773.2022.9976107","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976107","url":null,"abstract":"In order to allow Systems Engineers to utilize data produced in cyber-physical systems (CPS), they have to cooperate with data-scientists for custom data-extraction, data-preparation, and/or data-transformation mechanisms. While interfaces in CPS systems might be generic, the data that is produced for custom application needs has to be transformed and merged in very specific ways, to allow systems engineers proper interpretation and insight-extraction. In order to enable efficient cooperation between systems engineers and data scientists, the systems engineers have to provide a fine-grained specification that (a) describes all parts of the CPS, (b) how they might interact, (c) what data is exchanged between them, and (d) how the data inter-relates. A data scientists can then iteratively (including further refinements of the specification) prepare the necessary custom machine-learning models and components. Therefore, this work introduces a method supporting the collaborative definition of machine learning tasks by leveraging model-based systems engineering in the formalization of the systems modeling language SysML. The method supports the identification and integration of various data sources, the required definition of semantic connections between data attributes and the definition of the data processing steps within the machine learning support. Integrating machine learning-specific properties in systems engineering techniques allows non-data scientists to define a machine learning problem, document knowledge on the data, and further supports data scientists to use the formalized knowledge as input for an implementation.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134489936","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976073
Qing Tang, K. Jo
This paper studies the fully unsupervised object re-identification (re-ID) problem which can learn re-ID without any human-annotated labeled data. Recent works show that self-supervised momentum contrastive learning is an effective method for unsupervised object re-ID, but they neglect to optimize one important component - the similarity relationships among instances. Previous works focus on enforcing instance-to-centroid learning, which does not fully utilize the inter-instances information. Thus, we propose an Instances Correlation Loss (ICL) to enforce instance-to-instance learning in each training iteration. Experimental results show that the proposed ICL effectively boost the performance, which demonstrates that learning strategy is also a central importance to unsupervised re-ID task. Extensive experiments are performed on three mainstream person re-ID datasets and one vehicle re-ID dataset.
{"title":"Unsupervised Object Re-identification via Instances Correlation Loss","authors":"Qing Tang, K. Jo","doi":"10.1109/INDIN51773.2022.9976073","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976073","url":null,"abstract":"This paper studies the fully unsupervised object re-identification (re-ID) problem which can learn re-ID without any human-annotated labeled data. Recent works show that self-supervised momentum contrastive learning is an effective method for unsupervised object re-ID, but they neglect to optimize one important component - the similarity relationships among instances. Previous works focus on enforcing instance-to-centroid learning, which does not fully utilize the inter-instances information. Thus, we propose an Instances Correlation Loss (ICL) to enforce instance-to-instance learning in each training iteration. Experimental results show that the proposed ICL effectively boost the performance, which demonstrates that learning strategy is also a central importance to unsupervised re-ID task. Extensive experiments are performed on three mainstream person re-ID datasets and one vehicle re-ID dataset.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131759230","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976142
Dapeng Zhang, David Zhiwei Gao
Aiming to overcome the difficulty to obtain the fault data of practical system, a fault detection approach using health data only is proposed based on the whole space of the system being divided into the fault status and the fault-free status. Firstly the time series of observation window is generated by a deconvolutional network with an input of initial data obtained by Monte Carlo method. The probability distribution of generated data approximates to the actual sample data by discriminator of generative adversarial network. Through continuous iteration, the health probability distribution is finally obtained in the whole space. Concurrently the discriminator is evolved into a fault detector which realizes the detection of new data. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.
{"title":"A GAN-based fault detection for dynamic process with deconvolutional networks","authors":"Dapeng Zhang, David Zhiwei Gao","doi":"10.1109/INDIN51773.2022.9976142","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976142","url":null,"abstract":"Aiming to overcome the difficulty to obtain the fault data of practical system, a fault detection approach using health data only is proposed based on the whole space of the system being divided into the fault status and the fault-free status. Firstly the time series of observation window is generated by a deconvolutional network with an input of initial data obtained by Monte Carlo method. The probability distribution of generated data approximates to the actual sample data by discriminator of generative adversarial network. Through continuous iteration, the health probability distribution is finally obtained in the whole space. Concurrently the discriminator is evolved into a fault detector which realizes the detection of new data. The effectiveness of the algorithm is demonstrated by a numerical simulation example based on a wind turbine benchmark model.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123852126","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976118
Snigdha Mohanty, J. Abinahed, A. Al-Ansari, S. Mishra, S. Singh, S. Dakua
The delineation of liver difficult due to its similar intensity distributions in CT images. Additionally, there have been other challenges such that the variability in shape, size, and proximity to the other neighboring organs. The blurred liver edges and low contrast on the CT image make the segmentation further challenging. Furthermore, the patient movement during CT data acquisition along with spatial averaging lead to reconstruction artifacts; these are all reflected on the CT image complicating the segmentation task. In this paper, we have proposed a UNet-based automatic liver segmentation approach to delineate the boundaries between the liver and other abdominal organs. The algorithm is tested on publicly available datasets. The average values of Dice similarity coefficient (DC), Relative absolute volume difference (RAVD), Average symmetric surface distance (ASSD), Maximum symmetric surface distance (MSSD), Hausdorff distance (HD), and Precision are found to be 0.95±0.02, 0.04±0.02, 1.03±0.39, 1.15±0.5, 2.85±1.89, and 0.91±0.12, respectively.
