Neighborhood Components Analysis (NCA) is a popular method for learning a distance metric to be used within a k-nearest neighbors (kNN) classifier. A key assumption built into the model is that each point stochastically selects a single neighbor, which makes the model well-justified only for kNN with k = 1. However, kNN classifiers with k > 1 are more robust and usually preferred in practice. H...

KNN algorithm is a simple, effective, non-parametric classification, and has been widely used in text classification, pattern recognition, image and spatial classification. Research on improvements about KNN algorithm has broad application prospects and important scientific significance. Based on analysis about classic KNN and its improved algorithms, we find its over-reliance on the choice of ...

* U. Johansson and R. König are equal contributors to this paper. Abstract Standard kNN suffers from two major deficiencies, both related to the parameter k. First of all, it is well-known that the parameter value k is not only extremely important for the performance, but also very hard to estimate beforehand. In addition, the fact that k is a global constant, totally independent of the particu...

The k-Nearest Neighbor (kNN) classification approach is conceptually simple – yet widely applied since it often performs well in practical applications. However, using a global constant k does not always provide an optimal solution, e. g., for datasets with an irregular density distribution of data points. This paper proposes an adaptive kNN classifier where k is chosen dynamically for each ins...

The k-nearest neighbor (kNN) is one of the simplest classification methods used in machine learning. Since the main component of kNN is a distance metric, kernelization of kNN is possible. In this paper kNN and semi-supervised kNN algorithms are empirically compared on two data sets (the USPS data set and a subset of the Reuters-21578 text categorization corpus). We use a soft version of the kN...