基于链接重要性和数据场的链接预测算法

doi:10.11772/j.issn.1001-9081.2014.08.2179

计算机应用 ›› 2014, Vol. 34 ›› Issue (8): 2179-2183.DOI: 10.11772/j.issn.1001-9081.2014.08.2179

• 第五届中国数据挖掘会议(CCDM 2014)论文 • 上一篇下一篇

基于链接重要性和数据场的链接预测算法

陈巧玉,班志杰

内蒙古大学计算机学院，呼和浩特010021

收稿日期:2014-05-04 修回日期:2014-05-13 出版日期:2014-08-01 发布日期:2014-08-10
通讯作者: 班志杰
作者简介:陈巧玉(1986-)，女，河北唐山人，硕士研究生，主要研究方向：数据挖掘、复杂网络分析；班志杰（1976-），女，内蒙古赤峰人，副教授，主要研究方向：数据挖掘、在线网络分析。
基金资助:
国家自然科学基金资助项目;内蒙古自治区高等学校科学研究项目;内蒙古自治区自然科学基金资助项目

Link prediction algorithm based on link importance and data field

CHEN Qiaoyu,BAN Zhijie

School of Computer Science, Inner Mongolia University, Huhhot Nei Mongol 010021, China

Received:2014-05-04 Revised:2014-05-13 Online:2014-08-01 Published:2014-08-10
Contact: BAN Zhijie
Supported by:
;Inner Mongolia Autonomous Region Higher Scientific Research Project

摘要/Abstract

摘要：

针对现有基于节点相似性的链接预测方法忽略了网络拓扑本身链接强度的信息，带权的拓扑路径方法中权值较难确定等缺陷，提出一种基于链接重要性和数据场的链接预测算法。首先，将所有链接边赋予不同的链接权重；其次，考虑潜在链接节点间的相互影响，对部分没有链接的节点进行链接预估计；最后，利用数据场势函数计算两节点间的相似值。在典型的网络数据进行的实验结果表明，所提方法在分类指标和推荐指标中都有很好的表现：以AUC为评价指标时，比同复杂度的局部路径（LP）算法提高了3到6个百分点；以DCG为评价指标时比LP算法提高了1.5到2.5个DCG值。算法整体上提高了预测准确性，且由于参数确定简单，复杂度又不高，在实际中易于部署。

Abstract:

The existing link prediction methods based on node similarity usually ignore the link strength of network topology and the weight value in the typological path method with weight is difficult to set. To solve these problems, a new prediction algorithm based on link importance and data field was proposed. Firstly, this method assigned different weight for each link according to the topology graph. Secondly, it took into account the interaction between potential link nodes and pre-estimated the link values for the partial nodes without links. Finally, it calculated the similarity between two nodes with data field potential function. The experimental results on some typical data sets of the real-world network show that, the proposed method has good performance with both classification index and recommended index. In comparison to the Local Path (LP) algorithm with the same complexity, the proposed algorithm raises Area Under Curve (AUC) by 3 to 6 percentages, and raises Discounted Cumulative Gain (DCG) by 1.5 to 2.5 points. On the whole, it improves the prediction accuracy. Because of its easy parameter determination and low time complexity, this new approach can be deployed simply.

中图分类号:

TP391.4

陈巧玉班志杰. 基于链接重要性和数据场的链接预测算法[J]. 计算机应用, 2014, 34(8): 2179-2183.

CHEN Qiaoyu BAN Zhijie. Link prediction algorithm based on link importance and data field[J]. Journal of Computer Applications, 2014, 34(8): 2179-2183.

