Since there has been no systematic summary on this topic, in this paper, we review the main-stream graph construction/learning methods involved in both general machine learning algorithms (including semi-supervised learning, clustering, manifold learning, and spectral kernel learning, etc).
Consequently, efficient knowledge discovery algorithms for mining frequent subgraphs from these dynamic streaming graph structured data are in demand.
To keep low the overhead of partitioning graphs, even when processing the ever-increasing modern graphs, many previous studies use lightweight streaming graph-partitioning policies.
By analyzing this model under the load of realistic graph-data characteristics, we propose a method to identify important performance issues and then design new streaming graph-partitioning policies to address them.
The Gephi Streaming API allows graph streaming from a client to a server where Gephi is running.
Flow graphs are constructed for five types of applications: web, P2P download, P2P stream, video stream and instant messaging.
To achieve scalability, SCORE allows and encourages the programmer to ignore the hardware capacity of a particular platform and focus on capturing the fully spatial, streaming dataflow graph.
Unfortunately, ignoring or muting alerts are both much easier than managing your live-stream social graph so you only follow Periscopers or Meerkatters with a high signal to noise ratio.
This framework takes the graph stream as the initial input, and summarizes the original graph stream into the sketched graph by using Hashing function over the vertices; Based on the obtained sketched graph, formal concept analysis is utilized for mining the cliques from it.
Inspired by this problem, we propose (i) a specialized data structure called DSMatrix, which captures important data from dense graph streams onto the disk directly and (ii) stream mining algorithms that make use of such structure in order to mine frequent patterns effectively and efficiently.
