Unsupervised generation of data mining features from linked open data

By Heiko Paulheim, Johannes Fürnkranz. 2012

197 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Molecular Biology

Citations in the corpus

Modeling Relational Data with Graph Convolutional Networks (2017, background): “2011; de Vries and de Rooij 2015), and hand-designed feature extractors (Feat) (Paulheim and Fümkranz 2012).”
Schema-summarization in linked-data-based feature selection for recommender systems (2017, methodology): “The pre-pocessing step has been done following [20], we fixed two thresholds: one for the missing values pm = 99% and one for the distinct values pd = 98.”
Substructure counting graph kernels for machine learning from RDF data (2015, methodology): “However, they do not use kernel methods and use Linked Data as background knowledge to enhance data mining on ‘normal’ data [29].”
Knowledge graph refinement: A survey of approaches and evaluation methods (2016, methodology): “For internal methods, the features used for classification are usually the relations which connect an entity to other entities [80,86], i.”
Semantic Web in data mining and knowledge discovery: A comprehensive survey (2016, methodology): “Furthermore, elementary numerical or nominal features (such as the population of a city or the production studio of a movie) can be added [95].”
Deep feature synthesis: Towards automating data science endeavors (2015, background): “This work, and other related systems [16], are limited to datasets that are structured in knowledge bases or can be mined for entities that are in knowledge bases.”
RDF2Vec: RDF Graph Embeddings for Data Mining (2016, methodology): “FeGeLOD [20] and its successor, the RapidMiner Linked Open Data Exten-”
Recursive Feature Generation for Knowledge-based Learning (2018, background): “In many real-world applications, however, the given set of features is insufﬁcient for inducing a high quality classiﬁer (Levi & Weiss, 2004; Paulheim & F¨umkranz, 2012).”
Composition-based Multi-Relational Graph Convolutional Networks (2019, methodology): “Node and Graph Classiﬁcation: For node classiﬁcation, following Schlichtkrull et al. (2017), we compare with Feat (Paulheim & Fümkranz, 2012), WL (Shervashidze et al., 2011), and RDF2Vec (Ristoski & Paulheim, 2016).”
Relational Graph Attention Networks (2018, methodology): “Table 2: (a) Entity classiﬁcation results accuracy (mean and standard deviation over 10 seeds) for FEAT (Paulheim and Fümkranz, 2012), WL (Shervashidze et al., 2011; de Vries and de Rooij, 2015), RDF2Vec (Ristoski and Paulheim, 2016) and RGCN (Schlichtkrull et al., 2018), and (mean and standard…”