检索结果-内蒙古大学图书馆

2022 IEEE SmartWorld, 19th IEEE International Conference on Ubiquitous Intelligence and Computing, 2022 IEEE International Conference on Autonomous and Trusted Vehicles Conference, 22nd IEEE International Conference on Scalable Computing and Communications, 2022 IEEE International Conference on Digital Twin, 8th IEEE International Conference on Privacy Computing and 2022 IEEE International Conference on Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022

作者： He, Kun Liu, Qiyan Li, Qianmu Meng, Shunmei Nanjing University of Science and Technology Nanjing China University of Toronto Computer Engineering Canada Nanjing University State Key Lab. for Novel Software Technology Nanjing China

ISBN: (纸本)9798350346558

Sequential recommendation aims to learn the changes of users' interests according to their historical behaviors and predict the most likely next item. Since user's historical behavior are sequential actions, user's interest in different time periods has different emphases. For predicting a user's next item, not only the recent behavior is important, but also long-term preference with all historical behaviors could not be ignored. Recently, self-attention based user preference modeling has drawn much attention for its advantages of fewer parameters and parallelism. However, most of the existing self-attention based model do not make a good distinction between the long-term and short-term preferences of users. Based on the observations, we design a sequential recommendation model based on a combination of long-term and short-term preference. On the one hand, considering the changes of users' interests in different periods, we divide the sequence of users' behaviors into different temporal windows. Then, we use GRU to capture users' interests in different temporal window. On the other hand, to better combine the global and local information of user's, we adopt a locally constrained multi-head attention mechanism based on Transformer encoder. On three real-world public datasets, we finally validate the efficacy of our method. © 2022 IEEE.

关键词： User profile

来源：评论

学校读者我要写书评

暂无评论

Semi-Asynchronous Hierarchical Federated Deep Learning for Intrusion Detection in Ami Networks

SSRN

引用

SSRN 2023年

作者： Xia, Zhuoqun Zhou, Hongmei Hu, Zhenzhen Jiang, Qisheng Zhou, Kaixin School of Computer and Communication Engineering Changsha University of Science and Technology Changsha410114 China College of Computer Science and Software University of South China Hengyang421001 China

The emergence of smart grid brings great convenience to users and power companies, but also brings many new problems, among which the most prominent one is network attack security. Although federated learning works well in dealing with smart grid network attacks, it suffers from gradient leakage, client node failure and a single type of training model. Therefore, this paper proposes a new federated deep ensemble learning approach for advanced metering infrastructure (AMI) intrusion detection to protect data privacy and security. First, we design a hierarchical federated learning framework based on chained secure multi-party computation, which allows multiple concentrators to collaboratively train hierarchical intrusion detection models to protect local gradients. Second, we adapt the framework to the AMI network structure characteristics and intrusion detection real-time properties and design a semi-asynchronous model distribution protocol. Finally, we build an integrated model based on temporal convolutional networks and gated recursive units (TCN-GRU) to accurately detect AMI network attacks. The experimental results show that the proposed method can achieve 99.23% detection accuracy and 40% faster convergence than existing methods. © 2023, The Authors. All rights reserved.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Poster: Improving Spectrum Based Fault Localization For Python Programs Using Weighted Code Elements

Poster: Improving Spectrum Based Fault Localization For Pyth...

引用

International Conference on software Testing, Verification, and Validation, ICST

作者： Qusay Idrees Sarhan Árpád Beszédes Department of Software Engineering University of Szeged Szeged Hungary Department of Computer Science University of Duhok Duhok Iraq

In this paper, we present an approach for improving Spectrum-Based Fault Localization (SBFL) by integrating static and dynamic information about code elements. This is achieved by giving more importance to code elements that include mathematical operators compared to other types of elements (e.g., declaration, selection, iteration, or function call) and appear in failed tests. The intuition is that these elements are more likely to have bugs than others. The proposed approach is applicable to any SBFL formula without requiring any modifications to their structures because the weighting is done on the ranking list and not on the formulas. The experimental results of a preliminary study show that our approach achieved a much better performance in terms of average ranking compared to the underlying SBFL formulas. It also improved the Top-N categories; it doubled the number of cases in which the faulty method became the top-ranked element, and in all cases the fault became part of Top-5 of the ranking list.

