检索结果-内蒙古大学图书馆

A new entity extraction model based on journalistic brazilian portuguese language to enhance named entity recognition 1st

学校读者我要写书评

暂无评论

A new entity extraction model based on journalistic brazilia...

1st EAI International Conference on data and Information in Online Environments, DIONE 2020

作者： de Aquino Silva, Rogerio da Silva, Luana Dutra, Moisés Lima de Araujo, Gustavo Medeiros Engineering and Data Science Lab Federal University of Santa Catarina Florianópolis Brazil

ISBN: (纸本)9783030500719

Named Entity Recognition (NER) plays an important role on broad natural language processing applicability. According to the literature, the NER process applied to the English language reaches around 90% of accuracy. However, when applied to Portuguese, this accuracy is at most 83.38%. A wide range of algorithms based on LSTM (Long-Short Term Memory) architecture has being proposed to enhance the NER accuracy. However, a key component to a successful information extraction is the corpora used for NER training. In order to improve the NER in Portuguese language, this paper proposes a methodology for training text corpus based on Portuguese-language journalistic corpora. The Journalistic language has the best adherence to the contemporaneity of the language, since it preserves features such as objectivity, simplicity, impartiality, and is a reference of transmitting the information without ambiguity. The proposed methodology provides a model to extract entities and assess the obtained results with the use of Recurrent Neural Network architectures. At the best of our knowledge, with the proposed methodology, the NER task applied to the Portuguese language overcomes the average accuracy found in the literature, increased from 83.38% to 85.64%. Moreover, the use of this methodology could decrease the computational costs related to the NER processing tasks. © ICST Institute for Computer sciences, Social Informatics and Telecommunications engineering 2020.

关键词： Long short-term memory

Online Container Scheduling for Low-Latency IoT Services in Edge Cluster Upgrade: A Reinforcement Learning Approach

学校读者我要写书评

暂无评论

Online Container Scheduling for Low-Latency IoT Services in ...

2023 IEEE/CIC International Conference on Communications in China, ICCC 2023

作者： Cui, Hanshuai Tang, Zhiqing Lou, Jiong Jia, Weijia Beijing Normal University Institute of Artificial Intelligence and Future Networks Zhuhai519087 China Beijing Normal University School of Artificial Intelligence Beijing100875 China Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai200240 China BNU-HKBU United International College Guangdong Key Lab of Ai and Multi-Modal Data Processing Zhuhai519087 China

ISBN: (纸本)9798350345384

In Mobile Edge Computing (MEC), Internet of Things (IoT) devices offload computationally-intensive tasks to edge nodes, where they are executed within containers, reducing the reliance on centralized cloud infrastructure. Frequent up-grades are essential to maintain the efficient and secure operation of edge clusters. However, traditional cloud cluster upgrade strategies are ill-suited for edge clusters due to their geographically distributed nature and resource limitations. Therefore, it is crucial to properly schedule containers and upgrade edge clusters to minimize the impact on running tasks. In this paper, we propose a low-latency container scheduling algorithm for edge cluster upgrades. Specifically: 1) We formulate the online container scheduling problem for edge cluster upgrade to minimize the total task latency. 2) We propose a policy gradient-based reinforcement learning algorithm to address this problem, considering the unique characteristics of MEC. 3) Experimental results demonstrate that our algorithm reduces total task latency by approximately 27% compared to baseline algorithms. © 2023 IEEE.

关键词： Reinforcement learning

How Graph Neural Networks Learn: Lessons from Training Dynamics

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Chenxiao Wu, Qitian Wipf, David Sun, Ruoyu Yan, Junchi School of Artificial Intelligence Department of Computer Science and Engineering MoE Lab of AI Shanghai Jiao Tong University China Amazon Web Services United States School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen International Center for Industrial and Applied Mathematics Shenzhen Research Institute of Big Data China

A long-standing goal in deep learning has been to characterize the learning behavior of black-box models in a more interpretable manner. For graph neural networks (GNNs), considerable advances have been made in formalizing what functions they can represent, but whether GNNs will learn desired functions during the optimization process remains less clear. To fill this gap, we study their training dynamics in function space. In particular, we find that the gradient descent optimization of GNNs implicitly leverages the graph structure to update the learned function, as can be quantified by a phenomenon which we call kernel-graph alignment. We provide theoretical explanations for the emergence of this phenomenon in the overparameterized regime and empirically validate it on real-world GNNs. This finding offers new interpretable insights into when and why the learned GNN functions generalize, highlighting their limitations in heterophilic graphs. Practically, we propose a parameter-free algorithm that directly uses a sparse matrix (i.e. graph adjacency) to update the learned function. We demonstrate that this embarrassingly simple approach can be as effective as GNNs while being orders-of-magnitude faster. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

