Predictive Analytics (with R)

Input File (.csv only):
Read File

Condition(s) for Subset:
None = > >= < <= != Categorical Variable
None = > >= < <= != Categorical Variable
None = > >= < <= != Categorical Variable
None = > >= < <= != Categorical Variable

Dependent Variable:
None Change to Categorical Variable

Independent Variable(s):
None

Size of Training Data:
From Top From Bottom

Size of Validation Data:
% of Training Data

Include Prediction for Validation Data with Testing Data

Regression Techniques:
Linear Regression
Lasso Regression (i.e., L1 Regularization)
Ridge Regression (i.e., L2 Regularization)
Elastic Net Regression with alpha =
Decision Tree (with All 1 2 3 4 5 6 7 8 Levels)
Random Forest with trees with maximum nodes
Gradient Boosting with maximum depth = , number of trees =
XGBoost with maximum depth = , number of trees =
LightGBM with maximum depth = , number of trees =
Support Vector Machine with kernel = Linear Polynomial Radial Basis Sigmoid , cost = , gamma =

Classification Techniques:
Linear Discriminant Analysis
Logistic Regression for binary classification with cut-off =
Decision Tree with All 1 2 3 4 5 6 7 8 Levels
Random Forest with trees with maximum nodes
Adaptive Boosting (AdaBoost) with number of trees =
Gradient Boosting for binary classification with cut-off = , maximum depth = , number of trees =
XGBoost for binary classification with cut-off = , maximum depth = , number of trees =
LightGBM for binary classification with cut-off = , maximum depth = , number of trees =
Support Vector Machine with kernel = Linear Polynomial Radial Basis Sigmoid , cost = , gamma =
Naive Bayes
k-Nearest Neighbor with k = (Note: numeric independent variables only)
Neural Network (with hidden layers)

Additional Regression Metrics:
Mean Squared Error (MSE)
Mean Absolute Error (MAE)

Additional Binary Classification Metrics:
Precision
Recall or Sensitivity
Specificity
F1 Score



Code:

​Copyright © 2019 Abhimanyu Gupta. All rights reserved.