{"title":"Towards Developing a Liver Segmentation Method for Hepatocellular Carcinoma Treatment Planning","authors":"Snigdha Mohanty, J. Abinahed, A. Al-Ansari, S. Mishra, S. Singh, S. Dakua","doi":"10.1109/INDIN51773.2022.9976118","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976118","url":null,"abstract":"The delineation of liver difficult due to its similar intensity distributions in CT images. Additionally, there have been other challenges such that the variability in shape, size, and proximity to the other neighboring organs. The blurred liver edges and low contrast on the CT image make the segmentation further challenging. Furthermore, the patient movement during CT data acquisition along with spatial averaging lead to reconstruction artifacts; these are all reflected on the CT image complicating the segmentation task. In this paper, we have proposed a UNet-based automatic liver segmentation approach to delineate the boundaries between the liver and other abdominal organs. The algorithm is tested on publicly available datasets. The average values of Dice similarity coefficient (DC), Relative absolute volume difference (RAVD), Average symmetric surface distance (ASSD), Maximum symmetric surface distance (MSSD), Hausdorff distance (HD), and Precision are found to be 0.95±0.02, 0.04±0.02, 1.03±0.39, 1.15±0.5, 2.85±1.89, and 0.91±0.12, respectively.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127766778","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}
Intelligent transformation for traditional factories is a widely discussed topic. The key to this transformation is ensuring the integration between information technology and operational technology. However, it is a challenging task in industry owing to the communication heterogeneity of the underlying production equipment (horizontal communication), and inefficient interactions between the equipment and information decision center (vertical communication). In this paper, we explore asset administration shell (AAS), an asset virtualization technology, shielding heterogeneous physical communication protocol of production equipment. Besides, to promote inefficient communication between the equipment and information decision center, we adapt OPC UA protocol as the communication protocol of AAS for vertical communication. In addition, time-sensitive networking (TSN) is applied to ensure communication between the AAS and the corresponding physical device. Above operations ensure devices interconnection and interoperability. On this basis, we propose an AAS-based production management platform (AASPMP), which aims at the coverage from the demand side to the production side. Such an intelligent system characterizes three layers to decompose complicated system functionalities, and a visible client is provided for the convenience of remote operation and maintenance. We deploy our system on the actual production system and demonstrate the effectiveness of our design.
{"title":"AASPMP: Design and Implementation of Production Management Platform Based on AAS","authors":"Qihang Zhou, Yihao Wu, Chaojie Gu, Wenchao Meng, Shibo He, Zhiguo Shi","doi":"10.1109/INDIN51773.2022.9976146","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976146","url":null,"abstract":"Intelligent transformation for traditional factories is a widely discussed topic. The key to this transformation is ensuring the integration between information technology and operational technology. However, it is a challenging task in industry owing to the communication heterogeneity of the underlying production equipment (horizontal communication), and inefficient interactions between the equipment and information decision center (vertical communication). In this paper, we explore asset administration shell (AAS), an asset virtualization technology, shielding heterogeneous physical communication protocol of production equipment. Besides, to promote inefficient communication between the equipment and information decision center, we adapt OPC UA protocol as the communication protocol of AAS for vertical communication. In addition, time-sensitive networking (TSN) is applied to ensure communication between the AAS and the corresponding physical device. Above operations ensure devices interconnection and interoperability. On this basis, we propose an AAS-based production management platform (AASPMP), which aims at the coverage from the demand side to the production side. Such an intelligent system characterizes three layers to decompose complicated system functionalities, and a visible client is provided for the convenience of remote operation and maintenance. We deploy our system on the actual production system and demonstrate the effectiveness of our design.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121461900","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976091
Jia Miao, Jianwu Lin, Shenglei Hu, Guangling Liu
In the Internet era, due to the rapid development of investors communication with public companies, people have diversified ways to express their opinions, thus generating a large amount of data, which contains valuable information. In this paper, we use a combination of the financial sentiment dictionary and Bert to analyze the sentiment of investors’ questions based on the Q&R data of board secretaries on the platform "Easy Interactive" (http://irm.cninfo.com.cn/) launched by Shenzhen Stock Exchange, and the final accuracy rate is 92%, which is 16% higher than the traditional sentiment analysis methods. Compared with offline research, financial news, stock forums, social software, and other data, the Q&R data selected in this paper has less noise and is more intuitive. Moreover, this paper considers knowledge in the financial domain in sentiment analysis and has domain friendliness and model generalization in the financial domain by combining the financial domain sentiment lexicon with the Bert model with adversarial training.
