Other packages > Find by keyword >

shapley  

Weighted Mean SHAP for Feature Selection in ML Grid and Ensemble
View on CRAN: Click here


Download and install shapley package within the R console
Install from CRAN:
install.packages("shapley")

Install from Github:
library("remotes")
install_github("cran/shapley")

Install by package version:
library("remotes")
install_version("shapley", "0.4")



Attach the package and use:
library("shapley")
Maintained by
E. F. Haghish
[Scholar Profile | Author Map]
First Published: 2023-11-07
Latest Update: 2023-11-07
Description:
This R package introduces an innovative method for calculating SHapley Additive exPlanations (SHAP) values for a grid of fine-tuned base-learner machine learning models as well as stacked ensembles, a method not previously available due to the common reliance on single best-performing models. By integrating the weighted mean SHAP values from individual base-learners comprising the ensemble or individual base-learners in a tuning grid search, the package weights SHAP contributions according to each model's performance, assessed by the Area Under the Precision-Recall Curve (AUCPR) for binary classifiers (currently implemented). It further extends this framework to implement weighted confidence intervals for weighted mean SHAP values, offering a more comprehensive and robust feature importance evaluation over a grid of machine learning models, instead of solely computing SHAP values for the best-performing model. This methodology is particularly beneficial for addressing the severe class imbalance (class rarity) problem by providing a transparent, generalized measure of feature importance that mitigates the risk of reporting SHAP values for an overfitted or biased model and maintains robustness under severe class imbalance, where there is no universal criteria of identifying the absolute best model. Furthermore, the package implements hypothesis testing to ascertain the statistical significance of SHAP values for individual features, as well as comparative significance testing of SHAP contributions between features. Additionally, it tackles a critical gap in feature selection literature by presenting criteria for the automatic feature selection of the most important features across a grid of models or stacked ensembles, eliminating the need for arbitrary determination of the number of top features to be extracted. This utility is invaluable for researchers analyzing feature significance, particularly within severely imbalanced outcomes where conventional methods fall short. In addition, it is also expected to report democratic feature importance across a grid of models, resulting in a more comprehensive and generalizable feature selection. The package further implements a novel method for visualizing SHAP values both at subject level and feature level as well as a plot for feature selection based on the weighted mean SHAP ratios.
How to cite:
E. F. Haghish (2023). shapley: Weighted Mean SHAP for Feature Selection in ML Grid and Ensemble. R package version 0.4, https://cran.r-project.org/web/packages/shapley. Accessed 29 Mar. 2025.
Previous versions and publish date:
0.1 (2023-11-07 20:00), 0.3 (2024-05-30 09:00), 0.4 (2024-10-23 05:40)
Other packages that cited shapley R package
View shapley citation profile
Other R packages that shapley depends, imports, suggests or enhances
Complete documentation for shapley
Functions, R codes and Examples using the shapley R package
Some associated functions: h2o.get_ids . normalize . shapley . shapley.plot . shapley.test . shapley.top . test . 
Some associated R codes: h2o.get_ids.R . normalize.R . shapley.R . shapley.plot.R . shapley.test.R . shapley.top.R . test.R .  Full shapley package functions and examples
Downloads during the last 30 days
02/2702/2803/0103/0203/0303/0403/0503/0603/0703/0803/0903/1003/1103/1203/1303/1403/1503/1603/1703/1803/1903/2003/2103/2203/2303/2403/2503/2603/27Downloads for shapley051015202530354045505560TrendBars

Today's Hot Picks in Authors and Packages

gglgbtq  
Show Pride on 'ggplot2' Plots
Provides multiple palettes based on pride flags with tailored themes. ...
Download / Learn more Package Citations See dependency  
landmix  
Landmark Prediction for Mixture Data
Non-parametric prediction of survival outcomes for mixture data that incorporates covariates and a l ...
Download / Learn more Package Citations See dependency  
quickcode  
Quick and Essential 'R' Tricks for Better Scripts
The NOT functions, 'R' tricks and a compilation of some simple quick plus often used 'R' codes to im ...
Download / Learn more Package Citations See dependency  

23,842

R Packages

207,311

Dependencies

64,420

Author Associations

23,781

Publication Badges

© Copyright since 2022. All right reserved, rpkg.net.  Based in Cambridge, Massachusetts, USA