A car-sharing system has been considered the most promising solution for mitigating the waste of natural resources, traffic congestion, and greenhouse gas emission in the city. However, conventional car-sharing method relies on a trusted third party, which is facing the challenge of maximizing the benefits of C2C model and privacy disclosure risk. This work aims to design a brand-new car-sharing version based on blockchain network and attribute-based cryptosystem. The demander and supplier can use the smart contract to share the vehicle without the help of a centralized node; thus, avoiding the collusion manipulation to reach the optimal profit. In particular, even if the vehicle has been damaged, the supplier can trace the responsibility by accessing the order information stored in the blockchain. Security analysis has demonstrated the confirmation of robustness and privacy essentials, while experimental outcomes have shown the satisfactory feasibility of the new method.
{"title":"Preserving Privacy and Traceability in Car-sharing Blockchain Based on Attribute Cryptosystem","authors":"Tzu-Hao Chen Tzu-Hao Chen, Chit-Jie Chew Tzu-Hao Chen, Jung-San Lee Chit-Jie Chew","doi":"10.53106/160792642024032502001","DOIUrl":"https://doi.org/10.53106/160792642024032502001","url":null,"abstract":"\u0000 A car-sharing system has been considered the most promising solution for mitigating the waste of natural resources, traffic congestion, and greenhouse gas emission in the city. However, conventional car-sharing method relies on a trusted third party, which is facing the challenge of maximizing the benefits of C2C model and privacy disclosure risk. This work aims to design a brand-new car-sharing version based on blockchain network and attribute-based cryptosystem. The demander and supplier can use the smart contract to share the vehicle without the help of a centralized node; thus, avoiding the collusion manipulation to reach the optimal profit. In particular, even if the vehicle has been damaged, the supplier can trace the responsibility by accessing the order information stored in the blockchain. Security analysis has demonstrated the confirmation of robustness and privacy essentials, while experimental outcomes have shown the satisfactory feasibility of the new method.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"63 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140404331","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 : 2024-03-01DOI: 10.53106/160792642024032502004
Xiaoning Bo Xiaoning Bo, Zhiyuan Zhang Xiaoning Bo, Yipeng Wang Zhiyuan Zhang
Aiming at the problems of slow recognition speed and low recognition accuracy of arbitrarily placed workpiece by machine vision in traditional automated production lines, a workpiece recognition algorithm based on improved SSD is proposed. Firstly, the improved DarkNet53 is used to replace the backbone network in the original SSD network framework, and the network enhancement is used in the backbone network to solve the defect of small target missed detection. Then, channel attention module and deep semantic feature fusion module are added, in order to improve the recognition ability and detection accuracy of the small target features. Lastly, the loss function was optimized, and the problem caused by sample imbalance was solved by changing the weight distribution of positive and negative samples. In the experiment, image datasets of typical bolts, nuts, and connecting plates were constructed for the network training, the experimental results showed that, the recognition accuracy and speed have been optimized and meet the requirements of automatic work-piece detection in actual production, compared with traditional YOLOv4 and the original SSD algorithm in the work-piece recognition task.
{"title":"An Improved SSD Model for Small Size Work-pieces Recognition in Automatic Production Line","authors":"Xiaoning Bo Xiaoning Bo, Zhiyuan Zhang Xiaoning Bo, Yipeng Wang Zhiyuan Zhang","doi":"10.53106/160792642024032502004","DOIUrl":"https://doi.org/10.53106/160792642024032502004","url":null,"abstract":"\u0000 Aiming at the problems of slow recognition speed and low recognition accuracy of arbitrarily placed workpiece by machine vision in traditional automated production lines, a workpiece recognition algorithm based on improved SSD is proposed. Firstly, the improved DarkNet53 is used to replace the backbone network in the original SSD network framework, and the network enhancement is used in the backbone network to solve the defect of small target missed detection. Then, channel attention module and deep semantic feature fusion module are added, in order to improve the recognition ability and detection accuracy of the small target features. Lastly, the loss function was optimized, and the problem caused by sample imbalance was solved by changing the weight distribution of positive and negative samples. In the experiment, image datasets of typical bolts, nuts, and connecting plates were constructed for the network training, the experimental results showed that, the recognition accuracy and speed have been optimized and meet the requirements of automatic work-piece detection in actual production, compared with traditional YOLOv4 and the original SSD algorithm in the work-piece recognition task.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"47 38","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140400398","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 : 2024-03-01DOI: 10.53106/160792642024032502007
Zheng Yao Ng Zheng Yao Ng, Iftekhar Salam Zheng Yao Ng
We developed a blockchain-based multi-keyword searchable encryption scheme for securing COVID-19 contact tracing data. In this scheme, we used AES-GCM to encrypt the contact tracing data, guaranteeing that only authorized users can perform decryption. Our scheme employed a Ciphertext-Policy Attribute-based searchable encryption to encrypt the search index, ensuring only users with appropriate attributes can perform search operations. The scheme supports dynamic updates of the search index. The blockchain-based storage with smart contract ensures immutability and non-repudiation of storage and retrieval. Overall, the evaluation of the scheme shows that it works efficiently without compromising the security goals. This is one of the first works to implement a solution for secure storage and search of contact tracing data with blockchain-based searchable encryption. Compared to the existing searchable contact tracing schemes, it provides more features and maintains efficiency even if a large search index is used.
