To process data which increasingly become larger and more complicated, frequent graph mining was proposed, and numerous methods for this has been suggested with various approaches and applications. However, these methods do not consider characteristics of sub-graphs for each length in detail since they generally use a constant minimum support threshold for mining frequent sub-graphs. Small sub-graphs with a few vertices and edges tend to be interesting if their supports are high, while large ones with lots of the elements can be interesting even if their support are low. Motivated by this issue, we propose a novel frequent graph mining algorithm, Frequent Graph Mining with Length-Decreasing Support Constraints (FGM-LDSC). The algorithm applies various support constraints depending on lengths of sub-graphs, and thereby we can obtain more useful results. © 2013 Springer Science+Business Media Dordrecht(Outside the USA).
CITATION STYLE
Lee, G., & Yun, U. (2013). Frequent graph pattern mining with length-decreasing support constraints. In Lecture Notes in Electrical Engineering (Vol. 240 LNEE, pp. 185–192). https://doi.org/10.1007/978-94-007-6738-6_24
Mendeley helps you to discover research relevant for your work.