Structure/attribute computation of similarities between nodes of a RDF graph with application to linked data clustering

Khosravi-Farsani, Hadi; Nematbaksh, Mohammadali; Lausen, Georg

doi:10.3233/ida-130573

Cited by 3 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to clustering, we need measure similarity of node pairs. The similarity node pairs mainly be measured by the graph structure information [6] or graph node attribute and structural information [7]. P-Rank is an important measure model for structural similarity, which was widely used in data mining field, such as collaborative filtering, network graph clustering, KNN query.…”

Section: Introductionmentioning

confidence: 99%

Distributed Storage of RDF Based on Clustering

Leng¹,

Lu²

2015

Proceedings of the 2nd International Conference on Civil, Materials and Environmental Sciences

View full text Add to dashboard Cite

With the wide application of RDF(Resource Description Framework) data, the data volume grows rapidly. Therefore, RDF storage has become a hot research issue in data storage field currently. Distributed storage is an effective way to solve the storage and query of RDF data, and data partition is the premise of data distributed storage. In this paper we use graph clustering idea to realize the effective partition of RDF data. RDF can be described as a directed graph, so in this paper we use PRank (Penetrating Rank) algorithm to calculate the similarity of RDF graph node pairss, and then the improved K-means clustering algorithm is implemented to cluster the similarity results, so as to realize the distributed storage of RDF data. The experimental results show that, this method can complete the RDF data partition effectively, makes the intra-cluster similarity is smaller, and the larger the inter-cluster similarity.

show abstract

Section: Introductionmentioning

confidence: 99%

Distributed Storage of RDF Based on Clustering

Leng¹,

Lu²

2015

Proceedings of the 2nd International Conference on Civil, Materials and Environmental Sciences

View full text Add to dashboard Cite

show abstract

“…The key of Graph clustering is how to measure similarity between nodes, there are many similarity methods for measuring nodes in the graph, such as [4,5] method based on structure, attribute and structure/attribute [6,7]. Structural similarity measure does not need to analyze the information contained in the entity, is the most commonly method of similarity of graph nodes.…”

Section: Introductionmentioning

confidence: 99%

A Big Graph Clustering Algorithm Based on MapReduce

Leng

Zhang

2014

AMR

View full text Add to dashboard Cite

Graph clustering is an important technology in graph analysis area, the measure of similarity between node of graph is the presise for graph clustering. SimRank algorithm is a kind of universal structure similarity calculation model which is proposed by Jeh and Widom. SimRank algorithm using iterative method to calculate the similarity between nodes, so the time and space complexity is very high. With the rapid increase of data, the ability of single machine can not meet the requirement of the large-scale data calculation. In this paper, the distributed SimRank algorithm was proposed based on Mapreduce and was used to measure the similarity of graph. Then the distributed AP clustering algorithm was designed for clustering analysis graph nodes. The experimental was executed to compare the clustering running time and speedup and results show that the method can efficiently complete graph nodes similarity measure and clustering the large graph effectively.

show abstract

Faceted fusion of RDF data

Liu

Wang

et al. 2015

Information Fusion

View full text Add to dashboard Cite

Structure/attribute computation of similarities between nodes of a RDF graph with application to linked data clustering

Cited by 3 publications

References 17 publications

Distributed Storage of RDF Based on Clustering

Distributed Storage of RDF Based on Clustering

A Big Graph Clustering Algorithm Based on MapReduce

Faceted fusion of RDF data

Contact Info

Product

Resources

About