{"title":"Implementation of A Blockchain-based Searchable Encryption for Securing Contact Tracing Data","authors":"Zheng Yao Ng Zheng Yao Ng, Iftekhar Salam Zheng Yao Ng","doi":"10.53106/160792642024032502007","DOIUrl":"https://doi.org/10.53106/160792642024032502007","url":null,"abstract":"\u0000 We developed a blockchain-based multi-keyword searchable encryption scheme for securing COVID-19 contact tracing data. In this scheme, we used AES-GCM to encrypt the contact tracing data, guaranteeing that only authorized users can perform decryption. Our scheme employed a Ciphertext-Policy Attribute-based searchable encryption to encrypt the search index, ensuring only users with appropriate attributes can perform search operations. The scheme supports dynamic updates of the search index. The blockchain-based storage with smart contract ensures immutability and non-repudiation of storage and retrieval. Overall, the evaluation of the scheme shows that it works efficiently without compromising the security goals. This is one of the first works to implement a solution for secure storage and search of contact tracing data with blockchain-based searchable encryption. Compared to the existing searchable contact tracing schemes, it provides more features and maintains efficiency even if a large search index is used.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"221 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140402793","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 : 2024-03-01DOI: 10.53106/160792642024032502006
Dun Cao Dun Cao, Jiawen Liu Dun Cao, Qinghua Liu Jiawen Liu, Jin Wang Qinghua Liu, Min Zhu Jin Wang
In recent years, Internet of Vehicles (IoV), as a supporting technology for Intelligent Transportation System (ITS), is flourishing with the emergence and development of new technologies such as edge computing, 5G communication, and Artificial Intelligence (AI). However, the more complexity of wireless channels and vehicle distribution in 3D scenario brings a great challenge for relay-node selection in ITS. In this paper, we focus on how to alleviate the problem that the decline of two-hop distance and two-hop connection probability caused by the relative fading of inter-layer communication radius in 3D scenario with Vehicle-to-Vehicle (V2V) communication. To face this challenge, we develop a Relay-node Selection method based on Weighted Strategy for Overpass scenario (RSWSO). The simulation results show that RSWSO achieves improvement in two-hop distance, and an increase of up to 11.2% in terms of two-hop connection probability.
{"title":"Relay-node Selection Method Based on Weighted Strategy for 3D Scenario in Internet of Vehicles","authors":"Dun Cao Dun Cao, Jiawen Liu Dun Cao, Qinghua Liu Jiawen Liu, Jin Wang Qinghua Liu, Min Zhu Jin Wang","doi":"10.53106/160792642024032502006","DOIUrl":"https://doi.org/10.53106/160792642024032502006","url":null,"abstract":"\u0000 In recent years, Internet of Vehicles (IoV), as a supporting technology for Intelligent Transportation System (ITS), is flourishing with the emergence and development of new technologies such as edge computing, 5G communication, and Artificial Intelligence (AI). However, the more complexity of wireless channels and vehicle distribution in 3D scenario brings a great challenge for relay-node selection in ITS. In this paper, we focus on how to alleviate the problem that the decline of two-hop distance and two-hop connection probability caused by the relative fading of inter-layer communication radius in 3D scenario with Vehicle-to-Vehicle (V2V) communication. To face this challenge, we develop a Relay-node Selection method based on Weighted Strategy for Overpass scenario (RSWSO). The simulation results show that RSWSO achieves improvement in two-hop distance, and an increase of up to 11.2% in terms of two-hop connection probability.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"28 109","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140406217","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}
The rise of connected and autonomous vehicles has resulted in the development of vehicular ad hoc networks (VANETs) as a means of improving road safety, traffic efficiency, and passenger comfort. However, VANETs present challenges to achieving reliability and service quality in networks suffered highly by broadcast storms due to their mobility, scalability, and high node density. By addressing such challenges, this paper seeks to contribute to solving the problem of efficient information dissemination by diluting the broadcast storm and coverage issues by proposing a novel relay selection scheme based on the validation of the far-to-farthest zone technique. A flag control suppression mechanism is presented to overcome repeated information problems. RDVFF includes a separate algorithm for the optimization of participants for relay inclusion to address the latency issue during communication based on a multi-directional multiple-selection scheme. Compared with state-of-art protocols in terms of throughput, delivery ratio, collision, and latency experimental results show improvement of an average of 21.27%, 18.34%, 36.76%, and 38.04% respectively.
