Discovery of extreme events-related communities in contrasting groups of physical system networks
Data Mining and Knowledge Discovery
| Hi-index | 0.00 |
In this paper, aimed at the problem of detecting closed frequent subgraphs in biological networks, an improved FP-growth algorithm MaxFP is presented, which is based on the simplification model appropriate to biological networks. The defects of the algorithm based on item-set mining are analyzed when it is applied to biological networks, and which is overcome in MaxFP. In addition, MaxFP also takes the biological network characteristics into account. Experiment results show that MaxFP runs faster than the algorithms based on Apriori, and MaxFP not only detects maximal frequent subgraphs, but also finds more frequent subgraphs having biological meaning. The results got by performing Apriori based algorithms many times can be got by performing MaxFP once.