Add FP-GraphMiner algorithm
ZigRazor opened this issue · comments
FP-GraphMiner - A Fast Frequent Pattern Mining Algorithm for Network Graphs
A novel Frequent Pattern Graph Mining algorithm, FP-GraphMiner, that compactly
represents a set of network graphs as a Frequent Pattern Graph (or FP-Graph).
This graph can be used to efficiently mine frequent subgraphs including maximal
frequent subgraphs and maximum common subgraphs.
URL: https://www.researchgate.net/publication/235255851
You can take inspiration from this :
https://github.com/TheAlgorithms/Python/blob/master/graphs/frequent_pattern_graph_miner.py
@ZigRazor Could you assign me to this issue?
Sure @Lxixnxuxs !
As this is a subgraph pattern mining problem, the input here is not a single but a collection of graphs. Thus we have to decide on the representation. Are there already any concepts chosen/implemented?
In this moment there isn't nothing similar, so you can approach to the problem in the way you prefer.
We can think to add a folder for this kind of algorithm with more than one graph and create general static function inside a template.
Or if you have a better solution, you can implement it
Since the problem statement requires the number of graphs to be constant, we could add some kind of graph collection class with an (standard) array of graph objects in its core. Then the pattern-mining-function can be a method of this class, preserving the structure of the project.
Yes, good idea, you can apply it
@Lxixnxuxs are you working on it?