{"title":"RDVFF- Reliable Data Dissemination in Vehicular Ad Hoc Networks Based on Validation of Far to Farthest Zone","authors":"Deepak Gupta Deepak Gupta, Rakesh Rathi Deepak Gupta","doi":"10.53106/160792642024012501008","DOIUrl":"https://doi.org/10.53106/160792642024012501008","url":null,"abstract":"\u0000 The rise of connected and autonomous vehicles has resulted in the development of vehicular ad hoc networks (VANETs) as a means of improving road safety, traffic efficiency, and passenger comfort. However, VANETs present challenges to achieving reliability and service quality in networks suffered highly by broadcast storms due to their mobility, scalability, and high node density. By addressing such challenges, this paper seeks to contribute to solving the problem of efficient information dissemination by diluting the broadcast storm and coverage issues by proposing a novel relay selection scheme based on the validation of the far-to-farthest zone technique. A flag control suppression mechanism is presented to overcome repeated information problems. RDVFF includes a separate algorithm for the optimization of participants for relay inclusion to address the latency issue during communication based on a multi-directional multiple-selection scheme. Compared with state-of-art protocols in terms of throughput, delivery ratio, collision, and latency experimental results show improvement of an average of 21.27%, 18.34%, 36.76%, and 38.04% respectively.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"8 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139637401","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 : 2024-01-01DOI: 10.53106/160792642024012501004
Po-Yuan Su Po-Yuan Su, Chung-Ho Su Po-Yuan Su, Chao-Ming Wang Chung-Ho Su, Kuo-Kuang Fan Chao-Ming Wang
In recent years, the rapid expansion of digital media has revolutionized educational materials by incorporating engaging multimedia elements and gamification principles. These advancements in technology have transformed digital learning into a gamified experience, enhancing the creation and utilization of learning materials. The primary aim of this study was to develop a predictive model for measuring learning outcomes in the Air Quality Education (AQE) app. The study results revealed that the most influential factor affecting learning outcomes was Learning Outcome (R2 = 0.673), followed by Behavioral Intention (R2 = 0.541), Practicability (R2 = 0.456), Use Attitude (R2 = 0.431), and Entertainment (R2 = 0.328). The proposed model accounted for an impressive 67.29.8% of the variability in predicting learning outcomes. These research findings emphasize the significance of Behavioral Intention and Practicability as crucial factors in designing predictive models for learning outcomes within the AQE app. Furthermore, the study highlights the users’ strong emphasis on interface media richness and user experience design elements, indicating the need for their prioritized attention and enhancement. By focusing on these aspects, developers can heighten players’ expectations for interactive media elements and enable them to achieve a more immersive gaming experience through a user-friendly and interactive interface design.
