Bidirectional Action Rule Learning

Action rules specify recommendations which should be followed in order to transfer objects to the desired decision class. In this paper influence of employing information contained in source and target class examples in sequential covering based action rule induction method is examined. Results show that using source class for guiding the induction process produces best results.

[1]  Agnieszka Dardzinska.  Action Rules Mining , 2013, Studies in Computational Intelligence.

[2]  Marcin Blachnik,et al.  Instance Selection for Classifier Performance Estimation in Meta Learning , 2017, Entropy.

[3]  Salvatore Greco,et al.  Measuring expected effects of interventions based on decision rules , 2005, J. Exp. Theor. Artif. Intell..

[4]  Ayman Hajja,et al.  Reduction of Readmissions to Hospitals Based on Actionable Knowledge Discovery and Personalization , 2015, BDAS.

[5]  Salvatore Greco,et al.  Measuring Attractiveness of Rules from the Viewpoint of Knowledge Representation, Prediction and Efficiency of Intervention , 2005, AWIC.

[6]  Marcin Michalak,et al.  Rule Quality Measures Settings in Classification, Regression and Survival Rule Induction - an Empirical Approach , 2016, Fundam. Informaticae.

[7]  Zdzislaw Pawlak,et al.  Information systems theoretical foundations , 1981, Inf. Syst..

[8]  Zbigniew W. Ras,et al.  Action Rule Extraction from a Decision Table: ARED , 2008, ISMIS.

[9]  Ke Wang,et al.  Mining Actionable Patterns by Role Models , 2006, 22nd International Conference on Data Engineering (ICDE'06).

[10]  Johannes Fürnkranz,et al.  Foundations of Rule Learning , 2012, Cognitive Technologies.

[11]  Zbigniew W. Ras,et al.  Mining for interesting action rules , 2005, IEEE/WIC/ACM International Conference on Intelligent Agent Technology.

[12]  Urszula Stanczyk,et al.  On Combining Discretisation Parameters and Attribute Ranking for Selection of Decision Rules , 2017, IJCRS.

[13]  Ayman Hajja,et al.  Hierarchical object-driven action rules , 2013, Journal of Intelligent Information Systems.

[14]  Zbigniew W. Ras,et al.  Mining Surgical Meta-actions Effects with Variable Diagnoses' Number , 2014, ISMIS.

[15]  Véronique Masson,et al.  Building actions from classification rules , 2012, Knowledge and Information Systems.

[16]  Zbigniew W. Ras,et al.  Action-Rules: How to Increase Profit of a Company , 2000, PKDD.

[17]  Weidong Huang,et al.  Method for Discovering Actionable Rule , 2007, Fourth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD 2007).