Advances in Instance Selection for Instance-Based Learning Algorithms

By Henry Brighton, C. Mellish. Data mining and knowledge discovery, 2002

600 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/b955092c92fd4720108dfb2414772753d8f7b130

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

Citations in the corpus

Feature Selection: A literature Review (2014, background): “In instance selection, sampling methods have been developed to search for a set of instances that can perform in a focused way [127, 128].”
An instance level analysis of data complexity (2014, background): “Border points and instances that belong to a minority class have also been found to be more difficult to classify correctly (Brighton and Mellish 2002; van Hulse et al. 2007).”
Toward integrating feature selection algorithms for classification and clustering (2005, methodology): “In addition to sampling methods [33], a suite of methods have been developed to search for representative instances so that data mining is performed in a focused and direct way [11], [61], [76].”
k-Nearest Neighbour Classifiers - A Tutorial (2020, methodology): “Instance selection techniques can be categorised as competence preservation or competence enhancement techniques[10].”
Machine learning: a review of classification and combining techniques (2006, background): “Brighton and Mellish (2002) found that their ICF algorithm and RT3 algorithm (Wilson and Martinez 2000) achieved the highest degree of instance set reduction as well as the retention of classification accuracy: they are close to achieving unintrusive storage reduction.”
On Issues of Instance Selection (2002, methodology): “We can find three specific techniques (Madigan et al., 2002; Domingo et al., 2002; Brighton and Mellish, 2002) vs. a general framework LIU AND MOTODA (Reinartz, 2002); an on-line (incremental) approach (Domingo et al., 2002) vs. off-line (batch) ones (Brighton and Mellish, 2002; Madigan et al.,…”
Class Conditional Nearest Neighbor and Large Margin Instance Selection (2008, methodology): “In order to tackle these problems, various algorithms have been developed, such as those for instance/prototype selection [1], [2], [3], [8], [19], [22], [31], for feature selection [20], [25], [36], and for distance learning [29], [39], [40].”
Review of Deep Learning Algorithms and Architectures (2019, methodology): “Instead of training the network with all the data set, we can preselect a smaller but representative data set from the full training distribution set using instance selection methods [95] or Monte Carlo sampling [48].”
Educational data mining: A survey and a data mining-based analysis of recent works (2014, methodology): “ID Author of the EDM approach Discipline Model Task M: method A: algorithm T: technique E: equation F: frame 138 Maull, Saldivar, and Sumner (2010a) Machine learning,, probability Descriptive Clustering M: IBL,Bayes theorem A: k-means, EM 139 Vuong et al. (2011) Statistic Descriptive Functional depe”
k-Nearest Neighbour Classifiers (2007, methodology): “[3] evaluated their ICF against RT3 and found that neither algorithm consistently out performed the other and both represented the “cutting edge in instance set reduction techniques”.”
Optimal instance selection for improved decision tree (2007, methodology): “ICF Brighton and Mellish (2002) Iterative Case Filtering: ICF tries to select the instances which classify more prototypes correctly.”
Concept drift detection via competence models (2014, background): “This is important because suspicious cases are more likely to behave like noisy or novel cases [57].”
Comparison of Instances Seletion Algorithms I. Algorithms Survey (2004, methodology): “Iterative Case Filtering (ICF) was proposed by Brighton & Mellish in [10].”
Supervised Machine Learning: A Review of Classification Techniques (2007, neutral): “Brighton & Mellish (2002) found that their ICF algorithm and RT3 algorithm (Wilson & Martinez, 2000) achieved the highest degree of instance set reduction as well as the retention of classification accuracy: they are close to achieving unintrusive storage reduction.”