Complex Social network Partition for Balanced subnetworks---Hao Lan zhang,jiming liu,chunyu feng,chaoyi Pang,tongliang Li, Jing He reading

Summary: Abstract-complex Social network analysis methods has been applied extensively in various domains including online SOCI Al Media, biological complex networks, etc. Complex social networks is facing the challenge of information overload. The demands for efficient complex network analysis methods has been rising in recent years, particularly the extensive US E of online social applications, such as Flickr, Facebook and LinkedIn. This paper aims to simplify the network complexity through partitioning a large complex network into a set of less complex Networks. Existing Social network analysis methods is mainly based on complex network theory and data mining techniques. These methods is facing the challenges while dealing with extreme large social network data sets. Particularly, the difficulties of maintaining the statistical characteristics of partitioned Sub-networks has been Increa Sing dramatically. The proposed Normal distribution (ND) based method can balance the distribution Of the partitioned Sub-networks according to the original complex network. Therefore, each subnetwork can has its degree distribution similar to that of the original network. This can is very beneficial for analyzing sub-divided networks and potentially reducing the complexity in dynamic online s Ocial environment.

Understanding: This article mainly discusses the large-scale complex network encountered in the analysis of the problem, the network is too complex, so a better way is to decompose the original network into a smaller scale network can speed up the analysis.

One of the main problems in this is that the decomposition of the sub-network is not the same as the original network has the same nature.

This paper introduces several other methods of dividing network, then puts forward its own method, which seems to be an integration on the basis of predecessors. Divided into two parts, partition, merge.

Finally in a data set above the experiment gave the results, feel the data set a little bit, do not know to do research when it takes a large data set to do experiments.

It took about three hours to read this paper, the content is relatively concise, looked up some of the terminology, some information, feeling still a little harvest.

Harvest:

1. About the concept of power law distribution, power-law distribution, Link: http://blog.sina.com.cn/s/blog_49f6467e0100qh9l.html

2. Find out the meaning of some nouns

3. A little understanding of some of the terminology of the network, such as the Scale-free scale-free network,

In the network theory, the scale-free network (or non-scaling network) is a complex network with a class of characteristics, the typical feature is that most nodes in the network are connected with very few nodes, and very few nodes are connected with very many nodes. The presence of such a critical node (known as a " hub " or " distributed Node ") makes scale-free networks more resilient to unexpected failures, but vulnerable in the face of collaborative attacks. Many networks in reality have scale-free features, such as the Internet, financial system networks, social networking, and so on. ----Wikipedia, Link: https://zh.wikipedia.org/wiki/%E6%97%A0%E5%B0%BA%E5%BA%A6%E7%BD%91%E7%BB%9C

4. degree distribution: Degree distribution degree distribution refers to the distribution of the degree of the node. In the network, each node is connected to some other node, and the number of such connections is called the degree of the node. What is the degree of randomly extracting a node in a network? This probability distribution is called the node's degree distribution ^{[2]: 11.}

Complex Social network Partition for Balanced subnetworks---Hao Lan zhang,jiming liu,chunyu feng,chaoyi Pang,tongliang Li, Jing He reading