Stylianos Karagiannis, E. Magkos, Christoforos Ntantogian, Ricardo Cabecinha, Theo Fotis
Medical imaging plays a crucial role in modern healthcare, providing essential information for accurate diagnosis and treatment planning. The Digital Imaging and Communications in Medicine (DICOM) standard has revolutionized the storage, transmission, and sharing of medical images and related data. Despite its advantages, implementation and deployment of the DICOM protocol often suffers from incomplete understanding, leading to vulnerabilities within the healthcare ecosystem. This research paper presents an implementation of DICOM communication and the development of a practical demonstration for simulation purposes The simulation can be used for conducting cybersecurity tests in the context of DICOM communication. Overall, the simulation provides a digital environment that can help in retrieving valuable insights into the practical aspects of DICOM communication and PACS integration, serving as a valuable resource for medical imaging professionals, researchers, and developers. These research results provide practical insights, and the DICOM simulation can be used in realistic contexts to showcase a variety of security scenarios.
{"title":"Cybersecurity and Medical Imaging: A Simulation-Based Approach to DICOM Communication","authors":"Stylianos Karagiannis, E. Magkos, Christoforos Ntantogian, Ricardo Cabecinha, Theo Fotis","doi":"10.3390/app131810072","DOIUrl":"https://doi.org/10.3390/app131810072","url":null,"abstract":"Medical imaging plays a crucial role in modern healthcare, providing essential information for accurate diagnosis and treatment planning. The Digital Imaging and Communications in Medicine (DICOM) standard has revolutionized the storage, transmission, and sharing of medical images and related data. Despite its advantages, implementation and deployment of the DICOM protocol often suffers from incomplete understanding, leading to vulnerabilities within the healthcare ecosystem. This research paper presents an implementation of DICOM communication and the development of a practical demonstration for simulation purposes The simulation can be used for conducting cybersecurity tests in the context of DICOM communication. Overall, the simulation provides a digital environment that can help in retrieving valuable insights into the practical aspects of DICOM communication and PACS integration, serving as a valuable resource for medical imaging professionals, researchers, and developers. These research results provide practical insights, and the DICOM simulation can be used in realistic contexts to showcase a variety of security scenarios.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43053940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper addresses the issue of course keeping control (CKC) for unmanned surface vehicles (USVs) under network environments, where various challenges, such as network resource constraints and discontinuities of course and yaw caused by data transmission, are taken into account. To tackle the issue of network resource constraints, an event-sampled scheme is developed to obtain the course data, and a novel event-sampled adaptive neural-network-based state observer (NN–SO) is developed to achieve the state reconstruction of discontinuous yaw. Using a backstepping design method, an event-sampled mechanism, and an adaptive NN–SO, an adaptive neural output feedback (ANOF) control law is designed, where the dynamic surface control technique is introduced to solve the design issue caused by the intermission course data. Moreover, an event-triggered mechanism (ETM) is established in a controller–actuator (C–A) channel and a dual-channel event-triggered adaptive neural output feedback control (ETANOFC) solution is proposed. The theoretical results show that all signals in the closed-loop control system (CLCS) are bounded. The effectiveness is verified through numerical simulations.
{"title":"Event-Sampled Adaptive Neural Course Keeping Control for USVs Using Intermittent Course Data","authors":"Hongyang Zhi, Baofeng Pan, Guibing Zhu","doi":"10.3390/app131810035","DOIUrl":"https://doi.org/10.3390/app131810035","url":null,"abstract":"This paper addresses the issue of course keeping control (CKC) for unmanned surface vehicles (USVs) under network environments, where various challenges, such as network resource constraints and discontinuities of course and yaw caused by data transmission, are taken into account. To tackle the issue of network resource constraints, an event-sampled scheme is developed to obtain the course data, and a novel event-sampled adaptive neural-network-based state observer (NN–SO) is developed to achieve the state reconstruction of discontinuous yaw. Using a backstepping design method, an event-sampled mechanism, and an adaptive NN–SO, an adaptive neural output feedback (ANOF) control law is designed, where the dynamic surface control technique is introduced to solve the design issue caused by the intermission course data. Moreover, an event-triggered mechanism (ETM) is established in a controller–actuator (C–A) channel and a dual-channel event-triggered adaptive neural output feedback control (ETANOFC) solution is proposed. The theoretical results show that all signals in the closed-loop control system (CLCS) are bounded. The effectiveness is verified through numerical simulations.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47027343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Semi-supervised consensus clustering is a promising strategy to compensate for the subjectivity of clustering and its sensitivity to design factors, with various techniques being recently proposed to integrate domain knowledge and multiple clustering partitions. In this article, we present a new approach that makes double use of domain knowledge, namely to build the initial partitions, as well as to combine them. In particular, we show how to model and integrate must-link and cannot-link constraints into the objective function of a generic consensus clustering (CC) framework that maximizes the similarity between the consensus partition and the input partitions, which have, in turn, been enriched with the same constraints. In addition, borrowing from the theory of functional dependencies, the integrated framework exploits the notions of deductive closure and minimal cover to take full advantage of the logical implication between constraints. Using standard UCI benchmarks, we found that the resulting algorithm, termed CCC double-constrained consensus clustering), was more effective than plain CC at combining base-constrained partitions, with an average performance improvement of 5.54%. We then argue that CCC is especially well-suited for profiling counterfeit e-commerce websites, as constraints can be acquired by leveraging specific domain features, and demonstrate its potential for detecting affiliate marketing programs. Taken together, our experiments suggest that CCC makes the process of clustering more robust and able to withstand changes in clustering algorithms, datasets, and features, with a remarkable improvement in average performance.
