Efficient Clustering of Structured Documents Using Graph Self-Organizing Maps

Hagenbuchner, Markus; Tsoi, Ah Chung; Sperduti, Alessandro; Kc, Milly

doi:10.1007/978-3-540-85902-4_19

Cited by 11 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first experiment was visually evaluated and the second was evaluated using the metric CQ defined by (10). Also the results of the second experiment were compared with GraphClust method [12] (described in section I) using the option for sparse graphs.…”

Section: Resultsmentioning

confidence: 99%

“…Some groups in SOM are empty and most of the groups generated by GraphClust have only one element. Quality of clustering was evaluated using (10). When comparing SOM clustering with expert clustering .…”

Section: Resultsmentioning

confidence: 99%

“…For the case that both groups are empty, is not meaningful. Using (9), the clustering quality is defined as: (10) In the best case, when each cluster is identical to one and only one cluster , and completely different to the rest, CQ = 1. This is because for this case there is only one value "one" in each row and each column of matrix , and the rest of the values in are zero.…”

Section: Measuring the Quality Of Clusteringmentioning

confidence: 99%

“…To train the weights in the SOM, they define a function also based in the number of operations. Hagenbuchner et al [10] implemented an algorithm called "Graph Self-Organizing Maps" (GraphSOMs) which uses a recursive representation of graphs made of numerical labels. GraphSOMs is a generalization of "Self-organizing Maps for Structured Domain" (SOM-SD), which uses recursive callings to handle trees.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Finding patterns in labeled graphs using spectrum feature vectors in a SOM network

Fonseca

Gómez-Gil

Gonzalez

et al. 2011

The 2011 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Knowledge discovery in structured databases is very important nowadays. In the last years, graph-based data mining algorithms have used artificial neural networks as tools to support clustering. Several of these algorithms have obtained promising results, but they show expensive computational costs. In this work we introduce an algorithm for clustering graphs based on a SOM network, which is part of a process for discovering useful frequent patterns in large graph databases. Our algorithm is able to handle non-directed, cyclic graphs with labels in vertices and edges. An important characteristic is that it presents polynomial computational complexity, because it uses as input a feature vector built with the spectra of the Laplacian of an adjacent matrix. Such matrix contains codes representing the labels in the graph, which preserves the semantic information included in the graphs to be grouped. We tested our algorithm in a small set of graphs and in a large structured database, finding that it creates meaningful groups of graphs.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Measuring the Quality Of Clusteringmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Finding patterns in labeled graphs using spectrum feature vectors in a SOM network

Fonseca

Gómez-Gil

Gonzalez

et al. 2011

The 2011 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

show abstract

“…More precisely, as far as Wikipedia is concerned, the chosen competitors are XCFS [19], CRP as well as 4RP [28], SOM [17] and LSK [25]. On Sigmod, the comparison is performed against XCFS [19].…”

Section: B Competitorsmentioning

confidence: 99%

A Latent Semantic Approach to XML Clustering by Content and Structure Based on Non-negative Matrix Factorization

Costa

Ortale

2013

2013 12th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

Non-negative matrix factorization is intensively used in text clustering. We investigate its exploitation in the XML domain for clustering XML documents by structure and content into topically homogeneous groups. Non-negative matrix factorization is performed through an alternating least squares method, which incorporates expedients to attenuate the burden of largescale factorizations. This is especially relevant when massive textcentric XML corpora are processed. Empirical evidence from a comparative evaluation on real-world XML corpora reveals that our approach overcomes several state-of-the-art competitors in effectiveness.

show abstract

Mining Cluster Patterns in XML Corpora via Latent Topic Models of Content and Structure

Costa

Ortale

2019

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Efficient Clustering of Structured Documents Using Graph Self-Organizing Maps

Cited by 11 publications

References 7 publications

Finding patterns in labeled graphs using spectrum feature vectors in a SOM network

Finding patterns in labeled graphs using spectrum feature vectors in a SOM network

A Latent Semantic Approach to XML Clustering by Content and Structure Based on Non-negative Matrix Factorization

Mining Cluster Patterns in XML Corpora via Latent Topic Models of Content and Structure

Contact Info

Product

Resources

About