检索结果-内蒙古大学图书馆

8th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2019

作者： Ma, Fu-Yuan Chen, Wen-Qi Xiao, Min-Hao Wang, Xin Wang, Ying College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbol Computation and Knowledge Engineering Jilin University Ministry of Education Changchun130012 China Changchun Institute of Technology Changchun130012 China College of Software Jilin University Changchun130012 China

ISBN: (纸本)9783030322359

With the development of information society, Recommendation System has been an import tool to help users filter information and create more economic value for enterprises. However, it is difficult for traditional recommendation systems to interpret recommendation results. In order to improve users’ trust in recommendation results, interpretable recommendation models have attracted more and more attention. In this paper, we present the Explanation Chains Model based on the Fine-grained Data (F-ECM) to enhance the effectiveness of recommendation while achieving the interpretability of recommendation. First, we generate parsing trees from user comments and extract three key sentence structure information (i.e., aspects, features and sentiment tendency) from those generated parsing tree. The fine-grained similarity is computed based on the aspects and features of the products to be recommended, and users’ product satisfaction is predicted by combining sentiment tendency. Then the recommendation chain will be constructed according to the satisfaction degree in the recommendation list. Finally, we calculate recommendation chain scores of all the items to be recommended to the target user, generate the recommendation results and personal explanation for the user by the recommendation chains. Experiments in the Amazon data set show that the Explanation Chains Model based on the Fine-grained Data achieve better interpretability and performance of product recommendation systems. © 2019, Springer Nature Switzerland AG.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Incorporating Surprisingly Popular Algorithm and Euclidean Distance-based Adaptive Topology into PSO

arXiv

引用

arXiv 2021年

作者： Wu, Xuan Han, Jizong Wang, Di Gao, Pengyue Cui, Quanlong Chen, Liang Liang, Yanchun Huang, Han Lee, Heow Pueh Miao, Chunyan Zhou, You Wu, Chunguo Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University Changchun130012 China Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly Nanyang Technological University 639798 Singapore State Key Laboratory of Superhard Materials International Center for Computational Method and Software College of Physics Jilin University Changchun130012 China Commercial Quality and Efficiency Department Baidu Inc Beijing100085 China Department of Computer Science Shantou University Shantou515063 China School of Computer Science Zhuhai College of Science and Technology Zhuhai519041 China School of Software Engineering The South China University of Technology Guangzhou510641 China Department of Mechanical Engineering National University of Singapore 117575 Singapore School of Computer Science and Engineering Nanyang Technological University 639798 Singapore College of Software Jilin University Changchun130012 China

While many Particle Swarm Optimization (PSO) algorithms only use fitness to assess the performance of particles, in this work, we adopt Surprisingly Popular Algorithm (SPA) as a complementary metric in addition to fitness. Consequently, particles that are not widely known also have the opportunity to be selected as the learning exemplars. In addition, we propose a Euclidean distance-based adaptive topology to cooperate with SPA, where each particle only connects to k number of particles with the shortest Euclidean distance during each iteration. We also introduce the adaptive topology into heterogeneous populations to better solve large-scale problems. Specifically, the exploration sub-population better preserves the diversity of the population while the exploitation sub-population achieves fast convergence. Therefore, large-scale problems can be solved in a collaborative manner to elevate the overall performance. To evaluate the performance of our method, we conduct extensive experiments on various optimization problems, including three benchmark suites and two real-world optimization problems. The results demonstrate that our Euclidean distance-based adaptive topology outperforms the other widely adopted topologies and further suggest that our method performs significantly better than state-of-the-art PSO variants on small, medium, and large-scale problems. Copyright © 2021, The Authors. All rights reserved.

关键词： Small-world networks

来源：评论

学校读者我要写书评

暂无评论

Deep Bayesian optimization on attributed graphs

arXiv

引用

arXiv 2019年

作者： Cui, Jiaxu Yang, Bo Hu, Xia College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation Knowledge Engineering of Ministry of Education China Department of Computer Science and Engineering Texas A&M University College Station United States

Attributed graphs, which contain rich contextual features beyond just network structure, are ubiquitous and have been observed to benefit various network analytics applications. Graph structure optimization, aiming to find the optimal graphs in terms of some specific measures, has become an effective computational tool in complex network analysis. However, traditional model-free methods suffer from the expensive computational cost of evaluating graphs;existing vectorial Bayesian optimization methods cannot be directly applied to attributed graphs and have the scalability issue due to the use of Gaussian processes (GPs). To bridge the gap, in this paper, we propose a novel scalable Deep Graph Bayesian Optimization (DGBO) method on attributed graphs. The proposed DGBO prevents the cubical complexity of the GPs by adopting a deep graph neural network to surrogate black-box functions, and can scale linearly with the number of observations. Intensive experiments are conducted on both artificial and real-world problems, including molecular discovery and urban road network design, and demonstrate the effectiveness of the DGBO compared with the state-of-the-art. Copyright © 2019, The Authors. All rights reserved.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Automatic localization of macular area based on structure label transfer