Environment-aware dynamic graph learning for out-of-distribution generalization 23

学校读者我要写书评

暂无评论

Environment-aware dynamic graph learning for out-of-distribu...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Haonan Yuan Qingyun Sun Xingcheng Fu Ziwei Zhang Cheng Ji Hao Peng Jianxin Li Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University and School of Computer Science and Engineering Beihang University Key Lab of Education Blockchain and Intelligent Technology Guangxi Normal University Department of Computer Science and Technology Tsinghua University Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University

Dynamic graph neural networks (DGNNs) are increasingly pervasive in exploiting spatio-temporal patterns on dynamic graphs. However, existing works fail to generalize under distribution shifts, which are common in real-world scenarios. As the generation of dynamic graphs is heavily influenced by latent environments, investigating their impacts on the out-of-distribution (OOD) generalization is critical. However, it remains unexplored with the following two major challenges: (1) How to properly model and infer the complex environments on dynamic graphs with distribution shifts? (2) How to discover invariant patterns given inferred spatio-temporal environments? To solve these challenges, we propose a novel Environment-Aware dynamic Graph LEarning (EAGLE) framework for OOD generalization by modeling complex coupled environments and exploiting spatiotemporal invariant patterns. Specifically, we first design the environment-aware EADGNN to model environments by multi-channel environments disentangling. Then, we propose an environment instantiation mechanism for environment diversification with inferred distributions. Finally, we discriminate spatio-temporal invariant patterns for out-of-distribution prediction by the invariant pattern recognition mechanism and perform fine-grained causal interventions node-wisely with a mixture of instantiated environment samples. Experiments on real-world and synthetic dynamic graph datasets demonstrate the superiority of our method against state-of-the-art baselines under distribution shifts. To the best of our knowledge, we are the first to study OOD generalization on dynamic graphs from the environment learning perspective.

关键词：

ISSSR 2022 Keynote Speech II: Human-Machine Cooperation in Junior science Education: Development, Opportunities, and Challenges

学校读者我要写书评

暂无评论

ISSSR 2022 Keynote Speech II: Human-Machine Cooperation in J...

International Symposium on System and Software Reliability (ISSSR)

作者： Liang Luo Assistant Director of National & Local Joint Engineering Lab of Next Generation Internet Data Processing Technology School of Computer Science and Engineering (School of Cyberspace Security) University of Electronic Science and Technology of China

ISBN: (数字)9781665453974

ISBN: (纸本)9781665453981

Human-Machine-Environment (HME) Tri-Co (Coexisting, Cooperative, Cognitive) Education is an indispensable prerequisite and the main content of Artificial Intelligence empowering junior science education. Moreover, it is also a critical requirement for improving the quality of education and promoting the high-end development of education information. Building a unique learning environment with scene perception and an intelligent cognitive process is a tremendous challenge for diverse needs in junior scientific learning scenarios. Therefore, it is necessary to study human-machine cooperation scenarios and design advanced software and hardware. This talk will introduce some of our recent work in affective analysis, multimodal affection data processing methods, and reliability and security of affective computing systems, etc. We will also discuss some directions that are worthy of further studies.

关键词：

A Review of Algorithms, datasets, and Criteria in Word Sense Disambiguation with a View to its Use in Islamic Texts 8

学校读者我要写书评

暂无评论

A Review of Algorithms, Datasets, and Criteria in Word Sense...

8th Iranian Joint Congress on Fuzzy and Intelligent Systems, CFIS 2020

作者： Farahani, Yasamin Vasheghani Janfada, Behrooz Bidgoli, Behrooz Minaei University of Science and Technology Data Mining Lab Faculty of Computer Engineering Tehran Iran

ISBN: (数字)9781728173016

ISBN: (纸本)9781728173016

Most of the words in a natural language are polysemous;it means, they have multiple senses. Word sense disambiguation is the process of determining the exact sense of the polysemous word by analyzing the context in which the word occurs. Word sense disambiguation is one of the fundamental tasks in the field of natural language processing. However, There have been few studies in word sense disambiguation in Persian and Arabic languages. For the above reason and considering the abundance of ambiguous words and phrases in Islamic texts, the need for a special algorithm to help the disambiguation of such texts is felt. The present study attempted to have an overall review of word sense disambiguation approaches, strategies and issues, in order to open the way for future work on the presentation of similar algorithms for Islamic texts. We review approaches such as semi-supervised, unsupervised, supervised and knowledge-based approaches. The best strategy will be introduced, and the state-of-the-art systems related to it will be compared. © 2020 IEEE.

