How can I classify data with the nearest-neighbor algorithm using Python?

Question

I need to classify some data with (I hope) nearest-neighbor algorithm. I've googled this problem and found a lot of libraries (including PyML, mlPy, and Orange), but I'm unsure of where to start here.

 

How should I go about implementing k-NN using Python?

Answer 1

You can use Scikit learn for k nearest neighbours (KNN). 

KNN algorithm is used for both regression (returns a score) and classification (returns a class label).

Using the scikits.learn k-nearest neighbor module:
 

>>> import numpy as NP

>>> from sklearn import neighbors as kNN

>>> from sklearn import datasets

>>> iris = datasets.load_iris()

>>> data = iris.data

>>> class_labels = iris.target

>>> kNN1 = kNN.NeighborsClassifier()

>>> kNN1.fit(data, class_labels)

      NeighborsClassifier(n_neighbors=5, leaf_size=20, algorithm='auto')

 

K-nearest neighbors require an appropriate similarity metric (Euclidean distance). Scikits.learn includes modules comprised of various distance metrics as well as testing algorithms for the selection of the appropriate one.

Hope this answer helps.