Filters








128 Hits in 5.2 sec

Graph Kernels for RDF Data [chapter]

Uta Lösch, Stephan Bloehdorn, Achim Rettinger
2012 Lecture Notes in Computer Science  
Kernels, however, promise a more flexible approach by providing a powerful framework for decoupling the data representation from the learning task.  ...  We first review the problems that arise when conventional graph kernels are used for RDF graphs.  ...  Preliminaries: Data Mining Tasks for RDF The goal in this work is to make kernel machines available to learning tasks which use SW data as input.  ... 
doi:10.1007/978-3-642-30284-8_16 fatcat:frg62usjnrhktci74bmqdjt65q

RDF2Vec: RDF Graph Embeddings for Data Mining [chapter]

Petar Ristoski, Heiko Paulheim
2016 Lecture Notes in Computer Science  
We generate sequences by leveraging local information from graph substructures, harvested by Weisfeiler-Lehman Subtree RDF Graph Kernels and graph walks, and learn latent numerical representations of entities  ...  Our evaluation shows that such vector representations outperform existing techniques for the propositionalization of RDF graphs on a variety of different predictive machine learning tasks, and that feature  ...  -Kernels that count substructures in the RDF graph around the instance node.  ... 
doi:10.1007/978-3-319-46523-4_30 fatcat:hvt25fbunrcr7hqeuuxlnqxbim

Learning from biomedical linked data to suggest valid pharmacogenes

Kevin Dalleau, Yassine Marzougui, Sébastien Da Silva, Patrice Ringot, Ndeye Coumba Ndiaye, Adrien Coulet
2017 Journal of Biomedical Semantics  
Learning from these data, random forest enables identifying valid pharmacogenes with a F-measure of 0.73, on a 10 folds cross-validation, whereas graph kernel achieves a F-measure of 0.81.  ...  This identification relies on the classification of gene-drug pairs as either pharmacogenomically associated or not and was experimented with two machine learning methods -random forest and graph kernel  ...  Acknowledgements We acknowledge the participants of the SWAT4LS 2015 conference for their constructive feedback on the preliminary results of this work.  ... 
doi:10.1186/s13326-017-0125-1 pmid:28427468 pmcid:PMC5399403 fatcat:fhbddyuvwfe7bmoeeimskpc2bm

Skip Vectors for RDF Data: Extraction Based on the Complexity of Feature Patterns [article]

Yota Minami, Ken Kaneiwa
2022 arXiv   pre-print
Machine learning tasks for RDF graphs adopt three methods: (i) support vector machines (SVMs) with RDF graph kernels, (ii) RDF graph embeddings, and (iii) relational graph convolutional networks.  ...  In our evaluation experiments with RDF data, such as Wikidata, DBpedia, and YAGO, we compare our method with RDF graph kernels in an SVM.  ...  The graph kernels are defined by functions that calculate the distance between data by counting the common substructures for two graphs or nodes, enabling machine learning on graph data.  ... 
arXiv:2201.01996v3 fatcat:olqyuxmlkjesxmqvzewcjiwfxq

Biased graph walks for RDF graph embeddings

Michael Cochez, Petar Ristoski, Simone Paolo Ponzetto, Heiko Paulheim
2017 Proceedings of the 7th International Conference on Web Intelligence, Mining and Semantics - WIMS '17  
We generate sequences by exploiting local information from graph substructures, harvested by graph walks, and learn latent numerical representations of entities in RDF graphs.  ...  We evaluate our approach using di erent machine learning, as well as entity and document modeling benchmark data sets, and show that the naive RDF2Vec approach can be improved by exploiting Biased Graph  ...  presented in this paper has been partially funded by the Junior-professor funding programme of the Ministry of Science, Research and the Arts of the state of Baden-Wür emberg (project "Deep semantic models for  ... 
doi:10.1145/3102254.3102279 dblp:conf/wims/CochezRPP17 fatcat:2t3avsupirayvceccghev7maqu

A Linked Data Recommender System Using a Neighborhood-Based Graph Kernel [chapter]

Vito Claudio Ostuni, Tommaso Di Noia, Roberto Mirizzi, Eugenio Di Sciascio
2014 Lecture Notes in Business Information Processing  
In this paper we present a CB-RS that leverages LOD and profits from a neighborhood-based graph kernel.  ...  The boom of Linked Open Data (LOD) datasets with their huge amount of semantically interrelated data is thus a great opportunity for boosting CB-RSs.  ...  Due to the underlying data model of RDF datasets, we are particularly interested in those machine learning methods that are appropriate for dealing with objects structured as graphs.  ... 
doi:10.1007/978-3-319-10491-1_10 fatcat:bbytyvfqmneqlbenjupui5xcky

The knowledge graph as the default data model for learning on heterogeneous knowledge

Xander Wilcke, Peter Bloem, Victor de Boer, Michel Dumontier
2017 Data Science  
In modern machine learning, raw data is the preferred input for our models.  ...  Wilcke et al. / The knowledge graph as the default data model for learning on heterogeneous knowledge from their data, often creating a derivative of the original data in the process, they now prefer to  ...  . / The knowledge graph as the default data model for learning on heterogeneous knowledge Acknowledgements.  ... 
doi:10.3233/ds-170007 dblp:journals/datasci/WilckeBB17 fatcat:o5hclal77zayheldkar3lb3hf4

Not all Embeddings are created Equal: Extracting Entity-specific Substructures for RDF Graph Embedding [article]

