Skip to main content
Tchotchke's user avatar
Tchotchke's user avatar
Tchotchke's user avatar
Tchotchke
  • Member for 9 years, 4 months
  • Last seen more than 4 years ago
13 votes
Accepted

Best practices for coding categorical features for Decision Trees?

8 votes
Accepted

Machine Learning Book (Python)

8 votes
Accepted

Choosing complexity parameter in CART

7 votes
Accepted

Find close pairs in very high dimensional space with sparse vectors

4 votes

Why ROC increasing with more variable addition in logistic regression?

4 votes
Accepted

How to compare 2 classifers using confusion matrix?

4 votes
Accepted

A priori parameters to Random Forest based on n and p

3 votes

Variable Importance of R Random Forests

3 votes

Open-sourced pairwise learning models

3 votes
Accepted

What is a reasonable model measurement for a binary classification on an imbalanced data set? Accuracy? ROC?

3 votes

K-Fold Cross Validation for mixed-effect models: how to score them?

3 votes

Making a single decision tree from a random forest

2 votes

Avoiding overfitting with linear regression trees

2 votes

RandomForestClassifier Parameter Optimization

1 vote

How to report confusion matrix of Random Forest classifier on test set using R?

1 vote

Comparing CV Predictions across Folds for Random Forest

1 vote

How to use external data?

1 vote

receiver operating characteristic (ROC) on a test set

1 vote

Double (nested,wrapper) CrossValidation - final trained model

1 vote

Model based approaches to content based recommenders. How does this work?

1 vote

Decision trees in smaller datasets

1 vote

Feature selection when using cross validation

1 vote

Rare event bias techniques

1 vote

How to avoid overfitting in stratified k-fold cross validation methods for classification?

1 vote

How to validate k-fold cross validation results for classifiers?

1 vote

Cascade SVM in R

0 votes

Selection Bias - Taking a subset of a larger database

0 votes
Accepted

General question on the analysis design

0 votes

splitting a training data set based on classifier accuracy

0 votes

How to work a data set with a huge number of features?