A SURVEY OF COLLABORATIVE FILTERING BASED ON NEAREST-NEIGHBORS

Abstract

This paper explains the k-NN classification algorithm and its operator in RapidMiner. The Use Case of this chapter applies the k-NN operator on the Teacher Evaluation dataset. The operators explained in this chapter are: Read URL, Rename, Numerical to Binominal, Numerical to Polynominal, Set Role, Split Validation, Apply Model, and Performance. The k-Nearest Neighbor algorithm is based on learning by analogy, that is, by comparing a given test example with the training examples that are similar to it. The training examples are described by n attributes. Each example represents a point in an n-dimensional space. In this way, all of the training examples are stored in an n-dimensional pattern space. When given an unknown example, the k-nearest neighbor algorithm searches the pattern space for the k training examples that are closest to the unknown example. These k training examples are the k “nearest neighbors” of the unknown example. The “Closeness” is defined in terms of a distance metric, such as the Euclidean distance.