18 results on '"Graph"'
Search Results
2. 面向多方面的双通道知识增强图卷积网络模型.
- Author
-
陈景景, 韩 虎, and 徐学锋
- Abstract
Aspect-based sentiment analysis is a fine-grained sentiment analysis task, which aims to align aspects with the corresponding emotion words for aspect specific emotion polarity reasoning. In recent years, the graph neural network sentiment classification method based on syntactic dependent information has become a research hotspot in this field. However, due to the flexibility of comment sentences in content expression and syntactic structure, the modeling method using only syntactic dependent information still has some shortcomings. In order to enhance the comment sentences by affective knowledge and structural semantic information, a convolutional network model(DualSyn-GCN) of two channel knowledge enhancement graph is proposed. On one hand, the syntactic dependency adjacency matrix is enhanced according to the implicit relationship between aspect and aspect as well as aspect and context. On the other hand, the emotional dependency of aspect is learned from external emotional knowledge, and then the two different enhanced representations are fused to realize the sharing and complementarity between different representations. The experimental results show that, compared with the classical aspect based graph convolutional network model (ASGCN), this model improves the accuracy and MF1 value on LAP14 data set by 2.34% and 3.26% respectively. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
3. Cross-Subject ERP Detection Based on Graph and Dual Attention Mechanism.
- Author
-
XIANG Xiaojia, LAN Zhen, YAN Chao, LI Zixing, TANG Dengqing, and ZHOU Han
- Subjects
CONVOLUTIONAL neural networks ,RECURRENT neural networks ,EVOKED potentials (Electrophysiology) ,ELECTROENCEPHALOGRAPHY - Abstract
In order to improve the detection accuracy of event-related potential (ERP) in subject-independent scenarios, a convolutional recurrent neural network model based on graph embedding and dual attention mechanisms is proposed. The model uses a graph to represent the spatial information in electroencephalogram (EEG) signals, and uses the cascade framework of convolutional neural network (CNN) and long short-term memory network (LSTM) as the basic framework. By embedding dual attention mechanisms (i.e., selective kernel convolution and self-attention mechanism), it can fully extract the temporal and spatial features of EEG signals of different subjects, so as to improve the ERP detection accuracy in subject-independent scenarios. A large number of experiments carried out on the benchmark dataset based on rapid serial visual presentation paradigm demonstrate that the proposed method has significant superiority over 7 existing ERP detection methods in subject-independent scenarios. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF
4. 基于句法和全文信息增强的中文事件检测方法.
- Author
-
王 红 and 吴浩正
- Abstract
Copyright of Journal of Data Acquisition & Processing / Shu Ju Cai Ji Yu Chu Li is the property of Editorial Department of Journal of Nanjing University of Aeronautics & Astronautics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2022
- Full Text
- View/download PDF
5. 两类图的 Fractional 控制数.
- Author
-
李广, 徐保根, and 张君霞
- Abstract
Copyright of Journal of Guangxi Normal University - Natural Science Edition is the property of Gai Kan Bian Wei Hui and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2021
- Full Text
- View/download PDF
6. 两类乘积图的符号控制数.
- Author
-
徐保根, 张君霞, and 李 广
- Abstract
Copyright of Journal of Henan University of Science & Technology, Natural Science is the property of Editorial Office of Journal of Henan University of Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2020
- Full Text
- View/download PDF
7. 滚筒输送线路径规划方法研究.
- Author
-
薛志强 and 游有鹏
- Abstract
Aiming at the problem of path planning in roller conveying line systemꎬ the environment model of conveyor lineꎬ path planning al ̄ gorithm and optimization were researched. By extracting the entranceꎬ exitꎬ and transfer of the three modules as the node and taking the de ̄ livery routes between the nodes as the weighted routesꎬ a method of calculating the length of the links between the nodes by the conveyors e ̄ quivalent was proposedꎬ then a streamlined graph structure model was built. By using Dijkstra algorithm and improved ant colony algorithm to solve the small ̄scale and large ̄scale path planning problems respectivelyꎬ and according to the two constraint cases about the necessary pass ̄ ing nodes and the congestion road sections in the applied transmission lineꎬ the path planning algorithm of transmission line which combines Dijkstra algorithm and improved ant colony algorithm was proposedꎬ which two methods were blent with segmented splicing and path length weighting. The results indicate that the method provides an effective solution to the path planning of roller conveying line with different scales under two kinds of constraints. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF
8. 圆柱度误差可视化的理论研究.