{"title":"Sentiment Analysis of Board Secretaries’ Q&R Data","authors":"Jia Miao, Jianwu Lin, Shenglei Hu, Guangling Liu","doi":"10.1109/INDIN51773.2022.9976091","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976091","url":null,"abstract":"In the Internet era, due to the rapid development of investors communication with public companies, people have diversified ways to express their opinions, thus generating a large amount of data, which contains valuable information. In this paper, we use a combination of the financial sentiment dictionary and Bert to analyze the sentiment of investors’ questions based on the Q&R data of board secretaries on the platform \"Easy Interactive\" (http://irm.cninfo.com.cn/) launched by Shenzhen Stock Exchange, and the final accuracy rate is 92%, which is 16% higher than the traditional sentiment analysis methods. Compared with offline research, financial news, stock forums, social software, and other data, the Q&R data selected in this paper has less noise and is more intuitive. Moreover, this paper considers knowledge in the financial domain in sentiment analysis and has domain friendliness and model generalization in the financial domain by combining the financial domain sentiment lexicon with the Bert model with adversarial training.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114252923","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 : 2022-07-25DOI: 10.1109/INDIN51773.2022.9976093
A. Gkillas, A. Lalos
In multidimensional times series generated by sensor recordings of multiple dispersed IoT edge devices, missing measurements are commonplace due to sensing or communication failures, considered a thorny and perplexing problem in a wide range of Industry 4.0 applications. Existing studies for time series imputation focus on developing centralized deep learning approaches, which require massive amounts of data to be uploaded to a central server with adequate computational and power resources for the training of the models, since these approaches are unsuitable for edge and IoT devices characterized by limited computation resources. Different from the current literature, in this study, the time series imputation problem is studied from a federated learning perspective, which is able to surmount the above difficulties. In particular, a novel federated learning approach is proposed, assuming different IoT devices with varying sensing and computational capabilities, that trade-off accuracy with computational/communication/sensing complexity and minimize the operations that need to be performed during training and inferences phase. Furthermore, considering that the main computations are performed on the edge, where the IoT edge devices have limited computational capabilities and power resources, a lightweight yet effective autoencoder-based model is employed to address the examined problem, modified properly to capture the temporal dependencies of the time series data. Extensive evaluation studies with two open datasets have shown that both approaches minimize the data exchanges the need to be made for outperforming centralized approaches in the presence of limited training data.
{"title":"Missing Data Imputation for Multivariate Time series in Industrial IoT: A Federated Learning Approach","authors":"A. Gkillas, A. Lalos","doi":"10.1109/INDIN51773.2022.9976093","DOIUrl":"https://doi.org/10.1109/INDIN51773.2022.9976093","url":null,"abstract":"In multidimensional times series generated by sensor recordings of multiple dispersed IoT edge devices, missing measurements are commonplace due to sensing or communication failures, considered a thorny and perplexing problem in a wide range of Industry 4.0 applications. Existing studies for time series imputation focus on developing centralized deep learning approaches, which require massive amounts of data to be uploaded to a central server with adequate computational and power resources for the training of the models, since these approaches are unsuitable for edge and IoT devices characterized by limited computation resources. Different from the current literature, in this study, the time series imputation problem is studied from a federated learning perspective, which is able to surmount the above difficulties. In particular, a novel federated learning approach is proposed, assuming different IoT devices with varying sensing and computational capabilities, that trade-off accuracy with computational/communication/sensing complexity and minimize the operations that need to be performed during training and inferences phase. Furthermore, considering that the main computations are performed on the edge, where the IoT edge devices have limited computational capabilities and power resources, a lightweight yet effective autoencoder-based model is employed to address the examined problem, modified properly to capture the temporal dependencies of the time series data. Extensive evaluation studies with two open datasets have shown that both approaches minimize the data exchanges the need to be made for outperforming centralized approaches in the presence of limited training data.","PeriodicalId":359190,"journal":{"name":"2022 IEEE 20th International Conference on Industrial Informatics (INDIN)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126371794","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：