From f2e81162a19ba31b4b1d6c5fc9199b250cbb2137 Mon Sep 17 00:00:00 2001
From: Alexandre Croix <croix.alexandre@gmail.com>
Date: Thu, 23 May 2019 15:11:56 +0200
Subject: [PATCH] Update README.md
---
README.md | 15 +++++++++++----
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/README.md b/README.md
index 8e03f7e..7cab05e 100644
--- a/README.md
+++ b/README.md
@@ -80,6 +80,7 @@ SolutionDistance{
The **run** method returns a solution object, consisting of p weights and w weights to use with the WOWA operator, plus the total distance between the expected aggregated values that are given as parameter, and the aggregated values computed by WOWA using these weights.
+The method **run** can be used with ArrayList as the above example or with file name. One of these json file names contains the data and the second contains the expected results.
### Parameters description
- **population_size** : size of the population in the algorithm. Suggested value : 100
@@ -156,17 +157,23 @@ For each tested fold, the Area Under the Curve is also computed to evaluate the
The code above produces a result similar to:
```
-SolutionDistance{weights_w=[0.8673383311511217, 0.04564604584006219, 0.0647437341741078, 0.022271888834708403],
+SolutionDistance{
+weights_w=[0.8673383311511217, 0.04564604584006219, 0.0647437341741078, 0.022271888834708403],
weights_p=[0.5933035227430291, 0.10784413855996985, 0.03387258778518031, 0.26497975091182074],
-fitness score=2.2260299633096268}=0.16666666666666666
-SolutionDistance{weights_w=[0.7832984118592771, 0.12307744745817546, 0.07982187970335382, 0.013802260979193624],
+fitness score=2.2260299633096268}=
+0.16666666666666666
+
+SolutionDistance{
+weights_w=[0.7832984118592771, 0.12307744745817546, 0.07982187970335382, 0.013802260979193624],
weights_p=[0.01945033161182157, 0.3466399858254755, 0.18834296208558235, 0.44556672047712065],
-fitness score=1.7056044468736795}=0.4166666666666667
+fitness score=1.7056044468736795}=
+0.4166666666666667
```
As output, the method **runKFold** return a HashMap that contains the best solution for each fold and the AUC corresponding to this solution.
The method **runKFold** takes as argument the dataset (data and expected result) the number of folds used in the cross validation and a value that can increase the number of alert is this number is to low.
This method is interesting to increase the penalty to do not detect an alert.
+As for a classical learning, the method **runKFold** can be used as the example above or with json files. In this case, the arguments are String that are the file names.
## References
- [The WOWA operator : a review (V. Torra)](https://gitlab.cylab.be/cylab/wowa-training/raw/c3c3785c767ab8258df0fc585aec1e8d463851cd/doc/Torra%20-%202011%20-%20The%20WOWA%20Operator%20A%20Review.1007_978-3.pdf)
--
GitLab