Vertex unique labelled subgraph mining
نویسنده
چکیده
With the successful development of efficient algorithms for Frequent Subgraph Mining (FSM), this paper extends the scope of subgraph mining by proposing Vertex Unique labelled Subgraph Mining (VULSM). VULSM has a focus on the local properties of a graph and does not require external parameters such as the support threshold used in frequent pattern mining. There are many applications where the mining of VULS is significant, the application considered in this paper is error prediction with respect to sheet metal forming. More specifically this paper presents a formalism for VULSM and an algorithm, the Right-most Extension VULS Mining (REVULSM) algorithm, which identifies all VULS in a given graph. The performance of REVULSM is evaluated using a real world sheet metal forming application. The experimental results demonstrate that all VULS (Vertex Unique Labelled Subgraphs) can be effectively identified.
منابع مشابه
Minimal Vertex Unique Labelled Subgraph Mining
This paper introduces the concept of Vertex Unique Labelled Subgraph Mining (VULSM), a specialised form of subgraph mining. A VULS is a subgraph defined by a set of edge labels that has a unique vertex labelling associated with it. A minimal VULS is then a VULS which is not a supergraph of any other VULS. The application considered in this paper, for evaluation purposes, is error prediction wit...
متن کاملA Correspondence Between Maximal Complete Bipartite Subgraphs and Closed Patterns
For an undirected graph without self-loop, we prove: (i) that the number of closed patterns in the adjacency matrix of is even; (ii) that the number of the closed patterns is precisely double the number of maximal complete bipartite subgraphs of ; (iii) that for every maximal complete bipartite subgraph, there always exists a unique and distinct pair of closed patterns that matches the two vert...
متن کاملFrequent subgraph mining algorithms on weighted graphs
This thesis describes research work undertaken in the field of graph-based knowledge discovery (or graph mining). The objective of the research is to investigate the benefits that the concept of weighted frequent subgraph mining can offer in the context of the graph model based classification. Weighted subgraphs are graphs where some of the vertexes/edges are considered to be more significant t...
متن کاملLabeling Subgraph Embeddings and Cordiality of Graphs
Let $G$ be a graph with vertex set $V(G)$ and edge set $E(G)$, a vertex labeling $f : V(G)rightarrow mathbb{Z}_2$ induces an edge labeling $ f^{+} : E(G)rightarrow mathbb{Z}_2$ defined by $f^{+}(xy) = f(x) + f(y)$, for each edge $ xyin E(G)$. For each $i in mathbb{Z}_2$, let $ v_{f}(i)=|{u in V(G) : f(u) = i}|$ and $e_{f^+}(i)=|{xyin E(G) : f^{+}(xy) = i}|$. A vertex labeling $f$ of a graph $G...
متن کاملA Closed Frequent Subgraph Mining Algorithm in Unique Edge Label Graphs
Problems such as closed frequent subset mining, itemset mining, and connected tree mining can be solved in a polynomial delay. However, the problem of mining closed frequent connected subgraphs is a problem that requires an exponential time. In this paper, we present ECE-CloseSG, an algorithm for finding closed frequent unique edge label subgraphs. ECE-CloseSG uses a search space pruning and ap...
متن کامل