DESCRIPTION

---

Class for a K-nearest neighbours (Knn) classifier. It is a reasoning model that predicts the class to which belongs a given document based on the distance between its neighbours(instances).

PARAMETERS

---

Classifier Name: Element identifier for operations and Clipboard.

Neighbours number: The number of neighbours to use.

Cross validation: Permits to select the best k value to identify the class to which belongs a given document based on the hold-one-out algorithm.

Distance type: Gets the distance weighting method used to calculate the distance between neighbours. Allowed values: None, 1/distance (Inverse), 1-Distance (reverse).

REFERENCES

---

[1] D. Aha, D. Kibler (1991). Instance-based learning algorithms. Machine Learning. 6:37-66.