[1]杜阳阳,李华康,李涛.基于Node2vec 的改进算法的研究[J].计算机技术与发展,2018,28(07):6-10.[doi:10.3969/ j. issn.1673-629X.2018.07.002]
　DU Yang-yang,LI Hua-kang,LI Tao.Research on Improved Algorithm Based on Node2vec[J].,2018,28(07):6-10.[doi:10.3969/ j. issn.1673-629X.2018.07.002]
点击复制

基于Node2vec 的改进算法的研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年07期
页码:
6-10
栏目:
智能、算法、系统工程
出版日期:
2018-07-10

文章信息/Info

Title:
Research on Improved Algorithm Based on Node2vec
文章编号:
1673-629X(2018)07-0006-05
作者:
杜阳阳; 李华康; 李涛
南京邮电大学 计算机学院,江苏 南京 210046
Author(s):
DU Yang-yang; LI Hua-kang; LI Tao
School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210046,China
关键词:
数据挖掘; 随机游走; 节点表示; 多标签分类
Keywords:
data mining; random walk; node representation; multi-label classification
分类号:
TP181
DOI:
10.3969/ j. issn.1673-629X.2018.07.002
文献标志码:
A
摘要:
针对图节点的多标签分类任务,在 Node2vec 算法的基础上进行了改进,在原来随机游走的基础上加上部分标签信息的指导,然后对节点进行向量表示。 算法首先根据每一个图节点及其邻居节点的标签信息和事先设定好的游走参数的值,计算当前节点的邻居节点被游走的概率;然后由概率值和其他设定好的游走的参数开始游走,得到若干条路径;之后再调用 Word2vec 方法对若干条游走路径进行训练,将每个图节点表示成向量。 最后,通过使用逻辑分类模型对节点的特征表示进行多标签分类来验证算法的有效性。 实验结果证明,通过使用标签信息的指导,多标签分类的准确率有明显提升。
Abstract:
In view of the multi-label classification task of graph nodes,we carry on the improvement on the basis of theNode2vec algorithm,adding some label information to the original random walk,and then representing the nodes with vectors. The algorithm firstly calculates the probability that the neighbor node of the current node is traveling according to the label information of each graph node and its neighbor node and the value of the walking parameter set in advance. Then,the probability and other parameters begin to walk,and get a number of paths. After that,the Word2vec method is called to train a number of walking path,each node will be expressed as a vector. Finally,the validity of the algorithm is verified by using the logical classification model to label the feature representation of the nodes. The experiment shows that the accuracy of multi-label classification is remarkably improved by using the guidance of label information.

相似文献/References:

[1]项响琴 汪彩梅.基于聚类高维空间算法的离群数据挖掘技术研究[J].计算机技术与发展,2010,(01):120.
　XIANG Xiang-qin,WANG Cai-mei.Study of Outlier Data Mining Based on CLIQUE Algorithm[J].,2010,(07):120.
[2]李雷 丁亚丽 罗红旗.基于规则约束制导的入侵检测研究[J].计算机技术与发展,2010,(03):143.
　LI Lei,DING Ya-li,LUO Hong-qi.Intrusion Detection Technology Research Based on Homing - Constraint Rule[J].,2010,(07):143.
[3]吉同路 柏永飞 王立松.住宅与房地产电子政务中数据挖掘的应用研究[J].计算机技术与发展,2010,(01):235.
　JI Tong-lu,BAI Yong-fei,WANG Li-song.Study and Application of Data Mining in E-government of House and Real Estate Industry[J].,2010,(07):235.
[4]杨静 张楠男 李建 刘延明 梁美红.决策树算法的研究与应用[J].计算机技术与发展,2010,(02):114.
　YANG Jing,ZHANG Nan-nan,LI Jian,et al.Research and Application of Decision Tree Algorithm[J].,2010,(07):114.
[5]赵裕啸 倪志伟 王园园 伍章俊.SQL Server 2005数据挖掘技术在证券客户忠诚度的应用[J].计算机技术与发展,2010,(02):229.
　ZHAO Yu-xiao,NI Zhi-wei,WANG Yuan-yuan,et al.Application of Data Mining Technology of SQL Server 2005 in Customer Loyalty Model in Securities Industry[J].,2010,(07):229.
[6]张笑达 徐立臻.一种改进的基于矩阵的频繁项集挖掘算法[J].计算机技术与发展,2010,(04):93.
　ZHANG Xiao-da,XU Li-zhen.An Advanced Frequent Itemsets Mining Algorithm Based on Matrix[J].,2010,(07):93.
[7]王爱平 王占凤 陶嗣干 燕飞飞.数据挖掘中常用关联规则挖掘算法[J].计算机技术与发展,2010,(04):105.
　WANG Ai-ping,WANG Zhan-feng,TAO Si-gan,et al.Common Algorithms of Association Rules Mining in Data Mining[J].,2010,(07):105.
[8]张广路 雷景生 吴兴惠.一种改进的Apriori关联规则挖掘算法（英文）[J].计算机技术与发展,2010,(06):84.
　ZHANG Guang-lu,LEI Jing-sheng,WU Xing-hui.An Improved Apriori Algorithm for Mining Association Rules[J].,2010,(07):84.
[9]吴楠 胡学钢.基于聚类分区的序列模式挖掘算法研究[J].计算机技术与发展,2010,(06):109.
　WU Nan,HU Xue-gang.Research on Clustering Partition-Based Approach of Sequential Pattern Mining[J].,2010,(07):109.
[10]吴青 傅秀芬.水平分布数据库的正负关联规则挖掘[J].计算机技术与发展,2010,(06):113.
　WU Qing,FU Xiu-fen.Positive and Negative Association Rules Mining on Horizontally Partitioned Database[J].,2010,(07):113.

更新日期/Last Update: 2018-08-23