Network Based Community Detection By Using Bisecting Hierarchical Clustering

Author Name: Miss Snehal Prakash Mahajan, Prof Abhijeet R. Raipurkar
Author Email:


Social media is the platform where human interacts with each other, it is becoming part of our life. One of the task of detecting communities in social networks are real time networks that can be classified by using clustering techniques to extract and find hidden communities from the network. For example Facebook, the different communities can be extracted like people belonging to different interests and communities and this information can be used for marketing purposes. Nowadays, studies involving this clustering processes are basically composed of modularity maximization. In this paper the author proposes a bisecting hierarchical clustering based on a measure known as inter and inter cluster similarities to detect communities within the networks. The experiments indicates a successful performance.


Social Networks, Semantic Measures, Clustering.


Detection of communities in social sites or networks aims at the finding partitions of a graph with clusters with “similar” data points. It is the method that involves detecting communities in social networks in order to explain similar behaviors among groups of individuals [10]. Hence, the representation of social networks by using graphs that enables their analysis by using graph clustering methods. Apart from the sociology, there are also some other areas where the community detection problem might be suitable, for example, biology and many various areas [9].

To analyse the structure of the original systems, often the corresponding network structure is studied using groups of nodes having more intra group and less inter group edges. Such groups exist in most real world networks and influence the behaviour of the underlying system. The community detection is an important problem and it has the potential to solve many real world problems. Most of the existing community detection algorithms lack the ability to detect accurate community boundaries if the difference between the internal and the external node degree does not exceed a detect ability threshold. Community detection is the method of discovering groups in a network where different groups possesses different properties.

Need of community detection in social media:-

  • Human beings are social.
  • Easy-to-use social media allows people to extend their social life in unprecedented ways.
  • Find difficult to meet or contact friends in the physical world, but much easier to find friend online with similar interests.
  • Interactions between nodes can help determine communities.


This algorithm successfully classifies clusters into different community groups. Previous algorithms failed to gain knowledge intra cluster details, to overcome this problem this technique can be used to gain detailed intra cluster knowledge. For this project, we have taken data from Social Network Site Reddit pizza services which gives online food services and we successfully classified the different communities based on their texts, comments, orders etc.

74 total views, 1 views today

Download File

About the author: admin