检索结果-内蒙古大学图书馆

5th International Conference on Computer Vision and data Mining, ICCVDM 2024

作者： He, Ting Chen, Wei Zhang, Jun Li, Shuohao Hu, Shengze Laboratory for Big Data and Decision National University of Defense Technology Hunan Changsha410073 China

ISBN: (纸本)9781510682863

Scene Graph Generation (SGG) aims to bridge images and text. Addressing the issues of data distribution imbalance and long-tail effects in existing SGG methods, we propose a new Multi-feature Fusion Embedding Network(MFE-Net). This network optimizes relationship prediction through multi-level visual feature fusion and hard-constrained semantic feature classification, aiming to reduce inter-class confusion and strengthen intra-class consistency. Experiments on the Visual Genome dataset demonstrate that our model significantly enhances the accuracy of relationship recognition, achieving a new level of performance. The research not only optimizes SGG performance but also provides new strategies for addressing classification problems involving easily confused relationships. © 2024 SPIE.

关键词： Network embeddings

来源：评论

学校读者我要写书评

暂无评论

Learning-based Bundling Strategy for Two Products under Uncertain Consumer's Valuations

引用

IEEE Transactions on Engineering Management 2025年 72卷 1970-1982页

作者： Fang, Lang Tang, Jiafu Pan, Zhendong Dongbei University of Finance and Economics School of Management Science & Engineering Liaoning Key Laboratory of Big Data Management and Optimization Decision Dalian116025 China

Should a firm engage in bundling to boost revenue when consumer's valuations of products are heterogeneous and uncertain? In recent years, technological advances have made it possible for firms to use large amounts of available data to make decisions under demand-side information uncertainty. However, it remains unclear exactly how they can dynamically optimize bundling and pricing decisions by learning from uncertain consumer's valuations. To answer this question, we study the bundling and pricing decisions of a monopoly firm offering a basic product and a premium product over a finite T periods. We first analyze the equilibrium outcomes of pure component strategies (PCS) and pure bundling strategies (PBS) under deterministic consumer's valuations. We then introduce a learning-based bundling strategy (LBBS) framework to make decisions dynamically over time. It employs Thompson Sampling to estimate the fractions of low-valuation consumers (LVC) of two products, allowing the firm to adjust its decisions based on updated beliefs about consumer's valuation distributions. We demonstrate the robust performance of the LBBS and show the interesting findings. That is, the fraction of LVC of premium product, the ratio between high- and low-valuation (RHL) of basic product, the ratio of low-valuation (RLV) and the ratio of high-valuation (RHV) of two products wield significant influence on the adoption of bundling strategy and its possibility. These findings offer practical guidance to firm's practitioners regarding when and which PBS to adopt and how to improve PBS decisions according to the fractions of LVC of products. We also extend the model to uniform distribution of consumer's valuations and correlated consumer's valuations to test our results. © 1988-2012 IEEE.

关键词： Cost accounting Pricing Training data mining Artificial intelligence Uncertainty Profitability Supply chain management Sustainable development Surveys

来源：评论

学校读者我要写书评

暂无评论

Solving Combinatorial Optimization Problem Over Graph Through QUBO Transformation and Deep Reinforcement Learning 24

Solving Combinatorial Optimization Problem Over Graph Throug...

引用

24th IEEE International Conference on data Mining, ICDM 2024

作者： Pu, Tianle Chen, Chao Zeng, Li Liu, Shixuan Sun, Rui Fan, Changjun College of Systems Engineering National University of Defense Technology Laboratory for Big Data and Decision Changsha China

ISBN: (纸本)9798331506681

Many complex problems encountered in both production and daily life can be conceptualized as combinatorial optimization problems (COPs). Many ad-hoc deep learning methods have been proposed to solve these problems, but there still lacks an effective unified framework. In this work, we take a step towards this goal by designing an unified end-to-end deep reinforcement learning framework named CONQUER. CONQUER first translates various COPs into the unified QUBO (Quadratic Unconstrained Binary Optimization) formalization, and afterwards trains a neural QUBO solver to search high-quality strategies, which are then checked and repaired to obtain the solutions for original COPs. The QUBO solver is a purely data-driven neural model, which does not rely on any expert inputs. We adopt the graph transformer network to represent the QUBO graph, so as to effectively capture the features of problem objective and constraints. The Deep Q-network is utilized to train the QUBO solver to find long-sighted solution strategies. Despite being trained solely on small synthetic graphs, CONQUER exhibits promising generality on much larger instances, and can be easily adapted to real-world datasets. Experimentally, we show that CONQUER achieves the state-of-the-art performances on four COPs over graphs, including minimum vertex cover, maximum independent set, maximum clique and maximum cut problem. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Robustness Analysis and Evaluation Study of Chinese Text Event Detection Models 9

Robustness Analysis and Evaluation Study of Chinese Text Eve...

