Denna tjänst avvecklas 2026-01-19. Läs mer här (länk)

Denna tjänst avvecklas 2026-01-19. Läs mer här (länk)

moj.similarity.measures
Class CosineD_row

java.lang.Object
  moj.similarity.SimilarityMeasure
      moj.similarity.measures.CosineD_row

public class CosineD_row
extends SimilarityMeasure

COSINE DISTANCE MEASURE
Originally written by P. Cejchan, 1999/08/13. Algorithm taken from: Carbonell, J.G.& al. 1997 "Translingual Information Retrieval: A comparative evaluation. IJCAI'97". See also Salton, G. 1989 "Automatic text processing: The transformation, Analysis, and retrieval of information by computer". Addison-Wesley, Reading, Pennsylvania. (Jongman, et. al., 1995, page 178)--"More emphasis is given to qualitative aspects by not considering a site as point but as a vector. Understandably, the direction of this vector tells us something about the relative abundances of species. The similarity of two sites can be expressed as some function of the angle between the vector of these sites. Quite common is the use of the cosine (or Ochiai coefficient): cos=OS=sigma(k)Y(ki)Y(kj)/sqrt{[sigma(k)(Y(ki)^2)][sigma(k)(Y(kj))^2)]}" <-- this is obviously for distance between columns, not rows (P.C).

Version:

2004-Sep-03

Author:

Martin Hassel

Constructor Summary

CosineD_row()

Method Summary

double[][]

analyzeMatrix(float[][] data_matrix)

Methods inherited from class java.lang.Object

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

Constructor Detail

CosineD_row

public CosineD_row()

Method Detail

analyzeMatrix

public double[][] analyzeMatrix(float[][] data_matrix)

Specified by:

analyzeMatrix in class SimilarityMeasure

Parameters:

data_matrix - an array of float arrays denoting the rows in the matrix that are to be compared for similarity.

Returns:

an array of double arrays containing the similarities between all rows