关键词： Location awareness software testing Codes computer bugs Python

来源：评论

学校读者我要写书评

暂无评论

Partial Permutations Comparison, Maintenance and Applications

SSRN

引用

SSRN 2024年

作者： Levy, Avivit Shalom, Braha Riva Porat, Ely aDepartment of Software Engineering Shenkar College Ramat Gan 52526 Israel bDepartment of Computer Science Bar-Ilan University Ramat Gan 5290002 Israel cDepartment of Software Engineering Shenkar College Ramat Gan 52526 Israel

This paper studies partial permutations and their use in algorithmic tasks. A partial permutation over Σ is a bijection πpar: Σ1 -> Σ2 mapping a subset Σ1 of Σ to a subset Σ2 of Σ, where |Σ1| = |Σ2| (|Σ| denotes the size of a set Σ). Intuitively, two partial permutations agree if their mapping pairs do not form conflicts. We formally define this notion enabling a consistent as well as informatively rich comparison between partial permutations. We define the Partial Permutations Agreement problem (PPA), as follows. Given two sets A1, A2 of partial permutations over alphabet Σ, each of size n, output a pair (πi,πj), where πi is in A_1, πj is in A2 and πi agrees with πj, if exists. We study the existence of a data structure for efficiently maintaining a dynamic set of partial permutations enabling to retrieve agreement of partial permutations giving both negative and positive results. As applications we point out: (1) fruitful/futile methods for efficient genes sequences comparison in database, (2) an automatic color transformation data augmentation technique for image processing through neural networks, (3) negatively answer a recently posed open question on the strict parameterized dictionary matching with one gap (PDMOG) problem over general dictionary alphabets. © 2024, The Authors. All rights reserved.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Coordinated PI-based frequency deviation control of isolated hybrid microgrid: An online multi-agent tuning approach via reinforcement learning

Coordinated PI-based frequency deviation control of isolated...

引用

2022 IEEE PES Innovative Smart Grid Technologies Conference Europe, ISGT-Europe 2022

作者： Nosrati, K. Tepljakov, A. Petlenkov, E. Levron, Y. Skiparev, V. Belikov, J. Tallinn University of Technology Department of Computer Systems Tallinn Estonia Technion Faculty of Electrical Engineering Haifa Israel Tallinn University of Technology Department of Software Science Tallinn Estonia

ISBN: (数字)9781665480321

ISBN: (纸本)9781665480321

Numerous remote area applications welcome standalone renewable energy power generation systems or isolated microgrids (MGs). Due to the nature of solar and wind energy, the frequency deviation control (FDC) in hybrid MGs has become more complicated and critical than the conventional grid for power quality purposes. By using a coordination control strategy between a double-layered capacitor and a fuel cell, our mission here is to design a FDC system based on the PI controller which is tuned by an artificial neural network (ANN) in a multi-agent structure. To achieve this aim, a reinforcement learning technique is applied to train the ANN-based tuners. The performance of the proposed FDC system has been verified under different conditions by using real data to demonstrate the stability and robustness of the proposed controller. © 2022 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Asymptotic behavior of least squares estimators for nonlinear autoregressive models

引用

science China(Information sciences) 2023年第6期66卷 299-300页

作者： Zhaobo LIU Chanying LI College of Computer Science and Software Engineering Shenzhen University Key Laboratory of Systems and Control Academy of Mathematics and Systems ScienceChinese Academy of Sciences School of Mathematical Sciences University of Chinese Academy of Sciences

Dear editor,Estimations of nonlinear autoregressive(AR) models in the literature typically involve ergodic series. Based on this assumption,the asymptotic theory has been established accordingly(see [1–3]). However,t...

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mobility and Content Retrieval in Vehicular Named Data Network: Challenges and Countermeasures

Mobility and Content Retrieval in Vehicular Named Data Netwo...

引用

International Conference on Advancements in Computational sciences (ICACS)

作者： Muhammad Abdullah Ayesha Kiran Ubaid Azam Department of Computer Science & IT The University of Lahore Lahore Pakistan Department of Software Engineering The University of Lahore Lahore Pakistan

In the ecosystem of connected vehicles, the TCP/IP stack serves a significant role in terms of content dissemination, traffic control, and assignment of vehicle address. Recently, the Internet of Vehicles (IoV) has emerged drastically. Primarily, information sharing between vehicles is carried out for road safety, location sharing, hazard warning, and infotainment services. The traditional TCP/IP model is not appropriate for the transmission of bulky data in this dense and highly dynamic environment. Recently, a Vehicular Named Data Networking (VNDN) approach has been used for efficient information sharing between vehicles. Unlike currently used TCP/IP internet architecture, vehicles demand data in the form of Interest packet and disseminate requested data in a pull-based fashion. Interest packets need to be broadcasted to the potential data producers. Although, with the advent of VNDN, many challenges of IoV have been resolved; however, mobility and content retrieval are still major concerns that may lead to the problem of packet broadcast storm. In this paper, we discussed the major contributions that have been done to eliminate vehicle mobility and content retrieval issues in the VNDN paradigm. Besides, we mention the limitations of the proposed existing solutions implemented in the VNDN scenario. Furthermore, based on existing proposed solutions, we highlight the new challenges and directions for the design of new solutions.

关键词： Measurement Storms Information sharing Transportation TCPIP Traffic control Road safety

来源：评论

学校读者我要写书评

暂无评论

Leveraging AI and NLP in Chatbot Development: An Experimental Study 20

Leveraging AI and NLP in Chatbot Development: An Experimenta...

