Scribd
Upload
13 views

Types of Algorithm That Can Be Used in Machine Learning

machine learning

Uploaded by

Chet Palomares
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
13 views

Types of Algorithm That Can Be Used in Machine Learning

machine learning

Uploaded by

Chet Palomares
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Types Of algorithm that can be used in Machine learning

K nearest neighbor source:


http://www.saedsayad.com/k_nearest_neighbors.htm
K nearest neighbor is an algorithm that stores all available cases and classifies new cases based on the
similarity of measures.
Algorithm
A case is classified by its neighbors. One of the techniques that can be analyzed most was the k nearest
neighbor measured by a distance function. If K = 1, then the case is simply assigned to the appropriate
nearest neighbor. The first sets of equation are valid only for continuous variables.

Euclidean:

( x i y i )2
i=1

Manhattan:
k

|x i yi|
i=1

Minkowski:
k

1
q q

( (|xi y i|) )
i=1

For an instance of discrete variables the Hamming Distance equation must be used. Hamming Distance
equation standardizes the numerical variable between 0 and 1 when there is a combination of numerical
and categorical variables in data set.

Hamming Distance:
k

D H = |x i y i|
i=1

x= y D=0
x y D=1

SOM (Self Organizing Maps) John A. Bullinaria, 2004; Introduction to Neural Networks :
Lecture 17
SOM system also known as Konohen Network has a feed-forward structure with a single computational
layer of neurons arranged in rows and columns. Each neuron is fully connected to all the source units in
the input layer:
G9

One Dimensional Map will just have a single row or column in the computational layer.
Algorithm
The stages of the SOM algorithm that achieves this can be summarized as follows:
1. Initialization Choose random values for the initial weight vectors

wj

2. Sampling Draw a sample training input vector x from the input space.
3. Matching find the winning neuron I(x) that has weight vector closest to the input vector
4. Updating Apply the weight update equation

wij =n ( t ) T j ,l ( x ) ( t )( x iwij ) where

T j ,l ( x ) ( t ) is a Gaussian neighborhood and n ( t ) is the learning rate.


5. Continuation keep returning to step 2 until the feature map stops changing.

You might also like