A Parallel Algorithm for Community Detection in Social Networks, Based on Path Analysis and Threaded Binary Trees

Abstract

Several synchronous applications are based on the graph-structured data; among them, a very important application of this kind is community detection. Since the number and size of the networks modeled by graphs grow larger and larger, some level of parallelism needs to be used, to reduce the computational costs of such massive applications. Social networking sites allow users to manually categorize their friends into social circles (referred to as lists on Facebook and Twitter), while users, based on their interests, place themselves into groups of interest. However, the community detection and is a very effortful procedure, and in addition, these communities need to be updated very often, resulting in more effort. In this paper, we combine parallel processing techniques with a typical data structure like threaded binary trees to detect communities in an efficient manner. Our strategy is implemented over weighted networks with irregular topologies and it is based on a stepwise path detection strategy, where each step finds a link that increases the overall strength of the path being detected. To verify the functionality and parallelism benefits of our scheme, we perform experiments on five real-world data sets:

Hybrid CPU-GPU Community Detection in Weighted Networks

“…This causes changes to the irregular topologies mentioned just above, which pose another burden in community detection. This work extends our previous parallel community detection strategy [1], which introduced the use of threaded binary trees for community detection. One of the issues raised in this work was whether it is possible to implement the proposed algorithm or parts of it in the GPU and if such an implementation could improve the computational efficiency.…”

“…Section III gives a very brief theoretical background of the community detection problem and presents the criteria posed in our strategy to detect memberships and overlaps. More details can be found in [1]. Section IV describes our community detection algorithm and the CPU-GPU scheduling strategy, taking into account the four design principles presented above.…”

Taxonomy of Community Detection over Social Media

“…Experimental results show that the multi-spread model and LICD algorithm can solve the influence maximizing problem and local influenced community detection efficiently. Souravlas et al [27] exploited parallel processing and threaded binary tree data structure and presented a community detection algorithm. The main idea of the algorithm is to avoid race conditions while assuring the load balancing among executing processors.…”

Scheduling Fair Resource Allocation Policies for Cloud Computing through Flow Control

“…The idea of an application's dominant resource was initially presented by Ghodsi, in [8]. For example, there are applications which are CPU intensive in the sense that they mostly depend on CPU performance, like large, graph-based community detection schemes [9,10] or other applications which are I/O intensive, like date replication applications that require large disk spaces to accommodate the required replicas [11]. When the dominant resources vary, the resource scheduling problem becomes rather cumbersome, in the sense that, the policy employed should not only satisfy the users' requirements, but also balance each resource shared to be dominant for some applications and as non-dominant by other.…”

“…The two queue systems execute in parallel, implementing the three steps of our fair policy described in Paragraph II.A. We get V = [4,6,8,9,10]. From Equation 4…”