{"title":"Integrating Media Richness Theory and Technology Acceptance Model to Study the Learning Outcomes of Air Quality Education App","authors":"Po-Yuan Su Po-Yuan Su, Chung-Ho Su Po-Yuan Su, Chao-Ming Wang Chung-Ho Su, Kuo-Kuang Fan Chao-Ming Wang","doi":"10.53106/160792642024012501004","DOIUrl":"https://doi.org/10.53106/160792642024012501004","url":null,"abstract":"\u0000 In recent years, the rapid expansion of digital media has revolutionized educational materials by incorporating engaging multimedia elements and gamification principles. These advancements in technology have transformed digital learning into a gamified experience, enhancing the creation and utilization of learning materials. The primary aim of this study was to develop a predictive model for measuring learning outcomes in the Air Quality Education (AQE) app. The study results revealed that the most influential factor affecting learning outcomes was Learning Outcome (R2 = 0.673), followed by Behavioral Intention (R2 = 0.541), Practicability (R2 = 0.456), Use Attitude (R2 = 0.431), and Entertainment (R2 = 0.328). The proposed model accounted for an impressive 67.29.8% of the variability in predicting learning outcomes. These research findings emphasize the significance of Behavioral Intention and Practicability as crucial factors in designing predictive models for learning outcomes within the AQE app. Furthermore, the study highlights the users’ strong emphasis on interface media richness and user experience design elements, indicating the need for their prioritized attention and enhancement. By focusing on these aspects, developers can heighten players’ expectations for interactive media elements and enable them to achieve a more immersive gaming experience through a user-friendly and interactive interface design.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"335 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139632236","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}
With the characteristics on recording the trace of human life, cultural heritage is acknowledged as possessing significant humanistic, spiritual, and socio-economic values. However, there remain problems such as insufficient integration with science and technology, inadequate design innovation, and the lack of digital construction. Based on these problems, the article discusses the need of constructing a digital management platform with cultural heritage knowledge, the expression paradigm, the essential contents, and key technologies. By elaborating current status of its application and utilization, the study explores issues including scattered resources, simple transplantation and lacking digital approach. And it found that designers’ cognition on cultural heritage is relatively superficial, especially for images and literal meanings. Aiming for sharing design knowledge of cultural heritage, the study proposes a conceptual model digital platform of management based on the exploration of design process in extracting cultural elements and verifying the feasibility where design needs can be achievable. Taking Macao’s historic area as an example, the study demonstrates a way to construct the digital resource sharing and management platform using “open source” approach as a sort of design alternative where cultural elements and compositions can be applied efficiently.
{"title":"A Study on Constructing Digital Resource Sharing and Management Platform by Applying Cultural Compositions of the Historical Urban Area","authors":"Kuo-Hsun Wen Kuo-Hsun Wen, Xinyi Chen Kuo-Hsun Wen, Cheng-Cheong Lei Xinyi Chen","doi":"10.53106/160792642024012501001","DOIUrl":"https://doi.org/10.53106/160792642024012501001","url":null,"abstract":"\u0000 With the characteristics on recording the trace of human life, cultural heritage is acknowledged as possessing significant humanistic, spiritual, and socio-economic values. However, there remain problems such as insufficient integration with science and technology, inadequate design innovation, and the lack of digital construction. Based on these problems, the article discusses the need of constructing a digital management platform with cultural heritage knowledge, the expression paradigm, the essential contents, and key technologies. By elaborating current status of its application and utilization, the study explores issues including scattered resources, simple transplantation and lacking digital approach. And it found that designers’ cognition on cultural heritage is relatively superficial, especially for images and literal meanings. Aiming for sharing design knowledge of cultural heritage, the study proposes a conceptual model digital platform of management based on the exploration of design process in extracting cultural elements and verifying the feasibility where design needs can be achievable. Taking Macao’s historic area as an example, the study demonstrates a way to construct the digital resource sharing and management platform using “open source” approach as a sort of design alternative where cultural elements and compositions can be applied efficiently.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"56 44","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139634552","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}
With the widespread application of deep neural networks in image detection, adversarial sample attacks have gradually become a hot issue of concern for researchers. In this paper we propose a new adversarial sample detection approach called AdvDetector, which combines image generation through label fusion with image similarity detection. AdvDetector enhances sample quality and effectively identifies adversarial samples. Specifically, the method generates images by selecting seed pixels, the labels of deep neural network classification, and the pixel distribution learned from training data, and detects them using image similarity comparison methods. During the sample generation process, we introduce the AdvDetector method for adversarial sample detection to improve the quality of generated samples. We evaluated the effectiveness of the method on three publicly available image datasets, MNIST, Cifar-10, and GTSR, and the results show that the method is superior to existing baseline methods in terms of adversarial sample detection rate and sample generation quality.