关键词： Knowledge based systems

Vision-Based Multimodal Interfaces: A Survey and Taxonomy for Enhanced Context-Aware System Design

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hu, Yongquan Tang, Jingyu Gong, Xinya Zhou, Zhongyi Zhang, Shuning Elvitigala, Don Samitha Mueller, Florian Hu, Wen Quigley, Aaron J. School of Computer Science and Engineering University of New South Wales Sydney Australia The University of Tokyo Tokyo Japan Exertion Games Lab Department of Human-Centred Computing Monash University Melbourne Australia School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Tsinghua University Beijing China Department of Computer Science and Engineering South University of Science and Technology Shenzhen China CSIRO’s Data61 University of New South Wales Sydney Australia

The recent surge in artificial intelligence, particularly in multimodal processing technology, has advanced human-computer interaction, by altering how intelligent systems perceive, understand, and respond to contextual information (i.e., context awareness). Despite such advancements, there is a significant gap in comprehensive reviews examining these advances, especially from a multimodal data perspective, which is crucial for refining system design. This paper addresses a key aspect of this gap by conducting a systematic survey of data modality-driven Vision-based Multimodal Interfaces (VMIs). VMIs are essential for integrating multimodal data, enabling more precise interpretation of user intentions and complex interactions across physical and digital environments. Unlike previous task- or scenario-driven surveys, this study highlights the critical role of the visual modality in processing contextual information and facilitating multimodal interaction. Adopting a design framework moving from the whole to the details and back, it classifies VMIs across dimensions, providing insights for developing effective, context-aware *** Codes 68U35, 68T07 Copyright © 2025, The Authors. All rights reserved.

关键词： Taxonomies

A Joint Communication and Computation Framework for Digital Twin over Wireless Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yang, Zhaohui Chen, Mingzhe Liu, Yuchen Zhang, Zhaoyang The College of Information Science and Electronic Engineering Zhejiang University Zhejiang Hangzhou310027 China Zhejiang Hangzhou310007 China Zhejiang Lab Zhejiang Hangzhou311100 China The Department of Electrical and Computer Engineering Institute for Data Science and Computing University of Miami Coral GablesFL33146 United States Department of Computer Science North Carolina State University RaleighNC27606 United States

In this paper, the problem of low-latency communication and computation resource allocation for digital twin (DT) over wireless networks is investigated. In the considered model, multiple physical devices in the physical network (PN) needs to frequently offload the computation task related data to the digital network twin (DNT), which is generated and controlled by the central server. Due to limited energy budget of the physical devices, both computation accuracy and wireless transmission power must be considered during the DT procedure. This joint communication and computation problem is formulated as an optimization problem whose goal is to minimize the overall transmission delay of the system under total PN energy and DNT model accuracy constraints. To solve this problem, an alternating algorithm with iteratively solving device scheduling, power control, and data offloading subproblems. For the device scheduling subproblem, the optimal solution is obtained in closed form through the dual method. For the special case with one physical device, the optimal number of transmission times is reveled. Based on the theoretical findings, the original problem is transformed into a simplified problem and the optimal device scheduling can be found. Numerical results verify that the proposed algorithm can reduce the transmission delay of the system by up to 51.2% compared to the conventional schemes. © 2024, CC BY-SA.

关键词： Budget control

A Physics-based and data-driven Approach for Localized Statistical Channel Modeling

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Shutao Ning, Xinzhi Zheng, Xi Shi, Qingjiang Chang, Tsung-Hui Luo, Zhi-Quan School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Shenzhen Research Institute of Big Data Shenzhen China School of Software Engineering Tongji University Shanghai China Pengcheng Lab Shenzhen China Networking and User Experience Lab Huawei Technologies Shenzhen China

Localized channel modeling is crucial for offline performance optimization of 5G cellular networks, but the existing channel models are for general scenarios and do not capture local geographical structures. In this paper, we propose a novel physics-based and data-driven localized statistical channel modeling (LSCM), which is capable of sensing the physical geographical structures of the targeted cellular environment. The proposed channel modeling solely relies on the reference signal receiving power (RSRP) of the user equipment, unlike the traditional methods which use full channel impulse response matrices. The key is to build the relationship between the RSRP and the channel’s angular power spectrum. Based on it, we formulate the task of channel modeling as a sparse recovery problem where the non-zero entries of the sparse vector indicate the channel paths’ powers and angles of departure. A computationally efficient weighted non-negative orthogonal matching pursuit (WNOMP) algorithm is devised for solving the formulated problem. Finally, experiments based on synthetic and real RSRP measurements are presented to examine the performance of the proposed method. © 2023, CC BY.

关键词： Statistical Physics