Graph techniques in Pattern Recognition

grpah에서의 pattern 인식에는 다양한 종류의 기술이 존재한다.

이 중 연구에서는 전체 그래프 내에서 반복되는 subgraph를 찾아야하며, 정확히 일치하지 않더라도, 높은 유사도를 가지는 경우, 패턴으로 포함시켜야 하므로, Graph matching 중 inexact matching과, Graph learning 중 learning of graphs를 참고하여 적용 가능 여부를 살펴야 하겠다. 

---

A 2008 paper by U. Maulik 108 presents an algorithm for finding repeated subgraphs within large graphs, which can be considered an unsupervised form of graphbased learning. This can be used for data mining in domains suitable to a structural representation, e.g. web pages or molecular databases. The algorithm uses Evolutionary Programming to perform the search, with a fitness function based on the compression of the original graph attainable by the detection of the repeated substructure.

- Graph Matching and Learning in Pattern Recognition in the Last 10 Years