A Weighted Nearest Neighbor Algorithm for Learning with Symbolic Features

By Scott Cost, Steven L. Salzberg. 2004

531 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Molecular Biology, Model Architectures

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 Molecular Biology

In Model Architectures

Citations in the corpus

Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs (2016, background): “The maximum layer in which an element is present is selected randomly with an exponentially decaying probability distribution.”
SMOTE: Synthetic Minority Over-sampling Technique (2002, methodology): “…vectors is given by: ∆( X, Y ) = w x w y N i =1 δ ( x i , y i ) r (2) r = 1 yields the Manhattan distance, and r = 2 yields the Euclidean distance (Cost & Salzberg, 1993). w x and w y are the exemplar weights in the modiﬁed VDM. w y = 1 for a new example (feature vector), and w x is the bias…”
SMOTEBoost: Improving Prediction of the Minority Class in Boosting (2003, methodology): “In the nearest neighbor computations for the minority classes we use Euclidean distance for the continuous features and the Value Distance Metric (with the Euclidean assumption) for the nominal features [5, 25, 26].”
Top 10 algorithms in data mining (2007, methodology): “The PEBLS system by by Cost and Salzberg [14] is a well known example of such an approach.”
Feature Selection with Ensembles, Artificial Variables, and Redundancy Elimination (2009, methodology): “For example, the widely-used value difference metric (VDM) (Stanfill and Waltz, 1986) and its modified version (MVDM) (Cost and Salzberg, 1993) consider the conditional probability distribution of the respon e at a predictor value.”
A Comprehensive Survey and Experimental Comparison of Graph-Based Approximate Nearest Neighbor Search (2021, background): “Nearest Neighbor Search (NNS) is a fundamental building block in various application domains [7, 8, 38, 67, 70, 80, 108, 117], such as information retrieval [34, 118], pattern recognition [29, 57], data mining [44, 47], machine learning [24, 28], and recommendation systems [69, 82].”
Simlarity Search for Web Services (2004, background): “Most solutions to this problem (e.g. [10] [20] [27] [19]) are based on term frequency analysis.”
Using knowledge-based neural networks to improve algorithms: Refining the Chou-Fasman algorithm for protein folding (1992, methodology): “The protein-folding problem is also of interest because a number of machine learning techniques are currently being applied to this problem, including neural networks (Holley & Karplus, 1989; Qian & Sejnowski, 1988), inductive logic programming (Muggleton & Feng, 1991), case-based reasoning ( Cost &”
A Taxonomy of Recommender Agents on the Internet (2003, methodology): “PEBLS (Cost and Salzberg 1993) is a nearest neighbor algorithm which makes use of a modiﬁcation of the value difference metric (MVDM) for computing the distance between two examples.”
Combating imbalance in network intrusion datasets (2006, methodology): “We use the modification of Value Distance Metric (VDM) [21] to compute the nearest neighbors for the nominal valued features.”
Data Mining for Imbalanced Datasets: An Overview (2005, methodology): “We use the modification of Value Distance Metric (VDM) (Cost and Salzberg, 1993) to compute the nearest neighbors for the nominal valued features.”