{"title":"Research on Adversarial Sample Detection Method Based on Image Similarity","authors":"Xiaoxue Wu Xiaoxue Wu, Shuqi Zuo Xiaoxue Wu, Shiyu Weng Shuqi Zuo, Yongkang Jiang Shiyu Weng, Hao Huang Yongkang Jiang","doi":"10.53106/160792642024012501012","DOIUrl":"https://doi.org/10.53106/160792642024012501012","url":null,"abstract":"\u0000 With the widespread application of deep neural networks in image detection, adversarial sample attacks have gradually become a hot issue of concern for researchers. In this paper we propose a new adversarial sample detection approach called AdvDetector, which combines image generation through label fusion with image similarity detection. AdvDetector enhances sample quality and effectively identifies adversarial samples. Specifically, the method generates images by selecting seed pixels, the labels of deep neural network classification, and the pixel distribution learned from training data, and detects them using image similarity comparison methods. During the sample generation process, we introduce the AdvDetector method for adversarial sample detection to improve the quality of generated samples. We evaluated the effectiveness of the method on three publicly available image datasets, MNIST, Cifar-10, and GTSR, and the results show that the method is superior to existing baseline methods in terms of adversarial sample detection rate and sample generation quality.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"303 11","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139636024","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}
With advances in communication technology, modern society relies more than ever on the Internet and various user-friendly digital tools. It provides access to and enables the manipulation of files, trips, and the Windows API. Attackers frequently use various obfuscation techniques PowerShell scripts to avoid detection by anti-virus software. Their doing so can significantly reduce the readability of the script. This work statically analyzes PowerShell scripts. Thirty-three features that were based on the script’s keywords, format, and string combinations were used herein to determine the behavioral intent of the script. Ones are characteristic-based features that are obtained by calculation; the others are behavior-based features that determine the execution function of behavior using keywords and instructions. Behavior-based features can be divided into positive behavior-based features, neutral behavior-based features, and negative behavior-based features. These three types of features are enhanced by observing samples and adding keywords. The other type of characteristic-based feature is introduced into the formula from other studies in this work. The XGBoost model was used to evaluate the importance of the features that are proposed in this study and to identify the combination of features that contributed most to the detection of PowerShell scripts. The final model with the combined features is found to exhibit the best performance. The model has 99.27% accuracy when applied to the validation dataset. The results clearly indicate that the proposed malicious PowerShell script detection model outperforms previously developed models.
随着通信技术的进步，现代社会比以往任何时候都更加依赖互联网和各种用户友好型数字工具。它提供了对文件、行程和 Windows API 的访问和操作。攻击者经常使用各种混淆技术 PowerShell 脚本来避免被反病毒软件检测到。他们这样做会大大降低脚本的可读性。本研究对 PowerShell 脚本进行了静态分析。这里使用了基于脚本关键字、格式和字符串组合的 33 个特征来确定脚本的行为意图。其中一个是基于特征的特征，通过计算获得；其他是基于行为的特征，通过关键字和指令确定行为的执行功能。基于行为的特征可分为积极行为特征、中性行为特征和消极行为特征。这三类特征都是通过观察样本和添加关键字来增强的。另一种基于特征的特征是在本研究中从其他研究中引入到公式中的。XGBoost 模型用于评估本研究中提出的特征的重要性，并确定对检测 PowerShell 脚本贡献最大的特征组合。最终发现，具有组合特征的模型表现最佳。该模型应用于验证数据集时的准确率为 99.27%。结果清楚地表明，所提出的恶意 PowerShell 脚本检测模型优于之前开发的模型。
{"title":"Machine Learning Approaches to Malicious PowerShell Scripts Detection and Feature Combination Analysis","authors":"Hsiang-Hua Hung Hsiang-Hua Hung, Jiann-Liang Chen Hsiang-Hua Hung, Yi-Wei Ma Jiann-Liang Chen","doi":"10.53106/160792642024012501014","DOIUrl":"https://doi.org/10.53106/160792642024012501014","url":null,"abstract":"\u0000 With advances in communication technology, modern society relies more than ever on the Internet and various user-friendly digital tools. It provides access to and enables the manipulation of files, trips, and the Windows API. Attackers frequently use various obfuscation techniques PowerShell scripts to avoid detection by anti-virus software. Their doing so can significantly reduce the readability of the script. This work statically analyzes PowerShell scripts. Thirty-three features that were based on the script’s keywords, format, and string combinations were used herein to determine the behavioral intent of the script. Ones are characteristic-based features that are obtained by calculation; the others are behavior-based features that determine the execution function