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Chen, Haiwen Liu, Kaiyue Liu, Pengyu Wang, Ji Ge, Ningchao National University of Defense Technology Laboratory for Big Data and Decision Changsha China

ISBN: (纸本)9798350330076

Event detection aims to identify trigger words in given text and classify them into corresponding event types, making it an essential task in natural language processing. With the advancement of artificial intelligence, machine learning, and deep learning techniques have been further researched and applied in the field of event detection. However, current event detection methods heavily rely on manually labeled data, which is constrained by high-quality data requirements. When facing challenges such as imbalanced sample distribution, low-quality data labeling, and adversarial data contamination, the effectiveness of models is often compromised, leading to a lack of robustness and poor generalization capabilities. This significantly limits the universal applicability of event detection algorithms. Therefore, the robustness evaluation of text event detection algorithms is of paramount importance. In this paper, we have designed various text perturbation robustness datasets based on the characteristics of Chinese text and developed evaluation metrics specifically for robustness datasets to measure algorithm robustness. Building upon the robustness datasets, we have devised a resilient analysis framework. We have constructed a text event detection algorithm using multi-layer perceptrons, convolutional neural networks, and BERT. By comparing the model's evaluation results on the robustness test dataset with the high-quality initial dataset, we intuitively reflect the model's robustness. This approach holds strong theoretical value for further enhancing the model's generalization capabilities. © 2023 IEEE.

关键词： Multilayer neural networks

来源：评论

学校读者我要写书评

暂无评论

Few-shot Class Incremental Learning for Event Extraction 9

Few-shot Class Incremental Learning for Event Extraction

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Tian, Xuemeng Ge, Bin Xu, Chi National University of Defense Technology Laboratory for Big Data and Decision ChangSha China

ISBN: (纸本)9798350330076

In the field of natural language processing, research on incremental learning is relatively scarce compared to computer vision. Most existing studies are limited to tasks such as named entity recognition(NER), automatic question answering and text classification, and have not been applied to event extraction(EE). Considering the characteristics of EE (Continuously emerging new tasks and knowledge;Limited training samples for new tasks in real-life scenarios), our study focuses on a more challenging yet practical problem, i.e., fewshot class incremental learning for EE, which aims to conduct incremental learning under the condition of few samples, enabling it to perform well on new tasks while avoiding the phenomenon of "catastrophic forgetting". To address the fewshot problem, we generate a knowledge-enhanced EE model. and in terms of retaining old knowledge, inspired by knowledge distillation, regularizing constraints on multiple modules. Experimental results demonstrate that our approach effectively addresses the issues and improves the performance of event extraction models. © 2023 IEEE.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph Completion with Fused Factual and Commonsense Information 20th

Knowledge Graph Completion with Fused Factual and Commonse...

引用

Proceedings of the 20th Web Information Systems and Applications Conference, WISA 2023

作者： Liu, Changsen Tang, Jiuyang Zeng, Weixin Wu, Jibing Huang, Hongbin Laboratory for Big Data and Decision National University of Defense Technology ChangSha China

ISBN: (纸本)9789819962211

Knowledge graph contains a large number of factual triples, but it is still inevitably incomplete. Previous commonsense-based knowledge graph completion models used concepts to replace entities in triplets to generate high-quality negative sampling and link prediction from the perspective of joint facts and commonsense. However, they did not consider the importance of concepts and their correlation with relationships, resulting in a lot of noise and ample room for improvement. To address this problem, we designed commonsense for knowledge graph completion and filtered the commonsense knowledge based on the analytic hierarchy process. The obtained commonsense can further improve the quality of negative samples and the effectiveness of link prediction. Experimental results on four datasets of the knowledge graph completion (KGC) task show that our method can improve the performance of the original knowledge graph embedding (KGE) model. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Granular Sequential Three-Way decision for Specific decision Classes

引用

IEEE Transactions on Fuzzy Systems 2025年第5期33卷 1650-1663页

作者： Cheng, Yunlong Yang, Xiuhua Zhang, Qinghua Shao, Yabin Wang, Guoyin Chongqing University of Posts and Telecommunications School of Science Chongqing400065 China Chongqing University of Posts and Telecommunications Chongqing Key Laboratory of Tourism Multisource Data Perception and Decision Ministry of Culture and Tourism Key Laboratory of Big Data Intelligence Computing Chongqing Key Laboratory of Computational Intelligence Chongqing400065 China

Sequential three-way decision (S3WD) is an efficient granular computing paradigm for dealing with uncertain problems. However, it is primarily oriented to all decision classes, which contradicts the fact that decisions are typically for the specific decision classes. Meanwhile, most S3WD models hide the topological structure of the granules, leading to difficulties in semantic interpretation. To address the issues, integrating model construction, attribute reduction and knowledge extraction, a general framework of granular sequential three-way decision (GrS3WD) for the specific decision classes is proposed to improve semantic interpretation and computational efficiency. First, a two-stage trisecting strategy and a GrS3WD model are proposed to integrate model construction with attribute reduction. Its main advantage is that it retains the topological structure of granules, which not only enhances semantic interpretation but also avoids unnecessary double counting. Second, three acceleration strategies and a novel granular sequential three-way reduction (GrS3WR) algorithm are proposed to fast obtain a classification-based reduct or a class-specific reduct. Finally, the decision rules with multi-granularity can be directly extracted from the concept tree generated by GrS3WR. Experimental results demonstrate that a class-specific reduct usually has fewer attributes and better classification performance than a classification-based reduct. Moreover, GrS3WR can significantly improve the computational efficiency of attribute reduction. © 1993-2012 IEEE.

