Prototype Selection for Nearest Neighbor Classification: Taxonomy and Empirical Study

By Salvador Garcia, Joaquín Derrac, José Ramón Cano, Francisco Herrera. 2012

721 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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 Computer Vision and Pattern Recognition

Citations in the corpus

Multi-class texture analysis in colorectal cancer histology (2016, background): “The Euclidean-distance 1-nearest neighbour (1-NN) is a very simple classifier that is independent of tuning parameters, is easy to implement and has a low risk of overfitting 29 .”
Big data preprocessing: methods and prospects (2016, background): “Thus, instance selection is comprised by a series of techniques that must be able to choose a subset of data that can replace the original data set and also being able to fulfill the goal of a data mining application [33, 34].”
EUSBoost: Enhancing ensembles for highly imbalanced data-sets by evolutionary undersampling (2013, background): “Prototype selection [20] is a sampling process which, instead of aiming to balance the class distributions (as the ones shown in Section 2.3), reduces the reference set for the nearest neighbor (1NN) classi ﬁ er in order to improve its accuracy and reduce the storage necessity.”
k-Nearest Neighbour Classifiers - A Tutorial (2020, background): “It is known as Instance Selection or Prototype Selection by those who focus on nearest neighbour classification [20] and Case-base Editing or Case-base Maintenance by the CBR community [34].”
Tree Space Prototypes: Another Look at Making Tree Ensembles Interpretable (2016, methodology): “We point the reader to the review by [12] that suggested that prototype selection methods can be grouped into three categories: condensation [19], edition [49], or hybrid methods that remove both noisy and redundant points from the prototype selection set.”
Teacher Improves Learning by Selecting a Training Subset (2018, background): “The first thread is training set reduction [19, 43, 42], which during training time prunes items in S in an attempt to improve the learned model.”
Rethinking Semantic Segmentation: A Prototype View (2022, methodology): “With the unified view of parametric prototype learning, a few intrinsic yet long ignored issues in this field unfold: First, prototype selection [36] is a vital aspect in the design of a prototype based learner – prototypes should be typical for their classes.”
An efficient instance selection algorithm for k nearest neighbor regression (2017, methodology): “According to the type of selected instances, the existing instance selection algorithms for kNN classification are classified into three categories: condensation algorithms, edition algorithms and hybrid algorithms [3].”
Soft-Label Dataset Distillation and Text Dataset Distillation (2019, methodology): “In the context of nearest-neighbor classification, prototype selection (PS) and prototype generation (PG) are studied extensively as methods of reducing storage requirements and improving the efficiency of nearest-neighbor classification [17], [18].”
On the characterization of noise filters for self-training semi-supervised in nearest neighbor classification (2014, methodology): “We call local methods to those techniques in which the removal decision is based on a local neighborhood of instances [30] [31].”