{"title":"Double-Constrained Consensus Clustering with Application to Online Anti-Counterfeiting","authors":"Claudio Carpineto, Giovanni Romano","doi":"10.3390/app131810050","DOIUrl":"https://doi.org/10.3390/app131810050","url":null,"abstract":"Semi-supervised consensus clustering is a promising strategy to compensate for the subjectivity of clustering and its sensitivity to design factors, with various techniques being recently proposed to integrate domain knowledge and multiple clustering partitions. In this article, we present a new approach that makes double use of domain knowledge, namely to build the initial partitions, as well as to combine them. In particular, we show how to model and integrate must-link and cannot-link constraints into the objective function of a generic consensus clustering (CC) framework that maximizes the similarity between the consensus partition and the input partitions, which have, in turn, been enriched with the same constraints. In addition, borrowing from the theory of functional dependencies, the integrated framework exploits the notions of deductive closure and minimal cover to take full advantage of the logical implication between constraints. Using standard UCI benchmarks, we found that the resulting algorithm, termed CCC double-constrained consensus clustering), was more effective than plain CC at combining base-constrained partitions, with an average performance improvement of 5.54%. We then argue that CCC is especially well-suited for profiling counterfeit e-commerce websites, as constraints can be acquired by leveraging specific domain features, and demonstrate its potential for detecting affiliate marketing programs. Taken together, our experiments suggest that CCC makes the process of clustering more robust and able to withstand changes in clustering algorithms, datasets, and features, with a remarkable improvement in average performance.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47327134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A method for studying properties of the Earth’s surface tremor, measured by means of GPS, is proposed. The following tremor characteristics are considered: the entropy of wavelet coefficients, the Donoho–Johnston wavelet index, and two estimates of the spectral slope. The anomalous areas of tremor are determined by estimating the probability densities of extreme values of the studied properties. The criteria for abnormal tremor behavior are based on the proximity to, or the difference between, tremor properties and white noise. The greatest deviation from the properties of white noise is characterized by entropy minima and spectral slope and DJ index maxima. This behavior of the tremor is called “active”. The “passive” tremor behavior is characterized by the maximum proximity to the properties of white noise. The principal components approach provides weighted averaged density maps of these two variants of extreme distributions of parameters in a moving time window of 3 years. Singular points are the points of maximum average densities. The method is applied to the analysis of daily time series from a GPS network in California during the period 2009–2022. Singular points of tremor form well-defined clusters were found. The passive tremor could be caused by the activation of movement in fragments of the San Andreas fault.