引用

International Journal of Ophthalmology(English edition) 2018年第3期11卷 422-428页

作者： Xiao-Xin Guo Qun Li Chao Sun Yi-Nan LU Key Laboratory of Symbol Computation and Knowledge Engineering of Ministry of Education Changchun130012 Jilin Province China College of Computer Science and Technology Jilin University Changchun 130012 Jilin Province China

AIM： To explore feasibility and practicability of macula localization independent of macular morphological features. METHODS： A novel method was proposed to identify macula in fundus images by using structure label transfer. Its main idea was to match a processed image with the candidate images with known structures, and then transfer the structure label representing the macular to the processed image as a result of macula localization. In this way, macula localization couldn＇t be influenced by lesion or other interference any more. RESULTS： The average success rate in four datasets was 98.18%. For accuracy, the average error distance in four datasets was 0.151 optic disc diameter （ODD）. Even for severe lesion images, the proposed method can still maintain high success rate and high accuracy, e.g., 95.65% and 0.124 ODD in the case of STARE dataset, respectively, which indicated that the proposed method was highly robust and stable in the complicated situations. CONCLUSION： The proposed method can avoid the interference of lesion to macular morphological features in macula localization, and can locate macula with high accuracy and robustness, verifying its feasibility.

关键词： fundus image optic disc macula structurelabel transfer

来源：评论

学校读者我要写书评

暂无评论

Group-attention Based Neural Machine Translation

引用

IOP Conference Series: Materials Science and engineering 2020年第2期782卷

作者： Renyu Hu Hao Xu Yang Xiao Chenjun Wu Haiyang Jia College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Machine translation is a classic problem in natural language process (NLP). Recent years, the encoder and decoder through an attention mechanism has become a trend. Google proposed a new simple network architecture, the Transformer using attention mechanisms only rather than CNN or RNN in 2017. However, it may lose some important information (e.g., grammatical component, etc) when using attention mechanism for whole *** propose a new brand model based on transformer using Group attention layers and group position embedding. It absorbs the features of Group-CNN combines the algorithm in computer vision (CV) and NLP. The model not only pays more attention to the ingredients (e.g., subject, predicate and adverbial, etc), but also enhances the connection of phrase. It outperforms SofA Transformer in using more syntactic information.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Heuristic black-box adversarial attacks on video recognition models

arXiv

引用

arXiv 2019年

作者： Wei, Zhipeng Chen, Jingjing Wei, Xingxing Jiang, Linxi Chua, Tat-Seng Zhou, Fengfeng Jiang, Yu-Gang Jilin University Fudan University Beihang University National University of Singapore Health Informatics Lab College of Computer Science and Technology Key Laboratory of Symbolic Computation Knowledge Engineering of Ministry of Education Jilin University Changchun Jilin130012 China

We study the problem of attacking video recognition models in the black-box setting, where the model information is unknown and the adversary can only make queries to detect the predicted top-1 class and its probability. Compared with the black-box attack on images, attacking videos is more challenging as the computation cost for searching the adversarial perturbations on a video is much higher due to its high dimensionality. To overcome this challenge, we propose a heuristic black-box attack model that generates adversarial perturbations only on the selected frames and regions. More specifically, a heuristic-based algorithm is proposed to measure the importance of each frame in the video towards generating the adversarial examples. Based on the frames’ importance, the proposed algorithm heuristically searches a subset of frames where the generated adversarial example has strong adversarial attack ability while keeps the perturbations lower than the given bound. Besides, to further boost the attack efficiency, we propose to generate the perturbations only on the salient regions of the selected frames. In this way, the generated perturbations are sparse in both temporal and spatial domains. Experimental results of attacking two mainstream video recognition methods on the UCF-101 dataset and the HMDB-51 dataset demonstrate that the proposed heuristic black-box adversarial attack method can significantly reduce the computation cost and lead to more than 28% reduction in query numbers for the untargeted attack on both datasets. Copyright © 2019, The Authors. All rights reserved.

关键词： Heuristic methods

来源：评论

学校读者我要写书评

暂无评论

A Task Allocation Method for Stream Processing with Recovery Latency Constraint

引用

Journal of Computer Science & Technology 2018年第6期33卷 1125-1139页

作者： Hong-Liang Li Jie Wu Zhen Jiang Xiang Li Xiao-Hui Wei College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of the Ministry of Education Changchun 130012 China Department of Computer and Information Sciences Temple University Philadelphia PA 19122 U.S.A. Department of Computer Science West Chester University of Pennsylvania West Chester PA 19383 U.S.A.

