Add more help on explain predictions

inst/help/mlClassificationBoosting.md

@@ -19,7 +19,7 @@ Boosting works by sequentially adding features to an decision tree ensemble, eac
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -91,6 +91,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlClassificationKnn.md

@@ -19,7 +19,7 @@ K-nearest neighbors is a method of classification that looks at the *k* number o
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -89,6 +89,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlClassificationLda.md

@@ -22,7 +22,7 @@ Linear Discriminant Analysis (LDA) is a method of classification that aims to fi
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Coefficients: Shows the coefficients for the linear discriminants. 
 - Prior and posterior probabilities: Shows the prior and posterior group probabilities. Prior probabilities are the proportions in the training set.
 - Class means training data: Shows the means on every variable for every class in the training data.
@@ -80,6 +80,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlClassificationNeuralNetwork.md

@@ -19,7 +19,7 @@ Feedforward neural networks are predictive algorithms inspired by the biological
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Network weights: Shows the connections in the neural network together with their weights.
 
 #### Plots
@@ -108,6 +108,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 ---

inst/help/mlClassificationRandomForest.md

@@ -19,7 +19,7 @@ Random Forest is a method of classification that creates a set of decision trees
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -89,6 +89,7 @@ Generates a new column in your dataset with the class labels of your classificat
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
 - Breiman, Leo. (2001). Random forests. Machine Learning. Springer, 45(1), 5-32
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionBoosting.md

@@ -17,7 +17,7 @@ Boosting works by sequentially adding features to an decision tree ensemble, eac
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -77,6 +77,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionKnn.md

@@ -17,7 +17,7 @@ K-nearest neighbors is a method of regression that looks at the *k* number of fe
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -75,6 +75,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionLinear.md

@@ -13,7 +13,7 @@ Linear Regression
 
 #### Tables  
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Regression coefficients: Shows a table of the regression coefficients.
 
 #### Plots
@@ -53,6 +53,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionNeuralNetwork.md

@@ -17,7 +17,7 @@ Feedforward neural networks are predictive algorithms inspired by the biological
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Network weights: Shows the connections in the neural network together with their weights.
 
 #### Plots
@@ -93,6 +93,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionRandomForest.md

@@ -17,7 +17,7 @@ Random Forest is a method of regression that creates a set of decision trees tha
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 
 #### Plots
 - Data split: Shows how the data is split into training (and validation), and test set.
@@ -75,6 +75,7 @@ Generates a new column in your dataset with the values of your regression result
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
 - Breiman, Leo. (2001). Random forests. Machine Learning. Springer, 45(1), 5-32
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlRegressionRegularized.md

@@ -19,7 +19,7 @@ Regularized linear regression is an adaptation of linear regression in which the
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Regression coefficients: Gives the regression coefficient for each feature.
 
 #### Plots
@@ -72,6 +72,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlclassificationdecisiontree.md

@@ -19,7 +19,7 @@ Decision Trees is a supervised learning algorithm that uses a decision tree as a
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Splits in tree: Shows the split variables, their split point, and the number of observations (which are not missing and are of positive weight) sent left or right by the split. It also shows the improvement in deviance given by this split.
 
 #### Plots
@@ -77,6 +77,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlclassificationsvm.md

@@ -19,7 +19,7 @@ Support Vector Machines is a supervised learning algorithm that maps training ex
 - Model performance: Shows commonly used classification evaluation metrics like precision, recall, the F1-score, support and AUC (area under the ROC curve).
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Support vectors: Shows a table containing the data (points) indicated as support vectors by the algorithm.
 
 #### Plots
@@ -76,6 +76,7 @@ Generates a new column in your dataset with the class labels of your classificat
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 --- 

inst/help/mlregressiondecisiontree.md

@@ -17,7 +17,7 @@ Decision Trees is a supervised learning algorithm that uses a decision tree as a
 - Model performance: Shows commonly used classification evaluation metrics like mean squared error (MSE), root mean squared error (RMSE) and R<sup>2</sup>.
 - Feature importance: Shows the available feature importance metrics for the fitted model.
   - Permutations: Sets the number of permutations on which the mean dropout loss is based.
-- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables.
+- Explain predictions: Shows the decomposition of the model’s prediction into contributions that can be attributed to different explanatory variables. This feature uses the breakdown algoritm from the `ibreakdown` R package. For more details about this method, see Gosiewska and Biecek (2019).
 - Splits in tree: Shows the split variables, their split point, and the number of observations (which are not missing and are of positive weight) sent left or right by the split. It also shows the improvement in deviance given by this split.
 
 #### Plots
@@ -64,6 +64,7 @@ Generates a new column in your dataset with the values of your regression result
 ### References
 -------
 - James, G., Witten, D., Hastie, T., & Tibshirani, R. (2013). An introduction to statistical learning. Springer New York.
+- Gosiewska, A. & Biecek, P. (2019). Do Not Trust Additive Explanations. <i>ArXiv</i>. https://doi.org/10.48550/arXiv.1903.11420
 
 ### R-packages 
 ---