GUPIs: Global measures based on averages of indexes on the user and...
In ujaen-statistics/ThemAAs:

Description Usage Arguments Details Value References Examples

View source: R/GUPIs.R

General function that groups global measures based on averages of indexes on the user and producer perspective. This function is made up of: Average accuracy from producer's perspective, the average accuracy is an average of the accuracy of individual categories. Average accuracy from user's perspective, the average accuracy is an average of the accuracy of individual categories. Because the individual categories can be the user's or the producer's accuracy, it can be computed in both ways accordingly. Double average of user's and producer's perspective, is the average of the average accuracy from user's and producer's perspective. Average of Hellden's mean accuracy index, is the arithmetic mean of Hellden's mean accuracy index. Average of Short's mapping accuracy index, is the arithmetic mean of Short's mapping accuracy index. Classification succes index, applies to all classes and gives an overall estimation of classification effectiveness. Combined accuracy from producer's perspective, is the average of the overall accuracy and average accuracy. Combined accuracy from user's perspective, is the average of the overall accuracy and average accuracy. Combined accuracy from both user's and producer's perspectives, is the arithmetic mean of the overall accuracy and average of Hellden's mean accuracy index

1	GUPIs(object)

object

a coMa object (confusion matrix object)

Average accuracy from producer's perspective, is the sum of the diagonal values between its corresponding marginal (by colums) divided by the number of categories. Average accuracy from user's perspective, is the sum of the diagonal values between its corresponding marginal (by rows) divided by the number of categories. Double average of user's and producer's perspective, is the arithmetic mean of the average accuracy from user's and producer's perspective. Classification succes index, is the sum of average accuracy from user's and producer's perspective less 1. Combined accuracy from producer's perspective, is the arithmetic mean of the overall accuracy and Average accuracy from producer's perspective. Combined accuracy from user's perspective, is the arithmetic mean of the overall accuracy and Average accuracy from user's perspective. Combined accuracy from both user's and producer's perspectives,

GUPIs returns a list with the following elements:

Fung, T., & LeDrew, E. (1988). The determination of optimal threshold levels for change detection using various accuracy indices. Photogrammetric Engineering and Remote Sensing, 54, 1449-1454.

Liu, Canran, Paul Frazier, y Lalit Kumar. (2007). Comparative Assessment of the Measures of Thematic Classification Accuracy. Remote Sensing of Environment 107 (4): 606-616.

Koukoulas, S., & Blackburn, G. A. (2001). Introducing new indices for accuracy evaluation of classified images representing semi-natural woodland environments. Photogrammetric Engineering and Remote Sensing, 67, 499-510.

#Let evaluate averages of indexes on the user and producer perspective.
## Confusion matrix included in Congalton and Green (2009), pg. 108.
x <- coMa(cbind(c(65,6,0,4),c(4,81,11,7),c(22,5,85,3),c(24,8,19,90)))
## Averages of indexes on the user and producer perspective
Indexes <- GUPIs(x)