Stream processing applications continuously process large amounts of online streaming data in real time or nearreal time. They have strict latency constraints. However, the continuous processing makes them vulnerable to any failures,and the recoveries may slow down the entire processing pipeline and break latency constraints. The upstream backupscheme is one of the most widely applied fault-tolerant schemes for stream processing systems. It introduces complexbackup dependencies to tasks, which increases the difficulty of controlling recovery latencies. Moreover, when dependenttasks are located on the same processor, they fail at the same time in processor-level failures, bringing extra recovery latencies that increase the impacts of failures. This paper studies the relationship between the task allocation and therecovery latency of a stream processing application. We present a correlated failure effect model to describe the recoverylatency of a stream topology in processor-level failures under a task allocation plan. We introduce a recovery-latency awaretask allocation problem （RTAP） that seeks task allocation plans for stream topologies that will achieve guaranteed recoverylatencies. We discuss the difference between RTAP and classic task allocation problems and present a heuristic algorithmwith a computational complexity of O（n log2 n） to solve the problem. Extensive experiments were conducted to verify thecorrectness and effectiveness of our approach. It improves the resource usage by 15%-20% on average.

关键词： stream processing task allocation fault-tolerance upstream backup, recovery latency

来源：评论

学校读者我要写书评

暂无评论

Learning Entity and Relation Embeddings with Entity Description for knowledge Graph Completion

Learning Entity and Relation Embeddings with Entity Descript...

引用

2018 2nd International Conference on Artificial Intelligence: Technologies and Applications（ICAITA2018）

作者： Shaozhi Dai Yanchun Liang Shuyan Liu Ying Wang Wenle Shao Xixun Lin Xiaoyue Feng Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University Zhuhai Laboratory of Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Zhuhai College of Jilin University

With the growth of existing knowledge graph, the completion of knowledge graph has become a crucial problem. In this paper, we propose a novel model based on descriptionembodied knowledge representation learning framework, which is able to take advantages of both fact triples and entity description. Specifically, the relation projection is combined with description-embodied representation learning to learn entity and relation embeddings. Convolutional neural network and Trans R are adopted to get the description-based and structure-based representation of entity and relation, respectively. We employ FB15 K dataset generated from a large knowledge graph freebase, to evaluate the performances of the proposed model. Experimental results show that our proposed model greatly outperforms other existing baseline models.

关键词： knowledge graph completion entity description natural language processing

来源：评论

学校读者我要写书评

暂无评论

RiWalk: Fast Structural Node Embedding via Role Identification

RiWalk: Fast Structural Node Embedding via Role Identificati...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Xuewei Ma Geng Qin Zhiyang Qiu Mingxin Zheng Zhe Wang College of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbol Computation and Knowledge Engineering Ministry of Education Jilin University Changchun China

Nodes performing different functions in a network have different roles, and these roles can be gleaned from the structure of the network. Learning latent representations for the roles of nodes helps to understand the network and to transfer knowledge across networks. However, most existing structural embedding approaches suffer from high computation and space cost or rely on heuristic feature engineering. Here we propose RiWalk, a flexible paradigm for learning structural node representations as it decouples the structural embedding problem into a role identification procedure and a network embedding procedure. Through role identification, rooted kernels with structural dependencies kept are built to better integrate network embedding methods. To demonstrate the effectiveness of RiWalk, we develop two different role identification methods named RiWalk-SP and RiWalk-WL respectively and employ random walk based network embedding methods. Experiments on within-network classification tasks show that our proposed algorithms achieve comparable performance with other baselines while being an order of magnitude more efficient. Besides, we also conduct across-network role classification tasks. The results show potential of structural embeddings in transfer learning. RiWalk is also scalable, making it capable of capturing structural roles in massive networks.

关键词： Kernel Task analysis knowledge engineering Machine learning Bridges Proteins Network topology

来源：评论

学校读者我要写书评

暂无评论

Ordered proposition fusion based on consistency and uncertainty measurements

引用

Science China(Information Sciences) 2017年第8期60卷 117-135页

作者： Dayou LIU Yungang ZHU Ni NI Jie LIU College of Computer Science and Technology Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Beijing Entry-Exit Inspection and Quarantine Bureau

The fusion of ordered propositions is an important and widespread problem in artificial intelligence,but existing fusion methods have difficulty handling the fusion of ordered propositions. In this paper, we propose a solution based on consistency and uncertainty measurements. The main contributions of this paper are as follows. First, we propose the concept of convexity degree, mean, and center of basic support function to comprehensively describe the basic support function of ordered propositions. Second, we introduce entropy as a measure of uncertainty in the basic support function of ordered propositions. Third, we generalize the indeterminacy of the basic support function and propose a novel method to measure the consistency between two basic support functions. Finally, based on the above researches, we propose a novel algorithm for fusing ordered propositions. Theoretical analysis and experimental results demonstrate that the proposed method outperforms state-of-the-art methods.

关键词： uncertainty processing ordered proposition evidence theory dempster rule of combination information fusion

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：