检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zheng, Ziye Wu, Jiajing Lin, Dan Li, Quanzhong Ruan, Na School of Computer Science and Engineering Sun Yat-sen University Guangzhou China School of Software Engineering Sun Yat-sen University Zhuhai China School of Computer Science and Engineering Shanghai Jiaotong University Shanghai China

As the number of blockchain platforms continues to grow, the independence of these networks poses challenges for transferring assets and information across chains. Cross-chain bridge technology has emerged to address this issue, establishing communication protocols to facilitate cross-chain interaction of assets and information, thereby enhancing user experience. However, the complexity of cross-chain transactions increases the difficulty of security regulation, rendering traditional single-chain detection methods inadequate for cross-chain scenarios. Therefore, understanding cross-chain transaction semantics is crucial, as it forms the foundation for cross-chain security detection tasks. Although there are existing methods for extracting transaction semantics specifically for single chains, these approaches often overlook the unique characteristics of cross-chain scenarios, limiting their applicability. This paper introduces XSema, a novel cross-chain semantic extraction framework grounded in asset transfer and message-passing, designed specifically for cross-chain contexts. Experimental results demonstrate that XSema effectively distinguishes between cross-chain and non-cross-chain transactions, surpassing existing methods by over 9% for the generality metric and over 10% for the generalization metric. Furthermore, we analyze the underlying asset transfer patterns and message-passing event logs associated with cross-chain transactions. We offer new insights into the coexistence of multiple blockchains and the cross-chain ecosystem. Copyright © 2024, The Authors. All rights reserved.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

A Document-level Event Extraction Method Based on Mogrifier LSTM 4

A Document-level Event Extraction Method Based on Mogrifier ...

引用

4th International Conference on Robotics, Intelligent Control and Artificial Intelligence, RICAI 2022

作者： Ge, Junwei Qin, Zhixiang Fang, Yiqiu School of Software Engineering Chongqing University of Posts and Telecommunications Chongqing China School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (纸本)9781450398343

Element dispersion is a difficulty for document-level event extraction. While classic LSTM lacks the capability to interact between input and context while collecting long sequence features, previous document-level event extraction utilizes the entire document as input, leaving the sequence features of the document devoid of deeper contextual information. This paper suggests a document-level event extraction strategy based on Mogrifier LSTM to solve this issue. We divide the text into multiple paragraphs and then input each paragraph individually into the Mogrifier LSTM. To increase the context modeling capability of lengthy sequence text, the upgraded LSTM will allow the input of the present moment and the output of the preceding moment to be computed several times initially. Then an attention mechanism is introduced to capture the internal correlation of each paragraph and integrate the contextual semantics of each paragraph. Finally, sequence annotation is used to extract dispersed event elements and match event types. According to the experimental results on Chinese financial dataset, the method in this paper can effectively solve the problems of loss of depth information and scattered theoretical elements of long sequence features of documents, and improve the effectiveness of document-level event extraction. © 2022 ACM.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Ethics of AI: A Systematic Literature Review of Principles and Challenges 22

Ethics of AI: A Systematic Literature Review of Principles a...

引用

26th ACM International Conference on Evaluation and Assessment in software engineering, EASE 2022

作者： Khan, Arif Ali Badshah, Sher Liang, Peng Waseem, Muhammad Khan, Bilal Ahmad, Aakash Fahmideh, Mahdi Niazi, Mahmood Akbar, Muhammad Azeem M3S Empirical Software Engineering Research Unit University of Oulu Finland Faculty of Computer Science Dalhousie University Canada School of Computer Science Wuhan University China Department of Computer Science University of Loralai Pakistan College of Computer Science and Engineering University of Hail Saudi Arabia School of Business University of Southern Queensland Australia Department of Information and Computer Science King Fahd University of Petroleum and Minerals Finland Software Engineering Department Lappeenranta-Lahti University of Technology Finland

