Estimating Attributes: Analysis and Extensions of RELIEF

By Igor Kononenko. 1994

How researchers see this paper

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

Citations in the corpus

Benchmark and Survey of Automated Machine Learning Frameworks (2019, methodology): “Heuristic approaches follow an iterative procedure by adding single features (Kononenko, 1994).”
Filter Methods for Feature Selection - A Comparative Study (2007, unknown): “5 (??) 90 100 - 3,6,8,11,14,17,18 [1-11],[14-21],26,39,44,60,100 ,104,112,164,172,180,191”
Findings of the 2013 Workshop on Statistical Machine Translation (2013, methodology): “For English-Spanish, LogReg was used with L2 Regularisation (Lin et al., 2007) and two feature sets were devised after scoring features with ReliefF (Kononenko, 1994) and Information Gain (Hunt et al., 1966).”
Recent advances and emerging challenges of feature selection in the context of big data (2015, unknown): “ReliefF Multivariate Ranker Kononenko [25] n2m”
Computational Intelligence and Feature Selection - Rough and Fuzzy Approaches (2008, methodology): “The method has been extended to enable it to handle inconsistency, noise, and multi-class datasets [182].”
Searching for Interacting Features (2007, methodology): “Representative measures used for evaluating relevance include: distance measures [Kononenko, 1994; Robnik-Sikonja and Kononenko, 2003], information measures [Fleuret, 2004], and consistency measures [Almuallim and Dietterich, 1994], to name a few.”
Transfer Learning for EEG-Based Brain–Computer Interfaces: A Review of Progress Made Since 2016 (2020, unknown): “hannel are extracted. cTL ﬁrst computes the classiﬁcation accuracy by using the target subject’s data only, and performs transfer only if that accuracyis below the chance level. Then, it uses ReliefF [86] to select a few most emotion-relevantfeatures in the target domain, calculates their correlatio”
Feature Selection for Unsupervised Learning (2004, methodology): “These methods can be grouped as filter (Marill and Green, 1963; Narendra and Fukunaga, 1977; Almuallim and Dietterich, 1991; Kira and Rendell, 1992; Kononenko, 1994; Liu and Setiono, 1996; Cardie, 1993; Singh and Provan, 1995) or wrapper (John et al.”
Character Recognition Systems: A Guide for Students and Practitioners (2007, methodology): “Features were first evaluated by ReliefF [58], the extension of Relief [55].”
Conditional random fields for activity recognition (2007, background): “RELIEF is restricted to binary classification tasks, although others have introduced extended versions for multi-class problems [52].”