- Author
-
陈立杰, 张明耀, and 康亚栋
- Abstract
The basic theory of producing cylindricity error graph was studied, and a mathematical model for visualization of cylindricity error was established. The theory of computer graphics and three-dimensional wireframe models were used to realize the display of extracted integral feature, associated integral feature and associated derived feature of the measured cylinders. Axonometric projection transformation was adopted to fulfill the realization of basic geometric shapes. The space hidden method was used to solve the problem of geometrical graph hidden. According to the standard definition of cylindricity error, the mathematical model for cylindricity error graphic display was established. On the basis of theoretical study, software was developed to realize visualization of cylindricity error on the platform of LabWindows/CVI. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
9. 随机图的均匀边染色算法.
- Author
-
代素敏, 胡腾云, 尹波, and 李敬文
- Abstract
An equitable edge coloring of graph G is that adjacent edges of a graph have different color and the sizes of its color classes differ by at most 1. The minimum number of colors is called the equitable edge chromatic number. This paper proposed a new heuristic intelligent algorithm with the study of minimum equitable edge chromatic number. According to the conditions of edge coloring, it created object function, exchanged the color complement matrix of coloring matrix iteratively to find the optimum solution. It gave detailed algorithm, with lots of testing and analysis cases.For a graph with fixed number of vertices, this algorithm can find the minimum equitable edge chromatic number efficiently, while the time complexity of this algorithm is less than O(n3). [ABSTRACT FROM AUTHOR]
- Published
- 2016
- Full Text
- View/download PDF
10. 基于图的情境离群点检测方法.
- Author
-
李涛, 张芸, and 黄志宏
- Abstract
Copyright of Transactions of Beijing Institute of Technology is the property of Beijing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
- Published
- 2016
- Full Text
- View/download PDF
11. Random walk based node pair sampling algorithm in large graph.
- Author
-
WU Chun-qiong and YE Dong-yi
- Abstract
Node pair sampling in social networks is useful for friend recommendation and interest targeting. While the topology of the whole network is incomplete, or the cost of random generation of a user is very large, the performances of traditional uniform vertex sampling methods decrease quickly. So, this paper proposed a random walk based node pair sampling algorithm in large graph. Firstly, it analyzed systematically the problem of node pair sampling in social networks, and gave definitions of node pair in different hops. Secondly, it transformed the social network into an equivalent graph, whose nodes were edges in original graph, and whose edges contained two node having the same vertex. Finally, it applied random walk on the new graph and proposed a neighbor random walk sampling algorithm. The experiments show that, the proposed algorithm has less error, better performance, and is obviously better than uniform vertex sampling related methods. [ABSTRACT FROM AUTHOR]
- Published
- 2015
- Full Text
- View/download PDF
12. Random walls based approximate approach for topic search.
- Author
-
ZHANG Ling-xiao, LU Xin-hua, and LIU Ke-cheng
- Abstract
Hyperlink induced topic search (HITS) is one of the most authoritative and widely used personalized ranking algorithm on graphs. The HITS algorithm ranks nodes on graphs according to power iteration, and has high complexity of computation. In order to improve computation time, this paper analyzed and modeled the HITS algorithm with the random walls approach, and proposed Monte Carlo based approximation computation algorithms for the HITS ranking. Theoretical analysis and experiments show that the proposed random walk based approach of HITS ranking is not only efficient, but also has higher accuracy, and is significantly better than related works. [ABSTRACT FROM AUTHOR]