of behavior using keywords and instructions. Behavior-based features can be divided into positive behavior-based features, neutral behavior-based features, and negative behavior-based features. These three types of features are enhanced by observing samples and adding keywords. The other type of characteristic-based feature is introduced into the formula from other studies in this work. The XGBoost model was used to evaluate the importance of the features that are proposed in this study and to identify the combination of features that contributed most to the detection of PowerShell scripts. The final model with the combined features is found to exhibit the best performance. The model has 99.27% accuracy when applied to the validation dataset. The results clearly indicate that the proposed malicious PowerShell script detection model outperforms previously developed models.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"8 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139637364","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 : 2024-01-01DOI: 10.53106/160792642024012501011
Juan J. López-Jiménez Juan J. López-Jiménez, José A. García-Berná Juan J. López-Jiménez, Sofia Ouhbi José A. García-Berná, Begoña Moros Valle Sofia Ouhbi, José L. Fernández-Alemán Begoña Moros Valle
Audience Response Systems (ARS) can be used to increase students’ commitment and engagement. ARS are becoming popular at lectures, complementing traditional masterclasses and shedding light to a more profitability of the time. Several researchers studied the impact of ARS in the classroom. However, there is a lack of information about the current research landscape to identify paths towards the development of scientific research and projects in ARS field. This bibliometric study discusses a collection of bibliometric parameters on ARS literature that were calculated from data downloaded in Scopus database. A total of 2,015 publications were considered from Scopus database. Results showed that the number of publications is stable since 2010 with a noticeable decrease in 2019. The United States and the United Kingdom are the most productive countries with a total of 898 papers in the US and 179 in the UK. The most prolific author was Daniel Zingaro from the University of Toronto with a total of 10 manuscripts published. This study provides researchers who are interested in conducting research on ARS with insights on potential venues for publications and collaboration with research institutions and researchers that are more prolific in the field.
观众反应系统（ARS）可用于提高学生的投入度和参与度。受众反应系统在讲座中正变得越来越流行，它是对传统主讲课程的补充，并为更多的时间带来更多的利润。一些研究人员对 ARS 在课堂上的影响进行了研究。然而，目前缺乏有关研究现状的信息，因此无法确定 ARS 领域科学研究和项目的发展路径。本文献计量学研究讨论了从 Scopus 数据库下载的数据中计算得出的 ARS 文献的文献计量参数集。Scopus 数据库共收录了 2,015 篇出版物。结果显示，自 2010 年以来，出版物数量保持稳定，但在 2019 年出现明显下降。美国和英国是发表论文最多的国家，美国共发表了898篇论文，英国共发表了179篇论文。最多产的作者是多伦多大学的丹尼尔-津加罗（Daniel Zingaro），共发表了10篇手稿。这项研究为有兴趣开展 ARS 研究的研究人员提供了发表论文的潜在渠道，以及与该领域较多产的研究机构和研究人员合作的机会。
{"title":"A Quantitative Characterization of Audience Response System Research","authors":"Juan J. López-Jiménez Juan J. López-Jiménez, José A. García-Berná Juan J. López-Jiménez, Sofia Ouhbi José A. García-Berná, Begoña Moros Valle Sofia Ouhbi, José L. Fernández-Alemán Begoña Moros Valle","doi":"10.53106/160792642024012501011","DOIUrl":"https://doi.org/10.53106/160792642024012501011","url":null,"abstract":"\u0000 Audience Response Systems (ARS) can be used to increase students’ commitment and engagement. ARS are becoming popular at lectures, complementing traditional masterclasses and shedding light to a more profitability of the time. Several researchers studied the impact of ARS in the classroom. However, there is a lack of information about the current research landscape to identify paths towards the development of scientific research and projects in ARS field. This bibliometric study discusses a collection of bibliometric parameters on ARS literature that were calculated from data downloaded in Scopus database. A total of 2,015 publications were considered from Scopus database. Results showed that the number of publications is stable since 2010 with a noticeable decrease in 2019. The United States and the United Kingdom are the most productive countries with a total of 898 papers in the US and 179 in the UK. The most prolific author was Daniel Zingaro from the University of Toronto with a total of 10 manuscripts published. This study provides researchers who are interested in conducting research on ARS with insights on potential venues for publications and collaboration with research institutions and researchers that are more prolific in the field.\u0000 \u0000","PeriodicalId":442331,"journal":{"name":"網際網路技術學刊","volume":"54 22","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139634434","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：