Muhammad Rizwan Saeed, Charalampos Chelmis, Viktor K. Prasanna
2018 arXiv   pre-print
Several approaches have been recently proposed, for obtaining vector representations of KGs suitable for Machine Learning tasks, based on identifying and extracting relevant graph substructures using uniform  ...  graph embedding learned from the extracted substructures, outperform existing techniques in the task of entity recommendation in DBpedia.  ...  Acknowledgment This work is supported by Chevron Corp. under the joint project, Center for Interactive Smart Oilfield Technologies (CiSoft), at the University of Southern California.  ... 
arXiv:1804.05184v1 fatcat:o2nkt65zsngsnpyxmhk27bzzym

Mining the Web of Linked Data with RapidMiner

Petar Ristoski, Christian Bizer, Heiko Paulheim
2015 Journal of Web Semantics  
workflows without the need for expert knowledge in SPARQL or RDF.  ...  As an example, we show how statistical data from the World Bank on scientific publications, published as an RDF data cube, can be automatically linked to further datasets and analyzed using additional  ...  Graph kernels are used in kernel-based data mining algorithms, e.g., support vector machines.  ... 
doi:10.1016/j.websem.2015.06.004 fatcat:h4apkzrumfbjhehhxsryp6bvi4

Mining the Web of Linked Data with Rapidminer

Petar Ristoski, Christian Bizer, Heiko Paulheim
2015 Social Science Research Network  
workflows without the need for expert knowledge in SPARQL or RDF.  ...  As an example, we show how statistical data from the World Bank on scientific publications, published as an RDF data cube, can be automatically linked to further datasets and analyzed using additional  ...  Graph kernels are used in kernel-based data mining algorithms, e.g., support vector machines.  ... 
doi:10.2139/ssrn.3198927 fatcat:bzsr5eax35d4pommko6ydjcbdu

Mining the Web of Linked Data with Rapidminer

Petar Ristoski, Christian Bizer, Heiko Paulheim
2015 Social Science Research Network  
workflows without the need for expert knowledge in SPARQL or RDF.  ...  As an example, we show how statistical data from the World Bank on scientific publications, published as an RDF data cube, can be automatically linked to further datasets and analyzed using additional  ...  Graph kernels are used in kernel-based data mining algorithms, e.g., support vector machines.  ... 
doi:10.2139/ssrn.3199209 fatcat:ah3z74bycvbfdadocgezhwkdyq

Global RDF Vector Space Embeddings [chapter]

Michael Cochez, Petar Ristoski, Simone Paolo Ponzetto, Heiko Paulheim
2017 Lecture Notes in Computer Science  
Vector space embeddings have been shown to perform well when using RDF data in data mining and machine learning tasks.  ...  The approach generates sequences by leveraging local information from graph sub-structures, harvested by Weisfeiler-Lehman Subtree RDF Graph Kernels and graph walks, and then learns latent numerical representations  ...  presented in this paper has been partially funded by the Junior-professor funding programme of the Ministry of Science, Research and the Arts of the state of Baden-Württemberg (project "Deep semantic models for  ... 
doi:10.1007/978-3-319-68288-4_12 fatcat:sxum6ptcojeyjijdk5s6k6pkzi

A general representation scheme for crystalline solids based on Voronoi-tessellation real feature values and atomic property data

Randy Jalem, Masanobu Nakayama, Yusuke Noda, Tam Le, Ichiro Takeuchi, Yoshitaka Tateyama, Hisatsugu Yamazaki
2018 Science and Technology of Advanced Materials  
It was confirmed that the proposed feature vector from Voronoi real value binning generally outperforms the RDFbased one for the prediction of aforementioned properties.  ...  Together with electronegativitybased features, Voronoi-tessellation features from a given crystal structure that are derived from second-nearest neighbor information contribute significantly towards prediction  ...  Visualization for crystal structures was made using the VESTA software [56] , Voronoi cell figures were created using POV-Ray software [57] , data plot figures were generated using python Matplotlib  ... 
doi:10.1080/14686996.2018.1439253 pmid:29707064 pmcid:PMC5917445 fatcat:jshp37qu6zbmzhz26xdgj3l7bi

Leveraging Knowledge Graph Embedding Techniques for Industry 4.0 Use Cases [article]

Martina Garofalo and Maria Angela Pellegrino and Abdulrahman Altabba and Michael Cochez
2018 arXiv   pre-print
In this paper we discuss methods to convert (embed) the graph in a vector space, such that it becomes feasible to use traditional machine learning methods for Industry 4.0 settings.  ...  However, machine learning directly on graphs, needs feature engineering and has scalability issues.  ...  Part of this work was created as part of the Linked Data seminar at the I5 chair of the RWTH university.  ... 
arXiv:1808.00434v1 fatcat:pxa2tbw7tfgc7e4dbesena6lwu

lazar: a modular predictive toxicology framework

Andreas Maunz, Martin Gütlein, Micha Rautenberg, David Vorgrimmler, Denis Gebele, Christoph Helma
2013 Frontiers in Pharmacology  
lazar (lazy structure-activity relationships) is a modular framework for predictive toxicology.  ...  Model developers can choose between a large variety of algorithms for descriptor calculation and selection, chemical similarity indices, and model building.  ...  ACKNOWLEDGMENTS Financial support for the lazar version presented in this document was provided by the EU FP7 Projects OpenTox (Project Reference: 200787), ToxBank (Project Reference: 267042), and ModNanoTox  ... 
doi:10.3389/fphar.2013.00038 pmid:23761761 pmcid:PMC3669891 fatcat:x6gcytdetrhhhlfybjfuq6r3hu
« Previous Showing results 1 — 15 out of 128 results