检索结果-内蒙古大学图书馆

Embedding VLAD in Transformer for Video Question Answering

Jisuanji Xuebao/Chinese Journal of Computers 2023年第4期46卷 671-689页

作者： Guo, Dan Yao, Shen-Tao Wang, Hui Wang, Meng School of Computer and Information Engineering Hefei University of Technology Hefei230601 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230094 China Key Laboratory of Knowledge Engineering with Big Data（Hefei University of Technology） Ministry of Education Hefei230601 China Intelligent Interconnected Systems Laboratory of Anhui Province（Hefei University of Technology） Hefei230601 China

Video question answering（VideoQA）is a typical cross-modal understanding task. Its challenge lies in how to learn appropriate multimodal representation and cross-modal correlation for answer inference. Most existing video question answering methods focus on the latter，e. g.，relationship learning between each video frame or clip and word. In this work， we devote to advanced feature embedding of both video and query. We develop a clustering-based VLAD technique for VideoQA. The novelty of our work is the joint exploitation of temporal aggregation and correlation in multimodality. We propose an end-to-end trainable Transformed VLAD embedding network， named TVLAD-Net. TVLAD-Net constructs a differentiable aggregation network module（i. e.， convolutional Residual-less VLAD Block）to generate compact VLAD descriptors（transforming N frames，clips or words to compact K descriptors while K © 2023 Science Press. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Vulnerabilities Static Detection for Web Applications with False Positive Suppression

Vulnerabilities Static Detection for Web Applications with F...

引用

2010 IEEE International Conference on Information Theory and Information Security(2010 IEEE 国际信息论与信息安全会议)

作者： Bin Liang Qianqian Wang Jianjun Huang Jiagui Zhong Jingjing Cai Key Laboratory of Data Engineering and Knowledge Engineering (Renmin University of China) MOE Schoo School of Information Renmin University of China Beijing 100872 China Beijing Venustech Cybervision Co. Ltd Beijing 100094 China

Web applications become more and more important, and the corresponding security problems have been concerned about. This paper presents TASA, an ASP static analyzer, which employs a path-sensitive, inter-procedural and contextsensitive data flow analysis, mainly concerning the taint propagation and sanitization. This paper also discusses some techniques used in TASA, such as sanitization routines modeling, ASP specific features, alias analysis and path-related routines modeling, to prune false positives. Experiments on four open source applications show that TASA has a rate of false positive of 4.98% and it can avoid certain false warnings owing to the proposed approaches.

关键词： Security Detectors Computer bugs Portals Analytical models Browsers Merging

来源：评论

学校读者我要写书评

暂无评论

Global and Adaptive Local Label Correlation for Multi-label Learning with Missing Labels

Global and Adaptive Local Label Correlation for Multi-label ...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Qingxia Jiang Peipei Li Yuhong Zhang Xuegang Hu Key Laboratory of Knowledge Engineering with Big Data (the ministry of Education) Hefei University of Technology School of Computer Science and Information Engineering Hefei University of Technology Hefei Anhui Province China Anhui Province Key Laboratory of Industry Safety and Emergency Technology Hefei Anhui Province China

Label missing is a major challenge in multi-label learning. Many existing methods try to use label correlation to recover ground-truth labels, but they only focus on the label correlation within the original label space, however, the label correlation learned in this way is incomplete. Thus, inspired $b$ y the matrix adaptive column correlation, we propose a method to continuously adjust the label correlation matrix while the labels are filled in by adaptive column correlation learning method. Specifically, to reduce the impact of the missing label s on label correlation, the label space is firstly completed through manifold regularization while learning the local label information by adaptive column correlation learning in the complemented label space. Secondly, the global label correlation is utilized by adding a low-rank constraint to the entire label space. Finally, by jointly taking advantage of the global and adaptive local label correlation, our proposed approach achieves superior performance on both synthetic and real-world data sets from diverse domains compared to state-of-the art baselines.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Synchronous replication framework of databases servers supporting reliable electronic commerce applications

引用

Journal of Computational Information Systems 2009年第2期5卷 849-858页

作者： Liao, Guoqiong Chen, Shijun School of Information Technology Jiangxi University of Finance and Economics Nanchang 330013 China Jiangxi Key Laboratory of Data and Knowledge Engineering Nanchang 330013 China Siemens Ltd. China Corporate Technology Beijing 100102 China

To support electronic commerce applications with high reliability requirements, a synchronous replication framework of database servers is designed in this paper. At first, a three-tier architecture consisting of Web Servers (WS), Electronic Commerce Servers (ECS) and database Servers (DBS) for a reliable electronic commerce system is presented. Then, a primary copy synchronous replication model under the structure is given. Finally, a data synchronization method using one-phase commit (1PC) protocol for the model is discussed in details. The analysis result shows that 1PC can reduce response time than the two-phase commit protocol. © 2009 Binary Information Press March, 2009.