引用

20th International Conference on Frontiers of Information Technology, FIT 2023

作者： Paracha, Abdul Wahab Arshad, Usama Ali, Raja Hashim Ul Abideen, Zain Shah, Muhammad Huzaifa Ali Khan, Talha Ijaz, Ali Zeeshan Ali, Nisar Siddique, Abu Bakar Gik Institute of Engg. Sciences & Technology Faculty of Computer Science & Engineering Topi Pakistan BCMaterials Basque Center for Materials Applications & Nanostructure Basque Country Leioa Spain University of Europe for Applied Sciences Dept. of Technology & Software Engineering Berlin Germany University of Regina Faculty of Electronic Systems Engineering Regina Canada

ISBN: (纸本)9798350395785

In the current era of chatbots, this research delves into the advancements in AI chatbots, drawing on artificial intelligence (AI) and natural language processing (NLP) techniques to mimic human-like conversations. A particular focus is given to the potential of chatbots in facilitating multitasking dialogues, offering emotional support, and addressing complex subject matter, all the while respecting user privacy and trust. The implemented chatbot model is trained on a neural network, using Keras and TensorFlow libraries. This model's performance indicates a considerable dependence on the dataset's size, with larger datasets leading to better outcomes by providing more extensive language usage and context examples. Additionally, we also analyze the effect of varying architectures and hyperparameters on chatbot performance. The significance of localizing chatbots to adapt to different languages and cultures is also highlighted. While promising, the study identifies areas of improvement, suggesting future research directions in enhancing language capture techniques, expanding training datasets, and integrating emotional intelligence within chatbot systems. © 2023 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Foreign Object Recognition Model on Coal Mine Belt Conveyors Based on Improved MobileNetV3

Foreign Object Recognition Model on Coal Mine Belt Conveyors...

引用

IEEE International Conference on Communication and Information Systems (ICCIS)

作者： Yujia Hu Zhiwei Zhang Department of Artificial Intelligence College of Computer Science Sichuan University Chengdu China Department of Software Engineering College of Computer Science Xi'an Aeronautical Institute Xi'an China

Coal mine belt conveyors are commonly used equipment in coal mining enterprises which transport coal from the coal faces using belts. Currently, video image monitoring serves as the predominant approach to identify foreign objects within the transported coal, such as large gangue and rock bolts. In recent years, the utilization of Convolutional Neural Networks has become prevalent in automatically recognizing foreign objects within images. Nevertheless, existing models suffer from issues including low recognition accuracy and slow speed. In response to the aforementioned problems, this paper proposes an improved coal foreign object recognition model based on MobileNetV3. Firstly, the Coordinate Attention mechanism is introduced into MobileNetV3 to fully consider both channel and spatial features. Secondly, the ReLU6 activation function in MobileNetV3 is replaced with the ELU activation function to augment the model's capacity to express information within the negative value range. Finally, superfluous network layers are removed to reduce computational complexity and improve recognition speed. Experimental results demonstrate that the proposed improved MobileNetV3 model achieves an accuracy of 90.8% in recognizing foreign object images, which is 2.5 to 4.0 percentage points higher compared to commonly used neural networks including AlexNet and EfficientNet. Moreover, compared with MobileNetV3, the proposed model reduces the number of parameters and computational complexity by 40.3% and 12.0% respectively. These results indicate that the proposed model reaches higher recognition accuracy while reducing network size and computational load. As a result, it provides a feasible approach for enhancing safety in coal mining enterprises.

关键词：

来源：评论

学校读者我要写书评

暂无评论

PAIR: protein-aptamer interaction prediction based on language models and contrastive learning framework

PAIR: protein-aptamer interaction prediction based on langua...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Jun Zhang Zhiqiang Yan Hao Zeng Zexuan Zhu National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Aptamers are single-stranded DNA or RNA oligonucleotides that selectively bind to specific targets, making them valuable for drug design and diagnostic applications. Identifying the interactions between aptamers and target proteins is crucial for these applications. The systematic evolution of ligands by exponential enrichment process, traditionally used for this purpose, is challenging and time-consuming. The resulting aptamers often suffer from limitations in stability and diversity. Computational approaches have shown promise in aiding the discovery of high-performance aptamers, but existing methods are usually constrained by insufficient training data and limited generalizability. Recently, advancements in pre-training large language models have offered a new avenue to mitigate the dependency on large datasets. In this study, we propose a novel method to predict aptamer-protein interactions using large language models within a contrastive learning framework. Experimental results demonstrate that our method exhibits superior generalization and outperforms existing approaches. This method holds promise as a powerful tool for predicting aptamer-protein interactions.

关键词： Proteins Drugs Protein engineering Biological system modeling Large language models Computational modeling Training data Contrastive learning Predictive models Few shot learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：