{"title":"Singular Points of the Tremor of the Earth’s Surface","authors":"Alexey Lyubushin","doi":"10.3390/app131810060","DOIUrl":"https://doi.org/10.3390/app131810060","url":null,"abstract":"A method for studying properties of the Earth’s surface tremor, measured by means of GPS, is proposed. The following tremor characteristics are considered: the entropy of wavelet coefficients, the Donoho–Johnston wavelet index, and two estimates of the spectral slope. The anomalous areas of tremor are determined by estimating the probability densities of extreme values of the studied properties. The criteria for abnormal tremor behavior are based on the proximity to, or the difference between, tremor properties and white noise. The greatest deviation from the properties of white noise is characterized by entropy minima and spectral slope and DJ index maxima. This behavior of the tremor is called “active”. The “passive” tremor behavior is characterized by the maximum proximity to the properties of white noise. The principal components approach provides weighted averaged density maps of these two variants of extreme distributions of parameters in a moving time window of 3 years. Singular points are the points of maximum average densities. The method is applied to the analysis of daily time series from a GPS network in California during the period 2009–2022. Singular points of tremor form well-defined clusters were found. The passive tremor could be caused by the activation of movement in fragments of the San Andreas fault.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48212287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Owing to the special working systems of streak tube imaging lidar (STIL), the time and space dimensions are coupled together on the streak images. This coupling can cause measurement errors in 3D point clouds and can make measurement results more complicated to calibrate than other kinds of lidars. This paper presents a method to generate a time calibration array and an angle calibration array to separate the offset of the streak into time dimension and space dimension. The time and space information of the signal at any position on the streak image can be indexed through these two arrays. A validation experiment on aircraft was carried out, and the range error of the 3D point cloud was improved from 0.41 m to 0.27 m using the proposed calibration method. Thus, using the proposed calibration method can improve the accuracy of the point cloud produced by STIL.
{"title":"A Calibration Method for Time Dimension and Space Dimension of Streak Tube Imaging Lidar","authors":"Zhaodong Chen, Fangfang Shao, Zhigang Fan, Xing-shun Wang, Chaowei Dong, Zhi-wei Dong, R. Fan, Deying Chen","doi":"10.3390/app131810042","DOIUrl":"https://doi.org/10.3390/app131810042","url":null,"abstract":"Owing to the special working systems of streak tube imaging lidar (STIL), the time and space dimensions are coupled together on the streak images. This coupling can cause measurement errors in 3D point clouds and can make measurement results more complicated to calibrate than other kinds of lidars. This paper presents a method to generate a time calibration array and an angle calibration array to separate the offset of the streak into time dimension and space dimension. The time and space information of the signal at any position on the streak image can be indexed through these two arrays. A validation experiment on aircraft was carried out, and the range error of the 3D point cloud was improved from 0.41 m to 0.27 m using the proposed calibration method. Thus, using the proposed calibration method can improve the accuracy of the point cloud produced by STIL.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42561144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
An experiment was performed to investigate the movement time (MT) and subjective rating of difficulty for real and virtual pipe transferring tasks. Thirty adults joined as human participants. The HoloPipes app in a Microsoft® Hololens 2 augmented reality (AR) device was adopted to generate virtual pipes. The participants performed pipe transferring trials, from one location to another on a workbench, in both lateral and anterior–posterior directions. For the lateral transferring tasks, pipes in three diameters with three transferring distances and two origins were tested. For the anterior–posterior transferring tasks, pipes with a diameter of 2.2 cm with three transferring distances and two origins were tested. It was found that the MT of transferring a virtual pipe was significantly (p < 0.0001) shorter than that of transferring a real pipe. Moreover, male participants transferred the pipe significantly (p < 0.0001) faster than their female counterparts. Thus, the hypothesis that transferring a virtual pipe is less efficient than transferring a real pipe was rejected. It was also found that the MT of transferring both a real and a virtual object was dependent upon gender, handedness, and the transferring direction. In addition, the subjective rating of difficulty in pipe transferring is positively correlated (r = 0.48, p < 0.0001) with the MT. Based on Fitts’ law, additive MT models were proposed. These models could be used to predict the MT between handling real and virtual pipes under gender, handedness, and transferring direction conditions.