ISBN: (纸本)9781450396134

Ethics in AI becomes a global topic of interest for both policymakers and academic researchers. In the last few years, various research organizations, lawyers, think tankers, and regulatory bodies get involved in developing AI ethics guidelines and principles. However, there is still debate about the implications of these principles. We conducted a systematic literature review (SLR) study to investigate the agreement on the significance of AI principles and identify the challenging factors that could negatively impact the adoption of AI ethics principles. The results reveal that the global convergence set consists of 22 ethical principles and 15 challenges. Transparency, privacy, accountability and fairness are identified as the most common AI ethics principles. Similarly, lack of ethical knowledge and vague principles are reported as the significant challenges for considering ethics in AI. The findings of this study are the preliminary inputs for proposing a maturity model that assesses the ethical capabilities of AI systems and provides best practices for further improvements. © 2022 ACM.

关键词： Challenges AI Ethics Machine Ethics Systematic Literature Review Principles

来源：评论

学校读者我要写书评

暂无评论

Hybrid Convolutional Neural Networks for Enhanced Detection of Mango Leaf Diseases 6

Hybrid Convolutional Neural Networks for Enhanced Detection ...

引用

6th International Conference on Cybernetics, Cognition and Machine Learning Applications, ICCCMLA 2024

作者： Porna, Shweta Bhattacharjee Kabir, Md. Firoz Rana, Md. Imran Chowdhury Sajol, Md. Saiful Islam Roy, Tonmoy Khan, Mohammad Aman Ullah Adnan, Md. Al Bhavani, Girigula Durga Ahsanullah University of Science and Technology Department of Computer Science and Engineering Dhaka Bangladesh Master's in Information Technology University of the Cumberlands United States Bachelor of Business Administration International American university Los Angeles CA United States Louisiana State University Department of Computer Science and Engineering LA United States Data Analytics & Information Systems Utah State University Logan United States Brac University School of Data and Sciences Department of Computer Science and Engineering Dhaka1212 Bangladesh Noakhali Science And Technology University Bsc in software engineering Bangladesh Andhra Pradesh Amaravati522237 India

ISBN: (纸本)9798331505790

The classification of mango leaf diseases is critical for effective disease management and ensuring high-quality yields in mango cultivation. This paper presents a comprehensive study on using deep learning techniques to classify various mango leaf diseases, leveraging convolutional neural networks (CNNs) and hybrid models. A total of 7,524 images were used in our study. These included 4,000 training samples and 3,524 testing samples. The images were split into eight groups, which were powdery mildew, cutting weevil, anthracnose, bacterial canker, sooty mold, gall midge, healthy, and die back. The suggested method starts with feature extraction using VGG19 and MobileNetB1, then classification using both standalone models (ResNet50V2 + EfficientNetB1 and VGG16 + MobileNetB1). We employed data augmentation techniques like random brightness adjustment, rotation, and flipping to enhance the robustness of the model. We conducted hyperparameter tuning using hyperband and Bayesian optimization to optimize the model's performance. Experimental results demonstrate that the hybrid models achieved superior performance, with ResNet50V2 and EfficientNetB1 attaining a perfect accuracy of 100 % on the test set. These findings highlight the potential of deep learning techniques to improve the accuracy and reliability of mango leaf disease diagnosis, contributing significantly to the advancement of precision agriculture. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Automated Response Generation Using Language Models: An Approach to Enhancing User Interaction 7th

Automated Response Generation Using Language Models: An Appr...

引用

7th International Conference on Adaptive Instructional Systems, AIS 2025, held as part of the 27th HCI International Conference, HCII 2025

作者： Asaju, Christine Vadapalli, Hima School of Computer Science and Applied Mathematics University of Witwatersrand Johannesburg South Africa Academy of Computer Science and Software Engineering University of Johannesburg Johannesburg South Africa