- Published
- 2015
- Full Text
- View/download PDF
13. A CSGC improved algorithm of spectrum allocation.
- Author
-
TENG Zhijun and LI Ke
- Abstract
To solve the problem that the spectrum allocation algorithm based on graph theory coloring algorithm has not fully considered the actual bandwidth needs of users, this paper proposes a spectrum allocation based on user priority algorithm improved CSGC and the original algorithm. The algorithm introduces two time factors that are respectively called idle spectrum and user demand, by setting the user priority, the function can meet the needs of users during the second spectrum allocation. Simulation results show that the algorithm not only retains the performance of the original algorithm CSGC, but also greatly improves the spectrum utilization. [ABSTRACT FROM AUTHOR]
- Published
- 2014
14. Review on superpixel segmentation algorithms.
- Author
-
WANG Chun-yao, CHEN Jun-zhou, and LI Wei
- Subjects
- *
PIXELS , *IMAGE segmentation , *COMPUTER algorithms , *IMAGE processing , *COMPUTATIONAL complexity , *GRAPH theory - Abstract
Superpixel can capture redundancy of the image and reduce the complexity of subsequent processing tasks. These advantages make it receive more and more attentions from researchers at home and abroad. This paper first analyzed the development of the superpixel segmentation, and summarized the state-of-the-art superpixel segmentation algorithms in the view of graph-based and gradient-ascent-based methods. Then, it compared several superpixel segmentation algorithms by experiments, and illustrated their strengths and weaknesses respectively. At last, it introduced the latest applications of superpixel segmentation techniques with prospects. [ABSTRACT FROM AUTHOR]
- Published
- 2014
- Full Text
- View/download PDF
15. Graph-based near-duplicate video subsequence matching algorithm.
- Author
-
LIU Hong
- Subjects
- *
GRAPH theory , *COMPUTER algorithms , *IMAGE analysis , *COMPUTER systems , *COMPUTER programming , *COMPUTER engineering - Abstract
In order to improve the effectiveness and efficiency of near-duplicate video detection, this paper proposed the graph-based video subsequence matching algorithm. The algorithm constructed a matching results graph from the similarity search results based on the key frame features, and then converted the problem of near-duplicate video detection into the problem of finding the longest path in the matching results graph. The method has three main advantages: a) Graph-based method could find the best matching sequence in many messy match results, which effectively excluded false "high similarity" noise and compensated the limited description of image low level visual features. b)The graph-based method took fully into account the spatiotemporal characteristic of video sequence, and had high location accuracy. c)The graph-based sequence matching method could automatically detect the discrete paths in the matching result graph. Thus, it could detect more than one near-duplicate video. The proposed algorithm not only improves the detection accuracy, but improves the efficiency of detection, achieves good practical effect. [ABSTRACT FROM AUTHOR]
- Published
- 2013
- Full Text
- View/download PDF
16. Research on the Risk Awareness of Consumers in Fitness Clubs.
- Author
-
Shi-yan and Chert Yu-pu
- Published
- 2008
17. A class of Weierstrass function with Box dimension 2 of graphs.
- Author
-
LIU Chuntai
- Subjects
MATHEMATICAL functions ,MATHEMATICAL expansion ,MATHEMATICAL sequences ,INTEGERS ,MATHEMATICS - Abstract
The Box dimension of a class of Weierstrass function ... is discussed. Under the help of the λ-ary expansions, the Box dimension of Γ(f) is 2 when the sequence satisfies ..., where 1<λ is an integer and {a
n } is a summable sequence with positive terms. [ABSTRACT FROM AUTHOR]- Published
- 2013
18. Crossing numbers of cartesian products of stars with 5-vertex graphs.
- Author
-
HE Xiao-nian and HUANG Yuan-qiu
- Subjects
GRAPH theory ,HOMEOMORPHISMS ,MANIFOLDS (Mathematics) ,TOPOLOGICAL spaces ,TRANSFORMATION groups ,COMBINATORICS - Abstract
Garey and Johxon have proved that the problem to determine the crossing number of graphs is NP-complete. Because of its difficulty, we only know the crossing number of some classes of special graphs. The crossing number of Cartesian products of paths and cycles with 5-vertex graphs are mostly known, but only few cartesian products of 5-vertex graphs with star K
1, n are known. In this paper, we extend these results, and determine the crossing numbers of Cartesian products of two 5-vertex graphs with star K1, n . [ABSTRACT FROM AUTHOR]- Published
- 2009
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.