I found a precision and recall report table like as below
Precision Recall
.470±.009 .934±.013
.239±.010 .610±.013
I need the guidelines for ±.009 and ±.013 etc. What ± means and how to calculate it? Guide me for my situation. Thanks in advance.
$\begingroup$
$\endgroup$
3
-
$\begingroup$ probably obtained with bootstrapping, can you share where you saw it? $\endgroup$– gunesCommented Mar 25, 2019 at 6:39
-
$\begingroup$ Please go through the link : researchgate.net/figure/… $\endgroup$– user1999109Commented Mar 28, 2019 at 10:06
-
$\begingroup$ Guide me for my situation.. Thanks.. $\endgroup$– user1999109Commented Mar 28, 2019 at 10:07
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
1
In the paper, Page 4, it says:
...The detected cysts were then classified with texture descriptors based on Gray Level Co-occurrence Matrix (GLCM) and Gabor filters and by using classifiers such as Naive-Bayes (NB), Support Vector Machine (SVM) or Random Forest (RF) with 10-fold cross validation. The comparative evaluation in [21] indicates that GLCM + SVM and Gabor + SVM have segmentation accuracies of 0.8293 and 0.8244, respectively...
They do not specifically talk about it anywhere else in the paper, but the precision/recall mean/std values in the table you've shown were most probably calculated via these validation folds, i.e. an array of 10 precision values, and mean/std of them for example.
-
$\begingroup$ Thank you very much for your valuable reply. I will check with sample data and let you know. Thanks $\endgroup$ Commented Mar 30, 2019 at 4:31