ai.libs.jaicore.ml.tsc.distances

Class DynamicTimeWarping

java.lang.Object

ai.libs.jaicore.ml.tsc.distances.DynamicTimeWarping

All Implemented Interfaces:

ITimeSeriesDistance

public class DynamicTimeWarping
extends java.lang.Object
implements ITimeSeriesDistance

Implementation of the Dynamic Time Warping (DTW) measure as published in "Using Dynamic Time Warping to FindPatterns in Time Series" Donald J. Berndt and James Clifford. In DTW the time series are "warped" non-linearly in the time dimension to determine a measure of their similarity independent of certain non-linear variations in the time dimension. Given two time series A and B the dynamic programming formulation is based on the following recurrent definition: gamma(i,j) = delta(i,j) + min {gamma(i-1, j), gamma(i-1,j-1), gamma(i, j-1)} where gamma(i,j) is the cummulative distance up to i,j and delta(i,j) is the point distance between A_i and B_i.

Constructor Summary

Constructors

Constructor and Description
DynamicTimeWarping() Creates an instance with absolute distance as point distance.
DynamicTimeWarping(IScalarDistance delta) Creates an instance with a given scalar distance measure.

Method Summary

Modifier and Type	Method and Description
double	distance(double[] a, double[] b) Calculates the distance between two time series.
double	distanceWithWindow(double[] a, double[] b, int w)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DynamicTimeWarping

public DynamicTimeWarping()

Creates an instance with absolute distance as point distance.

DynamicTimeWarping

public DynamicTimeWarping(IScalarDistance delta)

Creates an instance with a given scalar distance measure.

Parameters:

delta - Scalar distance measure.

Method Detail

distance

public double distance(double[] a,
                       double[] b)

Description copied from interface: ITimeSeriesDistance

Calculates the distance between two time series.

Specified by:

distance in interface ITimeSeriesDistance

Parameters:

a - First time series.

b - Second time series.

Returns:

Distance between the first and second time series.

distanceWithWindow

public double distanceWithWindow(double[] a,
                                 double[] b,
                                 int w)