关键词： Granular computing

来源：评论

学校读者我要写书评

暂无评论

Research on Optimal Design of Reaction Conditions Based on Mixed Statistical Model 2

Research on Optimal Design of Reaction Conditions Based on M...

引用

2nd International Conference on Applied Statistics, Computational Mathematics, and Software Engineering, ASCMSE 2023

作者： Zhao, Runhao Zhang, Jiayu Wang, Xianrui Su, Bolun Tan, Junze Ben, Niu Laboratory for Big Data and Decision National University of Defense Technology Changsha China

ISBN: (数字)9781510668195

ISBN: (纸本)9781510668188

In the preparation of chemical raw materials, there are often some optimization problems that are difficult for people to solve. Due to the complex interaction between various factors and the great difference between the theoretical model and the practice, an effective optimal design model is imminent. In this paper, an innovative mixed statistical model is proposed for optimal design, including five sub-models of ANOVA model, polynomial optimal regression analysis model, design uniformity model, two-step regression model, and optimization model. In this paper, the mixed statistical model is also applied to the study of the combination of catalyst and temperature in the preparation of C4 olefin production, which has high practicability. © 2023 SPIE.

关键词： Optimal systems

来源：评论

学校读者我要写书评

暂无评论

KDDLSA:A Knowledge Extraction Framework Integrating Deep Learning and Syntactic Analysis 9

KDDLSA:A Knowledge Extraction Framework Integrating Deep Lea...

引用

9th International Conference on big data and Information Analytics, bigDIA 2023

作者： Li, Jiaxu Ge, Bin Zhang, Chong Duan, Haopeng Chen, Zhihao Liu, Kaiyue National University of Defense Technology Laboratory for Big Data and Decision Changsha China

ISBN: (纸本)9798350330076

Knowledge Extraction (KE) aims to extract entity and relationship information from text, serving as a foundational resource for downstream Natural Language Processing (NLP) tasks. While knowledge extraction methods in general domains are well-established, their performance diminishes when applied to specialized fields, as these domains often encompass unique professional information absent in broader contexts. In this paper, we introduce a knowledge extraction framework that amalgamates deep learning and syntactic analysis. This framework encompasses deep learning-based named entity recognition and relation extraction grounded in dependency parsing(KDDLSA). Specifically, we construct an external knowledge base tailored to specific fields through named entity recognition, thereby providing expert knowledge for relation extraction. Moreover, we devise a relation extraction algorithm by scrutinizing the nuances of Chinese grammar. Through the formulation of extraction rules, we pinpoint segments of the dependency syntax tree that adhere to these rules, enhancing the efficacy of knowledge extraction within specific domains. To validate our approach, we constructed a dataset in the maritime security domain using publicly available military news. Experimental results reveal a noteworthy performance enhancement when applying our method to the constructed dataset. © 2023 IEEE.

关键词： Syntactics

来源：评论

学校读者我要写书评

暂无评论

Joint Resource Optimization of Mobile Edge Computing in Disaster Areas Based on 5G Communication Network

Joint Resource Optimization of Mobile Edge Computing in Disa...

引用

2023 International Conference on Artificial Intelligence of Things and Systems, AIoTSys 2023

作者： Zhang, Shuai Yan, Hui Li, Xiaoqing Bao, Weidong Wang, Ji National University of Defense Technology Laboratory for Big Data and Decision Changsha China

ISBN: (纸本)9798350312270

In disaster-prone regions, cloud servers may face functionality impairments. Therefore, the emergence of the 5G era has led to the use of Unmanned Aerial Vehicles (UAVs) with edge servers as a promising alternative to centralized cloud servers for expedited restoration of computing capabilities in emergency situations. In this paper, we propose a multi-objective genetic particle swarm nested greedy algorithm (MOGAPSO-G) that optimizes energy usage and time delay within the system. By utilizing the Pareto optimal set obtained from MOGAPSO-G, one can comprehend the trade-offs between energy consumption and time delay, thus enabling the customization of UAV location deployment and task offloading strategies based on specific user requirements. Compared to the NSGA-II algorithm, MOPSO-G, the proposed MOGAPSO-G approach exhibits superior performance in terms of time delay and the Pareto frontier distribution. Notably, our method presents greater proximity to the true Pareto boundary, enhanced adaptability compared to the MOPSO algorithm, and increased robustness and practical applicability. This approach providing infrastructure support for AIoT computing that can significantly improve system efficiency and resilience during emergencies. © 2023 IEEE.

关键词： Time delay

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：