README.md
In kimbrianj/mlforsocialscience: Machine Learning for Social Sciences

Machine Learning for Social Science Tutorials

R package containing material for SURV 699: Machine Learning for Social Sciences

You can install the mlforsocialscience package using install_github within devtools.

devtools::install_github('kimbrianj/mlforsocialscience')

You can launch the tutorials using the "Tutorial" pane in RStudio, typically in the top right next to "Environment", "History", and "Connections". Navigate to the tutorial you want to start, then click on "Start Tutorial". To open a new tutorial, click on the stop sign at the top.

List of Tutorials Available: - 'introduction': Provides an introduction to R and RStudio. - 'ml-basics': Uses ordinary least squares regression to show some basics of how ML might be done in R. - 'knn': Shows how to do k-Nearest Neighbors in R, as well as discussing performance metrics. - 'regularized-regression-1': Lasso and Ridge regression using R. - 'regularized-regression-2': Further discussion of regularized regression, including elastic net and group lasso. - 'trees-1': Goes over the basics of implementation of Decision Trees in R. - 'trees-2': Goes over CTREE and Model-Based Recursive Partitioning in R. - 'ensemble': Shows the basics of using ensemble methods with tree-based methods. - 'interpretable-ml': Demonstrates how to implement interpretable ML techniques, such as PDP plots. - 'boosting-1': Applications of boosting with AdaBoost and Gradient Boosting - 'boosting-2': More boosting, with XGBoosting and Model-based Boosting. - 'ml-toolbox': Shows useful techniques in ML. - 'unsupervised-learning': Simple examples of implementing Latent Dirichlet Allocation for text analysis.