Analysis of the Modern Algorithms’ Accuracy for Communities Identification on Networks when Working with Graph Databases

封面

如何引用文章

全文:

开放存取 开放存取
受限制的访问 ##reader.subscriptionAccessGranted##
受限制的访问 订阅或者付费存取

详细

In this paper, we consider methods for extracting communities in networksusing various algorithms. The Girvan-Newman, Louvain, Walktrap and Leiden algorithms were presented and the results of their application on the Wikipedia graph were analyzed. Various metrics were used to assess the quality of the isolated communities, and the results were stored in the Neo4j graph database. The results showed that the Leiden and Louvain algorithms with a resolution equal to one showed the best results compared to other algorithms.

全文:

受限制的访问

作者简介

Ekaterina Kazakova

Financial University under the Government of the Russian Federation

编辑信件的主要联系方式.
Email: 191841@edu.fa.ru

student at the Faculty of Information Technology and Big Data Analysis of the Financial University under the Government of the Russian Federation

俄罗斯联邦, Moscow

参考

  1. Barabasi A.-L., Albert R. Emergence of scaling in Random networks. Science. 1999. No. 286. Pp. 509–512. doi: 10.1126/science.286.5439.509.
  2. Blondel V.D., Guillaume J.-L., Lambiotte R., Lefebvre E. Fast unfolding of communities in large networks. Journal of Statistical Mechanics Theory and Experiment. 2008. doi: 10.1088/1742-5468/2008/10/P10008.
  3. Bruna J., Li Xiang. Community detection with graph neural networks. 2017.
  4. Circulo library. URL: http://lab41.github.io/Circulo/
  5. Clauset A., Newman M.E.J., Moore C. Finding community structure in very large networks. Physical Review E. 2004. URL: http:// arxiv.org/abs/cond-mat/0408187
  6. Coscia M., Rossetti G., Giannotti F., Pedreschi D. Demon: A local-first discovery method for overlapping communities. KDD. 2012. URL: http://www. michelecoscia.com/wp-content/uploads/2012/08/cosciakdd12.pdf
  7. Blondel V.D., Guillaume J.-L., Lambiotte R., Lefebvre E. Fast unfolding of communities in large networks. J. Stat. Mech. 2008. URL: http://arxiv.org/abs/0803.0476
  8. Fortunato S. Community detection in graphs. Physics Reports. 2009. URL: http://arxiv.org/abs/0906.0612
  9. Girvan M., Newman M.E.J. Community structure in social and biological networks. Proceedings of the National Academy of Sciences. 2001. URL: http://arxiv.org/abs/cond-mat/0112110
  10. Gregory S. An algorithm to find overlapping community structure in networks. Proceeding PKDD 2007 Proceedings of the 11th European Conference on Principles and Practice of Knowledge Discovery in Databases. 2007. URL: http://www.cs.bris.ac.uk/Publications/Papers/2000689.pdf
  11. Girvan M., Newman M.G.M., Newman M.E.J. Community structure in social and biological networks. Proceedings of the National Academy of Sciences of the United States of America. 2002. No. 99. Pp. 7821–7826. doi: 10.1073/pnas.122653799.
  12. Hamilton W.L., Ying R., Leskovec J. Representation learning on graphs: Methods and applications. 2017.
  13. Nikhil M., Carin L., Rai P. Stochastic block models meet graph neural networks. 2019.
  14. Pons P., Latapy M. Computing communities in large networks using random walks. J. Graph Algorithms Appl. 2006. No. 10. Pp. 191–218. doi: 10.7155/jgaa.00124.

补充文件

附件文件
动作
1. JATS XML
下载
2. Fig. 1. Example of the simple graph

下载 (21KB)
索引源数据
3. Fig. 2. Graph retained from the networkx

下载 (250KB)
索引源数据
4. Fig. 3. Complete database

下载 (171KB)
索引源数据
5. Fig. 4. Node centrality graph

下载 (51KB)
索引源数据
6. Fig. 5. Identification of communities by the Girvan-Newman algorithm

下载 (236KB)
索引源数据
7. Fig. 6. Identification of communities by the Louvain algorithm

下载 (230KB)
索引源数据
8. Fig. 7. Identification of communities by the Leiden algorithm

下载 (233KB)
索引源数据
9. Fig. 8. Identification of communities by the Walktrap algorithm

下载 (227KB)
索引源数据
10. Algorithm quality metrics

下载 (542KB)
索引源数据
11. Fig. 9. Friedman rank test

下载 (48KB)
索引源数据
12. Fig. 10. Similarity matrix for random walk and Leiden algorithms

下载 (37KB)
索引源数据