Social network analysis, as an interdisciplinary research field, has drawn an increasing attention in last decade. Inspired by various interactions among inviduals, we developed this Social Network Computation Paltform (short for SNCP), which is capable of conducting algorithms on a variety of real-world datasets and providing vivid visualization tool for researchers.
We have developed some pages for exhibiting community detection, link prediction, SIR model, etc. We also provides a data formatting tool, aimming to converting the different datasets into an desired format.
Thanks for your participation in this platform. The following figure shows the relationships of registered users, where the link connected with A and B represent user A follows B or User B follows A. Put your mouse over the node to check its neighbors. The size of a focal node is determined by the number of its neighbors, i.e., the more neighbors, the larger node. The text alongside an link is the time of this followship occured.
Community structure (or Cluster), as a densely connected group of nodes, is widely existed in many complex systems, such as social networks, biological networks and so on. Numerous of algorithms have been proposed to solve the problem of detecting communities in complex networks, e.g, GN [1], which employs cutting maximum edge-betweenness strategy and has a good performance under the evaluation of Modularity(Q) [2]. The comparison of classic algorithms is represented here. Some of the datasets are from Mark Newman's blog .
Link Prediction [3] mainly includes the prediction of unknown links and future links. Unknown links prediction aims to predict the missing data based on small parts of a large-scale network to reduce the experimental costs, or recover data when the network is initially incomplete. Future links is to estimate the links that will be appear in the future. The link prediction page is here.
Reference: