Graph Sampling
Graph Sampling
In graph sampling we discover the all methods for patterns small graph from large no. of data. In data mining lots of data available but that all data represented with user requirement. The lots of patterns are used for representing data into graphical format like 2D, 3D method or pi chart, flowchart. In graph sampling all data are represented with use of graphs. The all mining data was show to user with use of graph method mainly. The graph mining is best research area within data mining.
Frequent Sub Graph Mining
The frequent sub graph mining gives a small number of graphs as a result from large graph database. In that mining lots of algorithms are used from data mining and create final output to user. The frequent sub graph mining comes under 2 different types mainly
1. Algorithm using BSF search strategy

 That all algorithm based on Apriori algorithm approach.
 The graph is divided into ‘K’ and ‘K+1’ formation.
 The size of graph defined by no. of vertices in that graph.
In that algorithm basically 2 algorithms occurs mainly
 AGM Algorithm
That algorithm is based on Apriori algorithm mainly.
That algorithm used adjacent matrix for graph representation.
 FSG Algorithm
That algorithm is based on Apriori algorithm mainly.
–Edges in that graphs are presented as a frequent items.
Every time additional edges are attached for finding frequent item in that graph technique.
2. Algorithm using DFS search strategy
 That type of algorithm comes under pattern graph approach.
 BSF graph technique is costly then DFS is used mainly.
That graph technique fallow 1 algorithm mainly.
 G SPAN Algorithm
That algorithm based on pattern search growth approach.
–Multiple candidate generation can be reduced in G Span.
It work on labeled sample graphs.
–Each graph has unique label for each edge and its vertices.
It finds frequent sub graph easily.