关键词： Electronic commerce

来源：评论

学校读者我要写书评

暂无评论

Efficient Processing of Global Top-k Aggregate Queries in Uncertain database

Efficient Processing of Global Top-k Aggregate Queries in Un...

引用

International Conference on Information engineering and Computer Science (ICIECS)

作者： Xianli Xing Dexi Liu Principal's Office Jiangxi University of Finance and Economics Nanchang China School of Information Technology Jiangxi Key Laboratory of Data and Knowledge Engineering Jiangxi University of Finance and Economics Nanchang China

ISBN: (纸本)9781424449934

Top-k queries in uncertain databases are quite popular and useful due to its wide application usage. However, compared to top-k in traditional databases, queries over uncertain database are more complicated because of the existence of exponential possible worlds. A Top-k aggregate query ranks groups of tuples by their aggregate values, sum or average for example, and returns k groups with the highest aggregate values. As a powerful semantic of top-k, global top-k, returns A highest-ranked tuples according to their probabilities of being in the top-k anTopswers in possible worlds. We propose a dynamic programming based method to process global top-k aggregate queries in uncertain database, where the number of retrieved tuples and group states generated on these tuples are minimized. Experiment results show that our algorithm is effective.

关键词： Aggregates databases Finance Power generation economics Dynamic programming Information technology Laboratories knowledge engineering Information retrieval Performance analysis

来源：评论

学校读者我要写书评

暂无评论

An Optimization-Based Truth Discovery Method with Claim Relation

An Optimization-Based Truth Discovery Method with Claim Rela...

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Jiazhu Xia Ying He Yuxin Jin Xianyu Bao Gongqing Wu Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Ministry of Education Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Shenzhen Academy of Inspection and Quarantine Shenzhen China

With the advent of the era of big data, the information from multi-sources often conflicts due to that errors and fake information are inevitable. Therefore, how to obtain the most trustworthy or true information (i.e. truth) people need gradually becomes a troublesome problem. In order to meet this challenge, a novel hot technology named truth discovery that can infer the truth and estimate the reliability of the source without supervision has attracted more and more attention. However, most existing truth discovery methods only consider that the information is either same or different rather than the fine-grained relation between them, such as inclusion, support, mutual exclusion, etc. Actually, this situation frequently exists in real-world applications. To tackle the aforementioned issue, we propose a novel truth discovery method named OTDCR in this paper, which can handle the fine-grained relation between the information and infer the truth more effectively through modeling the relation. In addition, a novel method of processing abnormal values is applied to the preprocessing of truth discovery, which is specially designed for categorical data with the relation. Experiments in real dataset show our method is more effective than several outstanding methods.

关键词： Task analysis knowledge engineering Big data Crowdsourcing Reliability engineering Meteorology

来源：评论

学校读者我要写书评

暂无评论

Idea-map: A Spatiotemporal View of Research Ideas

Idea-map: A Spatiotemporal View of Research Ideas

引用

2011 Eighth International Conference on Fuzzy System and knowledge Discovery(第八届模糊系统与知识发现国际会议 FSKD 2011)

作者： HeHu Xiaoyong Du Key Laboratory of Data Engineering and Knowledge Engineering(Renmin University of China) MOE Beijing 100872 P. R. China School of Information Renmin University of China Beijing 100872 P. R. China

in this paper, we describe idea-map, a mash-up application building on top of Linked data Cloud. It reads in user's keywords (about research ideas) and executes a SPARQL query against DBLP endpoint. Spatial and temporal information is extracted and parsed from the query results and is further transformed to SIMILE/EXHIBIT to show a spatiotemporal map for the research ideas. Idea-map shows the feasibility of combing various techniques such as YQL, SIMILE/Exhibit and SPARQL query answering to provide an insightful interface to better understand the interested research ideas.

关键词： Link data Cloud DBLP Spatiotemoral View YQL

来源：评论

学校读者我要写书评

暂无评论

Research on Crowdsourcing Truth Inference Method Based on Graph Embedding

Research on Crowdsourcing Truth Inference Method Based on Gr...

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Liangzhu Zhou Xingrui Zhuo Gongqing Wu Zan Zhang Xianyu Bao Key Laboratory of Knowledge Engineering with Big Data of Ministry of Education Hefei University of Technology Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Shenzhen Academy of Inspection and Quarantine Shenzhen China

ISBN: (纸本)9781665438599

Crowdsourcing is a cheap and popular method to solve problems that are difficult for computers to handle. Due to the differences in ability among workers on crowdsourcing platforms, existing research use aggregation strategies to deal with the labels of different workers to improve the utility of crowdsourcing data. However, most of these studies are based on probabilistic graphical models, which have problems such as difficulty in setting initial parameters. This paper proposes a novel crowdsourcing method Truth Inference based on Graph Embedding (TIGE) for single-choice questions, the method draws on the idea of graph autoencoder, constructs feature vectors for each crowdsourcing task, embeds the relationship between crowdsourcing tasks and workers in graphs, then uses graph neural networks to convert crowdsourcing problems into graph node prediction problems. The feature vectors are continuously optimized in the convolutional layer to obtain the final result. Compared with the six state-of-the-art algorithms on real-world datasets, our method has significant advantages in accuracy and F1-score.

关键词： Crowdsourcing Training Graphical models Probabilistic logic Prediction algorithms Graph neural networks Robustness

来源：评论

学校读者我要写书评

暂无评论

Ensemble Classification Method Based on Truth Discovery

Ensemble Classification Method Based on Truth Discovery

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Yuxin Jin Ze Yang Ying He Xianyu Bao Gongqing Wu Key Laboratory of Knowledge Engineering with Big Data Hefei University of Technology Ministry of Education Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Shenzhen Academy of Inspection and Quarantine Shenzhen China

Classification is a hot topic in such fields as machine learning and data mining. The traditional approach of machine learning is to find a classifier closest to the real classification function, while ensemble classification is to integrate the results of base classifiers, then make an overall prediction. Compared to using a single classifier, ensemble classification can significantly improve the generalization of the learning system in most cases. However, the existing ensemble classification methods rarely consider the weight of the classifier, and there are few methods to consider updating the weights dynamically. In this paper, we are inspired by the idea of truth discovery and propose a new ensemble classification method based on the truth discovery (named ECTD). As far as we know, we are the first to apply the idea of truth discovery in the field of ensemble learning. Experimental results demonstrate that the proposed method performs well in ensemble classification.

关键词： Reliability Machine learning Classification algorithms Task analysis Buildings Computational modeling knowledge engineering

来源：评论

学校读者我要写书评

暂无评论

Gaussian Model-Based Fully Convolutional Networks for Multivariate Time Series Classification

Gaussian Model-Based Fully Convolutional Networks for Multiv...

引用

IEEE International Conference on Big knowledge (ICBK)

作者： Changyang Tai Ze Yang Huicheng Zhang Gongqing Wu Junwei Lv Xianyu Bao Key Laboratory of Knowledge Engineering with Big Data of Ministry of Education Hefei University of Technology Hefei China School of Computer Science and Information Engineering Hefei University of Technology Hefei China Shenzhen Academy of Inspection and Quarantine Shenzhen China

ISBN: (纸本)9781665438599

Multivariate time series (MTS) classification has been regarded as one of the most challenging problems in data mining due to the difficulty in modeling the correlation of variables and samples. In addition, high-dimensional MTS modeling has a large time and space consumption. This paper proposes a novel method, Gaussian Model-based Fully Convolutional Networks (GM-FCN), to improve the performance of high-dimensional MTS classification. Each original MTS is converted into multivariate Gaussian model parameters as the input of FCN. These parameters effectively capture the correlation be-tween MTS variables and significantly reduce the data scale by aligning an MTS size to its dimension. FCN is designed to learn more in-depth features of MTS based on these parameters for modeling the correlation between samples. Thus, GM-FCN can not only model the correlation between variables, but also the correlation between samples. We compare GM-FCN with nine state-of-the-art MTS classification methods, INN-ED, INN-DTW-i, INN-DTW-D, KLD-GMC, MLP, ResNet, Encoder, MCNN, and MCDCNN, on four high-dimensional public datasets, experimen-tal results show that the accuracy of G M - FCN is significantly superior to the others. Besides, the training time of GM-FCN is dozens of times faster than FCN using the original equal-length MTS data as input.

关键词： Training knowledge engineering Deep learning Correlation Conferences Time series analysis data models

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：