Propositionalization approaches to relational data mining

By S. Kramer, Nada Lavrac, Peter A. Flach. 2001

336 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Information Systems

View Paper: https://www.semanticscholar.org/paper/5ae377db282d5b7ab971da802bf733d677a23020

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 Computational Theory and Mathematics

In Information Systems

Citations in the corpus

Tunneling Neural Perception and Logic Reasoning through Abductive Learning (2018, methodology): “Finally, the NLM transforms the locally consistent hypotheses into relational features using the propositionalization technique [24].”
Feature Construction Methods : A Survey (2009, background): “In subsequent literature, the problem of identifying good features with a first order representation has been covered under the propositionalization approaches [Kramer et al., 2000].”
The Synthetic Data Vault (2016, background): “In the past, several researchers have focused on statistically modeling data from a relational database for the purposes of feature engineering [1] or insight generation [2].”
FeatureHub: Towards Collaborative Data Science (2017, methodology): “In the databases community, a body of work has been developed for propositionalizing relational databases into “attribute-value” form for data-mining applications [17], [18].”
Learning Causal Models of Relational Domains (2010, methodology): “It was introduced as a way to continue to rely on existing algorithms instead of developing new ones to handle the increased complexity (Kramer, Lavrač, and Flach 2001).”
Fast relational learning using bottom clause propositionalization with artificial neural networks (2013, positive): “Finally, SINUS (Kramer et al., 2001) improved on DINUS by allowing unconstrained clauses, making use of language bias in the feature selection and verifying if it is possible to unify newly found literals with existing ones, while keeping consistency between pre-existing variable namings, thus…”
Link-Based Classification (2003, background): “2 IEssentially this is a propositionalization (Flach Lavrac, 2000; Kramer et al., 2001 ) of the aspects of the neighborhood of an object in the graph.”
Collective Classification in Network Data (2008, methodology): “Within the inductive logic programming community, aggregation has been studied as a means for propositionalizing a relational classification problem (Knobbe, deHaas, and Siebes 2001; Kramer, Lavrac, and Flach 2001).”
Knowledge Infusion (2006, background): “Then, in effect, a three-valued logic will be realized using two binary variables for each relation.”
From Local Patterns to Global Models: The LeGo Approach to Data Mining (2008, background): “Typical problems with large feature spaces include text mining [12], propositionalization approaches to relational learning [25], and others.”
Margin-based first-order rule learning (2007, background): “It can be configured to base its rule generation process on different kinds of information, thus bridging the gap between propositionalization (Kramer et al. 2001) and relational learning approaches that generate and evaluate discriminative features dynamically.”
Classification in Networked Data: A Toolkit and a Univariate Case Study (2007, neutral): “…are many more, for example from the rich literature of inductive logic programming (ILP, such as, De Raedt et al., 2001; Dzeroski and Lavrac, 2001; Kramer et al., 2001; Flach and Lachiche, 2004; Richardson and Domingos, 2006), or based on using relational database joins to generate relational…”
Collective Classi!cation in Network Data (2008, methodology): “Within the inductive logic programming community, aggregation has been studied as a means for propositionalizing a relational classification problem (Knobbe, deHaas, and Siebes 2001; Kramer, Lavrac, and Flach 2001).”
Substructure counting graph kernels for machine learning from RDF data (2015, result): “The framework for graph kernels for RDF that we define in this paper has some similarities to the comparison of propositionalization [27] strategies for Linked Data in [28], in the sense that some of the propositionalizations are similar to the graph substructures that our graph kernels work with.”