参考文献

［1］SRIKANT R, AGRAWAL R. Mining quantitative association rules in large relational tables ［C］// Proceedings of the 1996 ACM SIGMOD International Conference on Management of Data. New York: ACM Press, 1996: 1-12.
［2］CATLETT J. On changing continuous attributes into ordered discrete attributes ［C］// Proceedings of the European Working Session on Learning on Machine Learning, LNCS 482. Berlin: Springer, 1991: 164-178.
［3］MEHTA S, PARTHASARATHY S, YANG H. Toward unsupervised correlation preserving discretization ［J］. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(9): 1174-1185.
［4］KERBER R. ChiMerge: discretization of numeric attributes ［C］// Proceedings of the Tenth National Conference on Artificial Intelligence. Menlo Park: AAAI Press, 1992: 123-128.
［5］LIU H, SETIONO R. Chi2: feature selection and discretization of numeric attributes ［C］// Proceedings of the Seventh IEEE International Conference on Tools with Artificial Intelligence. Washington, DC: IEEE Computer Society, 1995: 388-391.
［6］YANG Y, WEBB G I. Discretization for naive-Bayes learning: managing discretization bias and variance ［J］. Machine Learning, 2009, 74(1): 39-74.
［7］RUIZ F J, ANGULO C, AGELL N. IDD: a supervised interval distance-based method for discretization ［J］. IEEE Transactions on Knowledge and Data Engineering, 2008, 20(9): 1230-1238.
［8］DOUGHERTY J, KOHAVI R, SAHAMI M. Supervised and unsupervised discretization of continuous features ［C］// Proceedings of the Twelfth International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers, 1995: 194-202.
［9］LI G. An unsupervised discretization algorithm based on mixture probabilistic model ［J］. Chinese Journal of Computers, 2002, 25(2): 158-164.(李刚.基于混合概率模型的无监督离散化算法［J］.计算机学报,2002,25(2):158-164.)
［10］KURGAN L A, CIOS K J. CAIM discretization algorithm ［J］. IEEE Transactions on Knowledge and Data Engineering, 2004, 16(2): 145-153.
［11］TSAI C J, LEE C I, YANG W. A discretization algorithm based on class-attribute contingency coefficient ［J］. Information Sciences, 2008, 178(3): 714-731.
［12］SCHMIDBERGER G, FRANK E. Unsupervised discretization u-sing tree-based density estimation ［C］// PKDD 2005: Proceedings of the 9th European Conference on Principles and Practice of Knowledge Discovery in Databases, LNCS 3721. Berlin: Springer, 2005: 240-251.
［13］BIBA M, ESPOSITO F, FERILLI S, et al. Unsupervised discretization using kernel density estimation ［C］// Proceedings of the 20th International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann Publishers, 2007: 697-701.
［14］BORIAH S, CHANDOLA V, KUMAR V. Similarity measures for categorical data: a comparative evaluation ［C］// Proceedings of the 8th SIAM International Conference on Data Mining. Philadelphia: SIAM, 2008: 243-254.
［15］ZHANG S, WONG H S, SHEN Y. Generalized adjusted rand indices for cluster ensembles ［J］. Pattern Recognition, 2012, 45(6): 2214-2226.
［16］NG A Y, JORDAN M I, WEISS Y. On spectral clustering: analysis and an algorithm ［C］// Advances in Neural Information Processing Systems. Cambridge: MIT Press, 2002: 849-856.
［17］THEODORIDIS S, KOUTROUMBAS K. Pattern recognition ［M］. Waltham: Academic Press, 2003.

[1]	罗圣钦陈金怡李洪均. 基于注意力机制的多尺度残差UNet实现乳腺癌灶分割[J]. 计算机应用, 0, (): 0-0.
[2]	杨鼎康黄帅王顺利翟鹏李一丹张立华. 基于对抗生成网络和网络集成的面部表情识别方法EE-GAN[J]. 计算机应用, 0, (): 0-0.
[3]	秦庭威赵鹏程秦品乐曾建朝柴锐黄永琦. 基于残差注意力机制的点云配准算法[J]. 计算机应用, 0, (): 0-0.
[4]	鲁永帅唐英杰马鑫然. 基于深度特征融合的无纺布低对比度浆丝缺陷检测方法[J]. 计算机应用, 0, (): 0-0.
[5]	裴仪瑶, 郭会明, 张丹普, 陈文博. 基于定位不确定性的鲁棒3D目标检测方法[J]. 计算机应用, 2021, 41(10): 2979-2984.
[6]	许学斌, 张佳达, 刘伟, 路龙宾, 赵雨晴. 融合空间和通道特征的高精度乳腺癌分类方法[J]. 计算机应用, 2021, 41(10): 3025-3032.
[7]	汪虹余, 张彧, 杨恒, 穆楠. 基于蚁群优化算法的弱光图像显著性目标检测[J]. 计算机应用, 2021, 41(10): 2970-2978.
[8]	魏淳武, 赵涓涓, 唐笑先, 强彦. 基于多时期蒸馏网络的随访数据知识提取方法[J]. 计算机应用, 2021, 41(10): 2871-2878.
[9]	韩建栋, 李晓宇. 基于多尺度特征融合的行人重识别方法[J]. 计算机应用, 2021, 41(10): 2991-2996.
[10]	张凯悦, 张鸿. 基于注意力机制网络的航运监控图像识别模型[J]. 计算机应用, 2021, 41(10): 3010-3016.
[11]	李自强, 王正勇, 陈洪刚, 李林怡, 何小海. 基于外观和动作特征双预测模型的视频异常行为检测[J]. 计算机应用, 2021, 41(10): 2997-3003.
[12]	高世伟, 张长柱, 王祝萍. 基于可分离金字塔的轻量级实时语义分割算法[J]. 计算机应用, 2021, 41(10): 2937-2944.
[13]	卓禹心, 韩素雅, 张榆锋, 李支尧, 董毅峰. 基于超声谐波包络Nakagami参数图像的微波消融区域自动分割方法[J]. 计算机应用, 2021, 41(10): 3089-3096.
[14]	李敬虎邢前国郑向阳李琳王丽丽. 基于深度学习的无人机影像夜光藻赤潮提取方法[J]. 计算机应用, 0, (): 0-0.
[15]	任炜白鹤翔. 基于全局与局部的多标签图像分类方法[J]. 计算机应用, 0, (): 0-0.

基于链接重要性和数据场的链接预测算法

Link prediction algorithm based on link importance and data field

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics