Mining social applications network from business perspective using modularity maximization for community detection

There are different social applications available for different purposes. A lot of information about different fields including politics, sports, business, movie industry, etc., pass by and people are not well informed about most important happenings taking place in the world. Social applications usage varies among people in different parts of the world. A social application in a community may be popular for a particular purpose such as Twitter that may be used as a core application for political use among people in one part of the world, whereas other people may use Facebook, WeChat or YouTube for entertainment and other purposes and may not be aware of the important political changes taking place in the world. Social media usage by businesses can be improved by knowing the maximum usage of particular social applications among different communities of people so that targeted contents including information, advertisements, services and recommendations can be forwarded to them. In this paper, we mine social applications network by extracting knowledge according to the popularity of social applications. r -neighborhood technique is used for removal of edges from social applications network. Users are assigned to different communities based on the modularity scores. Optimal communities are found using divisive clustering approach that partitions the graph until maximum modularity score is achieved. Community detection method is also performed in gephi tool and using k -nearest neighbors graph. The trends of the social applications are analyzed among different communities, and it is seen that r -neighborhood, k -nearest neighbors and gephi tool result in Twitter, YouTube and Facebook as the most popular applications among other social applications. Related contents can be forwarded to the respective communities as well as people of a community defined by popularity of a social application can also be well informed about other happenings in the world such as Twitter and YouTube communities that may advertise about different products, whereas Facebook and YouTube communities are advertised with political news. The modularity function of k -nearest neighbors has the highest value and gives better interpretation of communities than other two techniques.

[1]  David Kempe,et al.  Modularity-maximizing graph communities via mathematical programming , 2007, 0710.2533.

[2]  A. Medus,et al.  Detection of community structures in networks via global optimization , 2005 .

[3]  Susan Bastani,et al.  Fuzzy community detection on the basis of similarities in structural/attribute in large-scale social networks , 2021, Artificial Intelligence Review.

[4]  Xingwang Zhao,et al.  A community detection algorithm based on graph compression for large-scale social networks , 2020, Inf. Sci..

[5]  M E J Newman,et al.  Modularity and community structure in networks. , 2006, Proceedings of the National Academy of Sciences of the United States of America.

[6]  Yuzhong Chen,et al.  Local community detection algorithm based on local modularity density , 2021, Applied Intelligence.

[7]  M E J Newman,et al.  Community structure in social and biological networks , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[8]  Roger Guimerà,et al.  Extracting the hierarchical organization of complex systems , 2007, Proceedings of the National Academy of Sciences.

[9]  Abdallah Abarda,et al.  The greedy coupled-seeds expansion method for the overlapping community detection in social networks , 2021, Computing.

[10]  R. Guimerà,et al.  Functional cartography of complex metabolic networks , 2005, Nature.

[11]  Kun He,et al.  Hidden Community Detection in Social Networks , 2017, Inf. Sci..

[12]  Meng Wang,et al.  Community Detection in Social Networks: An In-depth Benchmarking Study with a Procedure-Oriented Framework , 2015, Proc. VLDB Endow..

[13]  S. Fortunato,et al.  Resolution limit in community detection , 2006, Proceedings of the National Academy of Sciences.

[14]  Padhraic Smyth,et al.  A Spectral Clustering Approach To Finding Communities in Graph , 2005, SDM.

[15]  Santo Fortunato,et al.  Community detection in graphs , 2009, ArXiv.

[16]  Charu C. Aggarwal,et al.  A Survey of Clustering Algorithms for Graph Data , 2010, Managing and Mining Graph Data.

[17]  Paulo Shakarian,et al.  Mining for geographically disperse communities in social networks by leveraging distance modularity , 2013, KDD.

[18]  Béla Bollobás,et al.  Modern Graph Theory , 2002, Graduate Texts in Mathematics.

[19]  A. Levine,et al.  Gene assessment and sample classification for gene expression data using a genetic algorithm/k-nearest neighbor method. , 2001, Combinatorial chemistry & high throughput screening.

[20]  Julita Vassileva,et al.  Semantic Adaptive Social Web , 2011, UMAP Workshops.

[21]  Ken Wakita,et al.  Finding community structure in mega-scale social networks: [extended abstract] , 2007, WWW '07.

[22]  Philip S. Yu,et al.  Top 10 algorithms in data mining , 2007, Knowledge and Information Systems.