{"title":"Movement Time and Subjective Rating of Difficulty in Real and Virtual Pipe Transferring Tasks","authors":"Kaiway Li, Thi Lan Anh Nguyen","doi":"10.3390/app131810043","DOIUrl":"https://doi.org/10.3390/app131810043","url":null,"abstract":"An experiment was performed to investigate the movement time (MT) and subjective rating of difficulty for real and virtual pipe transferring tasks. Thirty adults joined as human participants. The HoloPipes app in a Microsoft® Hololens 2 augmented reality (AR) device was adopted to generate virtual pipes. The participants performed pipe transferring trials, from one location to another on a workbench, in both lateral and anterior–posterior directions. For the lateral transferring tasks, pipes in three diameters with three transferring distances and two origins were tested. For the anterior–posterior transferring tasks, pipes with a diameter of 2.2 cm with three transferring distances and two origins were tested. It was found that the MT of transferring a virtual pipe was significantly (p < 0.0001) shorter than that of transferring a real pipe. Moreover, male participants transferred the pipe significantly (p < 0.0001) faster than their female counterparts. Thus, the hypothesis that transferring a virtual pipe is less efficient than transferring a real pipe was rejected. It was also found that the MT of transferring both a real and a virtual object was dependent upon gender, handedness, and the transferring direction. In addition, the subjective rating of difficulty in pipe transferring is positively correlated (r = 0.48, p < 0.0001) with the MT. Based on Fitts’ law, additive MT models were proposed. These models could be used to predict the MT between handling real and virtual pipes under gender, handedness, and transferring direction conditions.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44630101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wenyi Hu, Wei Hong, Hongkun Wang, Meilin Liu, Shan Liu
In recent years, with the rapid development of artificial intelligence technology, computer vision-based pest detection technology has been widely used in agricultural production. Tomato diseases and pests are serious problems affecting tomato yield and quality, so it is important to detect them quickly and accurately. In this paper, we propose a tomato disease and pest detection model based on an improved YOLOv5n to overcome the problems of low accuracy and large model size in traditional pest detection methods. Firstly, we use the Efficient Vision Transformer as the feature extraction backbone network to reduce model parameters and computational complexity while improving detection accuracy, thus solving the problems of poor real-time performance and model deployment. Second, we replace the original nearest neighbor interpolation upsampling module with the lightweight general-purpose upsampling operator Content-Aware ReAssembly of FEatures to reduce feature information loss during upsampling. Finally, we use Wise-IoU instead of the original CIoU as the regression loss function of the target bounding box to improve the regression prediction accuracy of the predicted bounding box while accelerating the convergence speed of the regression loss function. We perform statistical analysis on the experimental results of tomato diseases and pests under data augmentation conditions. The results show that the improved algorithm improves mAP50 and mAP50:95 by 2.3% and 1.7%, respectively, while reducing the number of model parameters by 0.4 M and the computational complexity by 0.9 GFLOPs. The improved model has a parameter count of only 1.6 M and a computational complexity of only 3.3 GFLOPs, demonstrating a certain advantage over other mainstream object detection algorithms in terms of detection accuracy, model parameter count, and computational complexity. The experimental results show that this method is suitable for the early detection of tomato diseases and pests.
近年来，随着人工智能技术的快速发展，基于计算机视觉的害虫检测技术在农业生产中得到了广泛的应用。番茄病虫害是影响番茄产量和品质的严重问题，对其进行快速、准确的检测具有重要意义。本文提出了一种基于改进的YOLOv5n的番茄病虫害检测模型，克服了传统病虫害检测方法精度低、模型尺寸大的问题。首先，我们使用高效视觉变压器作为特征提取骨干网络，在降低模型参数和计算复杂度的同时提高检测精度，从而解决实时性差和模型部署问题。其次，我们用轻量级的通用上采样算子Content-Aware ReAssembly of FEatures取代原来的最近邻插值上采样模块，以减少上采样过程中特征信息的丢失。最后，我们用Wise-IoU代替原来的CIoU作为目标边界框的回归损失函数，提高了预测边界框的回归预测精度，同时加快了回归损失函数的收敛速度。对数据扩增条件下番茄病虫害试验结果进行统计分析。结果表明，改进后的算法将mAP50和mAP50:95分别提高了2.3%和1.7%，模型参数数量减少了0.4 M，计算复杂度降低了0.9 GFLOPs。改进后的模型参数数仅为1.6 M，计算复杂度仅为3.3 GFLOPs，在检测精度、模型参数数和计算复杂度方面都比其他主流目标检测算法有一定的优势。实验结果表明，该方法适用于番茄病虫害的早期检测。
{"title":"A Study on Tomato Disease and Pest Detection Method","authors":"Wenyi Hu, Wei Hong, Hongkun Wang, Meilin Liu, Shan Liu","doi":"10.3390/app131810063","DOIUrl":"https://doi.org/10.3390/app131810063","url":null,"abstract":"In recent years, with the rapid development of artificial intelligence technology, computer vision-based pest detection technology has been widely used in agricultural production. Tomato diseases and pests are serious problems affecting tomato yield and quality, so it is important to detect them quickly and accurately. In this paper, we propose a tomato disease and pest detection model based on an improved YOLOv5n to overcome the problems of low accuracy and large model size in traditional pest detection methods. Firstly, we use the Efficient Vision Transformer as the feature extraction backbone network to reduce model parameters and computational complexity while improving detection accuracy, thus solving the problems of poor real-time performance and model deployment. Second, we replace the original nearest neighbor interpolation upsampling module with the lightweight general-purpose upsampling operator Content-Aware ReAssembly of FEatures to reduce feature information loss during upsampling. Finally, we use Wise-IoU instead of the original CIoU as the regression loss function of the target bounding box to improve the regression prediction accuracy of the predicted bounding box while accelerating the convergence speed of the regression loss function. We perform statistical analysis on the experimental results of tomato diseases and pests under data augmentation conditions. The results show that the improved algorithm improves mAP50 and mAP50:95 by 2.3% and 1.7%, respectively, while reducing the number of model parameters by 0.4 M and the computational complexity by 0.9 GFLOPs. The improved model has a parameter count of only 1.6 M and a computational complexity of only 3.3 GFLOPs, demonstrating a certain advantage over other mainstream object detection algorithms in terms of detection accuracy, model parameter count, and computational complexity. The experimental results show that this method is suitable for the early detection of tomato diseases and pests.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43253967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Explainable artificial intelligence (XAI) methods aim to explain to the user on what basis the model makes decisions. Unfortunately, general-purpose approaches that are independent of the types of data, model used and the level of sophistication of the user are not always able to make model decisions more comprehensible. An example of such a problem, which is considered in this paper, is a predictive maintenance task where a model identifying outliers in time series is applied. Typical explanations of the model’s decisions, which present the importance of the attributes, are not sufficient to support the user for such a task. Within the framework of this work, a visualisation and analysis of the context of local explanations presenting attribute importance are proposed. Two types of context for explanations are considered: local and global. They extend the information provided by typical explanations and offer the user greater insight into the validity of the alarms triggered by the model. Evaluation of the proposed context was performed on two time series representations: basic and extended. For the extended representation, an aggregation of explanations was used to make them more intuitive for the user. The results show the usefulness of the proposed context, particularly for the basic data representation. However, for the extended representation, the aggregation of explanations used is sometimes insufficient to provide a clear explanatory context. Therefore, the explanation using simplification with a surrogate model on basic data representation was proposed as a solution. The obtained results can be valuable for developers of decision support systems for predictive maintenance.
{"title":"Contextual Explanations for Decision Support in Predictive Maintenance","authors":"Michał Kozielski","doi":"10.3390/app131810068","DOIUrl":"https://doi.org/10.3390/app131810068","url":null,"abstract":"Explainable artificial intelligence (XAI) methods aim to explain to the user on what basis the model makes decisions. Unfortunately, general-purpose approaches that are independent of the types of data, model used and the level of sophistication of the user are not always able to make model decisions more comprehensible. An example of such a problem, which is considered in this paper, is a predictive maintenance task where a model identifying outliers in time series is applied. Typical explanations of the model’s decisions, which present the importance of the attributes, are not sufficient to support the user for such a task. Within the framework of this work, a visualisation and analysis of the context of local explanations presenting attribute importance are proposed. Two types of context for explanations are considered: local and global. They extend the information provided by typical explanations and offer the user greater insight into the validity of the alarms triggered by the model. Evaluation of the proposed context was performed on two time series representations: basic and extended. For the extended representation, an aggregation of explanations was used to make them more intuitive for the user. The results show the usefulness of the proposed context, particularly for the basic data representation. However, for the extended representation, the aggregation of explanations used is sometimes insufficient to provide a clear explanatory context. Therefore, the explanation using simplification with a surrogate model on basic data representation was proposed as a solution. The obtained results can be valuable for developers of decision support systems for predictive maintenance.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43424489","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Recommender systems play a crucial role in personalizing online user experiences by creating user profiles based on user–item interactions and preferences. Knowledge graphs (KGs) are intricate data structures that encapsulate semantic information, expressing users and items in a meaningful way. Although recent deep learning-based recommendation algorithms that embed KGs have demonstrated impressive performance, the richness of semantics and explainability embedded in the KGs are often lost due to the opaque nature of vector representations in deep neural networks. To address this issue, we propose a novel user profiling method for recommender systems that can encapsulate user preferences while preserving the original semantics of the KGs, using frequent subgraph mining. Our approach involves creating user profile vectors from a set of frequent subgraphs that contain information about user preferences and the strength of those preferences, measured by frequency. Subsequently, we trained a deep neural network model to learn the relationship between users and items, thereby facilitating effective recommendations using the neural network’s approximation ability. We evaluated our user profiling methodology on movie data and found that it demonstrated competitive performance, indicating that our approach can accurately represent user preferences while maintaining the semantics of the KGs. This work, therefore, presents a significant step towards creating more transparent and effective recommender systems that can be beneficial for a wide range of applications and readers interested in this field.
{"title":"Enhancing Recommender Systems with Semantic User Profiling through Frequent Subgraph Mining on Knowledge Graphs","authors":"Haemin Jung, Heesung Park, Kwangyon Lee","doi":"10.3390/app131810041","DOIUrl":"https://doi.org/10.3390/app131810041","url":null,"abstract":"Recommender systems play a crucial role in personalizing online user experiences by creating user profiles based on user–item interactions and preferences. Knowledge graphs (KGs) are intricate data structures that encapsulate semantic information, expressing users and items in a meaningful way. Although recent deep learning-based recommendation algorithms that embed KGs have demonstrated impressive performance, the richness of semantics and explainability embedded in the KGs are often lost due to the opaque nature of vector representations in deep neural networks. To address this issue, we propose a novel user profiling method for recommender systems that can encapsulate user preferences while preserving the original semantics of the KGs, using frequent subgraph mining. Our approach involves creating user profile vectors from a set of frequent subgraphs that contain information about user preferences and the strength of those preferences, measured by frequency. Subsequently, we trained a deep neural network model to learn the relationship between users and items, thereby facilitating effective recommendations using the neural network’s approximation ability. We evaluated our user profiling methodology on movie data and found that it demonstrated competitive performance, indicating that our approach can accurately represent user preferences while maintaining the semantics of the KGs. This work, therefore, presents a significant step towards creating more transparent and effective recommender systems that can be beneficial for a wide range of applications and readers interested in this field.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45018053","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Daekook M. Nekar, DongYeop Lee, Ji-Heon Hong, JinSeop Kim, S. Kim, Yeon-Gyo Nam, Jaeho Yu
The present study investigated the feasibility and effectiveness of pseudo-weight resistance training using mixed-reality (MR) technology for shoulder muscle activation in healthy individuals. Thirty-two healthy students aged 20~35 years participated in this preliminary study and were divided into two groups. Participants in the MR group received 30 min of training three times a week for 4 weeks using a customized MR-based pseudo-weight resistance training system. Those in the control group performed the same exercises using a conventional training program. Muscle activation of the deltoids, upper trapezius, infraspinatus, and supraspinatus were measured before and after the intervention. There was a statistically significant difference in middle deltoid, upper trapezius, and supraspinatus muscle activation in the MR group (p < 0.05), while the control group showed a significant difference in the anterior and middle deltoid, upper trapezius, and supraspinatus (p < 0.05). Regarding the between-group comparison, no statistically significant difference was observed for all six muscles (p > 0.05). Without any superiority of physical weight resistance training in the pseudo-weight training program, an MR-based pseudo-weight resistance training system can potentially be used for muscle-strengthening training, especially for early rehabilitation programs. However, further study using a large sample size with a long experimental duration is needed for more evidence of the presented technology and its use in home training.
{"title":"Effects of Pseudo-Weight Resistance Training Using Mixed-Reality Technology on Muscle Activation in Healthy Adults: A Preliminary Study","authors":"Daekook M. Nekar, DongYeop Lee, Ji-Heon Hong, JinSeop Kim, S. Kim, Yeon-Gyo Nam, Jaeho Yu","doi":"10.3390/app131810021","DOIUrl":"https://doi.org/10.3390/app131810021","url":null,"abstract":"The present study investigated the feasibility and effectiveness of pseudo-weight resistance training using mixed-reality (MR) technology for shoulder muscle activation in healthy individuals. Thirty-two healthy students aged 20~35 years participated in this preliminary study and were divided into two groups. Participants in the MR group received 30 min of training three times a week for 4 weeks using a customized MR-based pseudo-weight resistance training system. Those in the control group performed the same exercises using a conventional training program. Muscle activation of the deltoids, upper trapezius, infraspinatus, and supraspinatus were measured before and after the intervention. There was a statistically significant difference in middle deltoid, upper trapezius, and supraspinatus muscle activation in the MR group (p < 0.05), while the control group showed a significant difference in the anterior and middle deltoid, upper trapezius, and supraspinatus (p < 0.05). Regarding the between-group comparison, no statistically significant difference was observed for all six muscles (p > 0.05). Without any superiority of physical weight resistance training in the pseudo-weight training program, an MR-based pseudo-weight resistance training system can potentially be used for muscle-strengthening training, especially for early rehabilitation programs. However, further study using a large sample size with a long experimental duration is needed for more evidence of the presented technology and its use in home training.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.7,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45173476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"综合性期刊","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：