An Outlier Detection Method Based on Ranking and Clustering in Bi-typed Heterogeneous Network

PENG Tao; YANG Ni-ya; XU Yuan-bo; WANG Bing-bing; LIU Lu

doi:10.3969/j.issn.0372-2112.2018.02.004

您当前的位置：

首页 >

文章列表页 >

An Outlier Detection Method Based on Ranking and Clustering in Bi-typed Heterogeneous Network

更新时间：2025-07-16

- An Outlier Detection Method Based on Ranking and Clustering in Bi-typed Heterogeneous Network
- Acta Electronica Sinica Vol. 46, Issue 2, Pages: 281-288(2018)
- 作者机构：
  
  1. 吉林大学计算机科学与技术学院,吉林,长春,130012
  2. 符号计算与知识工程教育部重点实验室(吉林大学),吉林,长春,130012
  3. 吉林大学计算机科学与技术学院,吉林,长春,130012
  4. 符号计算与知识工程教育部重点实验室(吉林大学),吉林,长春,130012
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.60903098);Postgraduate Innovation Fund of Jilin University (No.2016183, No.2016184)
- DOI：10.3969/j.issn.0372-2112.2018.02.004
  CLC： TP391
- Published Online：25 February 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
PENG Tao, YANG Ni-ya, XU Yuan-bo, et al. An Outlier Detection Method Based on Ranking and Clustering in Bi-typed Heterogeneous Network[J]. Acta Electronica Sinica, 2018, 46(2): 281-288.
DOI：

PENG Tao, YANG Ni-ya, XU Yuan-bo, et al. An Outlier Detection Method Based on Ranking and Clustering in Bi-typed Heterogeneous Network[J]. Acta Electronica Sinica, 2018, 46(2): 281-288. DOI： 10.3969/j.issn.0372-2112.2018.02.004.

摘要

挖掘隐藏在网络中不同于正常数据对象的离群点是数据挖掘的重要任务之一.目前，针对双类型异质信息网络离群点检测的研究工作相对较少，原本适用于同质网络的离群点检测方法将很难适用于双类型异质网络.为此，提出了异质信息网络中基于排序和聚类的离群点检测方法（RKBOutlier）.从异质信息网络中抽取两种类型的对象以及链接两种对象的语义信息，将待检测的数据作为属性对象，将另一类型数据作为目标对象，对目标对象进行聚类来检测属性对象在各个聚类中的分布情况，数据分布异常的对象即为离群点.将排序和聚类相结合来显著提高聚类的准确度.实验结果表明，RKBOutlier可以在双类型异质信息网络中有效地检测出离群点.

Abstract

Mining the outliers that are different from normal data objects in the network is one of the important tasks in data mining. At present

the research aiming at outlier detection in bi-typed heterogeneous information network is relatively small. The methods which are applicable to homogeneous network can not be applied to bi-typed heterogeneous networks. Therefore

we propose a Rank-Kmeans Based Outlier detection method

called RKBOutlier

in heterogeneous information network. The two kinds of the objects and the connected semantic information are extracted from the heterogeneous information network. One type of the objects is regarded as the attribute objects

another type of the objects is regarded as the target objects. We perform cluster partitioning on target objects to detect the distribution of the attribute objects in each cluster. The objects which are abnormal at data distribution are considered to be the outliers. Ranking and clustering are combined to significantly improve the accuracy of clustering. The experimental results show that RKBOutlier can effectively detect outliers in bi-typed heterogeneous information network.

关键词

Keywords

references

Views

298

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Temporal-Pulse Amplitude Clustering-Based Unknown Radar Emitter Signal Sorting Method

Continuous Relation Extraction Technique Based on Dynamic Relational Prototypes

Pseudo-label Denoising and SAM Optimization for Large-scale Unsupervised Semantic Segmentation

Heuristic Clustering Algorithm Based on Sub-Game Perfect Equilibrium

Density Peaks Clustering Algorithm Based on Geodesic Distance and Cosine Mutual Reverse Nearest Neighbors for Manifold Datasets

Related Author

LIN Jin-jian

HUANG Ming-jun

SUN Hui-bo

LIN Zi-han

XIE Kai

ZHONG Jiang

DAI Qi-zhu

LI Xue

Related Institution

Deptment of Electronics and Communication Engineering, Sun Yat-sen University of China

The 29th Research Institute of China Electronics Technology Group Corporation

School of Computer Science, Chongqing University

Key Laboratory of Dependable Service Computing in Cyber Physical Society, Chongqing University

School of Electrical Engineering and Computer Science, The University of Queensland

⁰