ISBN: (纸本)9783031929663

Structured feedback based on specific queries is crucial for users and service providers in various sectors, such as education, industry, entertainment, and healthcare. This process enables all stakeholders to obtain specific and direct feedback, helping them gauge their interaction with the resources/material provided and improving overall human-computer interactions. Incorporating general and specific feedback mechanisms, especially in e-learning, must be strengthened to enhance student and teacher satisfaction while interacting with e-learning material, e.g., lecture videos. The proposed work explores deep learning language models that can take in narrative reports (student/teacher feedback reports) built using user feedback and generate responses to specific questions posed by students/teachers. This process supports the requirements of students and lecturers who want to reflect on particular aspects of their learning/delivery. Usability studies reported that a large percentage of the responses (80% and 90% for single and group reports, respectively) generated during the experimental evaluation were in line with the questions posed, suggesting that the proposed pipeline performed well in response generation. Automating responses by synthesizing narrative reports by utilising a language model has the potential to provide insights into student learning affect. The proposed model is limited by the narrative reports produced by the previous models in the cascade. When incorporated, other modalities linked to learning can improve outcomes and result in a robust system. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Human computer interaction

来源：评论

学校读者我要写书评

暂无评论

Akane: Perplexity-Guided Time Series Data Cleaning

引用

Proceedings of the ACM on Management of Data 2024年第3期2卷 1-26页

作者： Xiaoyu Han Haoran Xiong Zhenying He Peng Wang Chen Wang X. Sean Wang School of Computer Science Fudan University Shanghai China Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China National Engineering Research Center for Big Data Software EIRI Tsinghua University Beijing China

Dirty data are prevalent in time series, such as energy consumption or stock data. Existing data cleaning algorithms present shortcomings in dirty data identification and unsatisfactory cleaning decisions. To handle these drawbacks, we leverage inherent recurrent patterns in time series, analogize them as fixed combinations in textual data, and incorporate the concept of perplexity. The cleaning problem is thus transformed to minimize the perplexity of the time series under a given cleaning cost, and we design a four-phase algorithmic framework to tackle this problem. To ensure the framework's feasibility, we also conduct a brief analysis of the impact of dirty data and devise an automatic budget selection strategy. Moreover, to make it more generic, we additionally introduce advanced solutions, including an ameliorative probability calculation method grounded in the homomorphic pattern aggregation and a greedy-based heuristic algorithm for resource savings. Experiments on 12 real-world datasets demonstrate the superiority of our methods.

关键词： data cleaning perplexity recurrent patterns time series

来源：评论

学校读者我要写书评

暂无评论

Homogeneous Architecture Augmentation and Confidence Prediction for Evolutionary Neural Architecture Search

Homogeneous Architecture Augmentation and Confidence Predict...

引用

Congress on Evolutionary Computation

作者： Pengcheng Jiang Yu Xue Ferrante Neri School of Software Nanjing University of Information Science and Technology Nanjing China School of Computer Science and Electronic Engineering University of Surrey Guildford United Kingdom

ISBN: (数字)9798331534318

ISBN: (纸本)9798331534325

Evolutionary neural architecture search (ENAS) automates the design of high-performing neural networks but is often hindered by the high computational cost of evaluating individual architectures. Surrogate models mitigate this issue by predicting performance, yet their accuracy depends on the quality of training data and their ability to utilise insights from real evaluations. This paper presents homogeneous encoding-based ENAS (HENAS), a novel method addressing these challenges through two key innovations: homogeneous architecture augmentation and confidence-based prediction. Through homogeneous architecture augmentation, HENAS exploits redundant encodings in the MobileNetV3 search space to generate multiple representations of the same architecture, enhancing the surrogate model’s training data without additional cost. Confidence-based prediction introduces a mechanism to identify architectures with uncertain performance estimates, prioritising them for evaluation. Integrated into an evolutionary framework, these techniques improve search efficiency and exploration. Experiments on CIFAR-10, CIFAR-100, and ImageNet show that HENAS achieves state-of-the-art performance with reduced computational expense. Ablation studies confirm the contributions of its core components, highlighting the value of redundancy exploitation and uncertainty management in surrogate-assisted ENAS.

关键词： Technological innovation Uncertainty Computational modeling Redundancy Training data computer architecture Evolutionary computation Predictive models Data models Neural architecture search

来源：评论

学校读者我要写书评

暂无评论

Dina: Deformable Interaction Analogy

SSRN

引用

SSRN 2024年

作者： Huang, Zeyu Dai, Sisi Xu, Kai Zhang, Hao Huang, Hui Hu, Ruizhen College of Computer Science and Software Engineering Shenzhen University China School of Computer Science National University of Defense Technology China School of Computing Science Simon Fraser University Canada

We introduce Deformable Interaction Analogy (DINA) to create similar interactions between 3D objects. Our goal is to generate multiple analogous 3D interactions between an anchor object and different target objects based on a single demo interaction. To achieve this, we optimize the anchor object's pose or shape to mimic the demo interaction with new target objects. We propose using an Interaction Interface (ITF) as a descriptive and robust representation of interactions. For interaction analogy, we transform the ITF to guide feature matching during the anchor object's repositioning and deformation. Qualitative and quantitative experiments demonstrate the effectiveness of our ITF-guided deformable interaction analogy, even when compared to state-of-the-art methods that utilize more complex interaction representations and feature learning from large datasets. © 2024, The Authors. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A Multiple-Path Learning Neural Network Model for Code Completion

A Multiple-Path Learning Neural Network Model for Code Compl...

引用

2023 IEEE International Conference on Web Services, ICWS 2023

作者： Liu, Yi Liu, Jianxun Zhang, Xiangping Hu, Haize Hunan University of Science and Technology Hunan Provincial Key Laboratory for Services Computing and Novel Software Technology Xiangtan411201 China Hunan University of Humanities Science and Technology School of Informationn Loudi417000 China Hunan University of Science and Technology School of Computer Science and Engineering Xiangtan411201 China

ISBN: (纸本)9798350304855

Code completion, which can accelerate the software development process and improve the quality of software products, is an essential part of today's integrated development environments. It has become an important research topic in the field of software engineering. Recent studies have shown that the method of code completion based on the Abstract Syntax Tree (AST) learns syntactic information about the code, which helps to improve the accuracy of code completion. However, when modeling neural networks for ASTs, the sequencing operation of nodes leads to the loss of their hierarchical structure information. Meanwhile, traditional neural networks cannot predict many Out-of-Vocabulary (OoV) words in the terminal node values of AST. To alleviate the above problem, in this paper, we propose a Multiple-Path Learning neural network model for code completion (MPL) based on an AST by learning from a large-scale corpus. In this model, multiple paths such as context path, root path, and terminal node path are established to understand different code features required for node prediction and improve code representation ability. Based on the principle of program local repeatability, it also adopts a replication mechanism to copy the appropriate OoV words from the local terminal node path as the prediction result, further improving the prediction accuracy. The experimental results show that the MPL model has better performance than existing methods on the code completion task. © 2023 IEEE.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

Interpretability Research of Variational Autoencoder Generation Process Based on Feature Disentanglement 5

Interpretability Research of Variational Autoencoder Generat...

引用

5th International Seminar on Artificial Intelligence, Networking and Information Technology, AINIT 2024

作者： Xi, Sicheng Peng, Yanhui College of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing Key Laboratory of Computational Intelligence Chongqing China School of Software Engineering Chongqing University of Posts and Telecommunications Chongqing Key Laboratory of Computational Intelligence Chongqing China

ISBN: (纸本)9798350385557

Variational Autoencoder (VAE), as one of the main generative models, has a powerful representation learning capability. However, the hidden space representation learned by VAE is a high-dimensional and complex vector space, which makes it difficult to explain how the model gradually learns and composes the final generated results on different semantic features. To address this problem, firstly, this paper increases the degree of decoupling between different semantic features by increasing the independence between the hidden variables of the modal hermitian space, and explains the learning process of the model on different hermitian spaces by visualization based on the feature decoupling model. In addition, this paper also proposes a hidden variable contribution index to measure the influence of different dimensional hidden variables on the generation results, so as to explain the learning process of the model. © 2024 IEEE.

关键词： Learning systems

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：