Data Science Drills

Test Yourself to Enhance Your Skills

Home
Contact Us

Random Forests – Missing Data

Leave a reply

Welcome to your Random Forests - Missing Data

After the initial guess for missing values in the training data, the missing value is then imputed based on the

Most frequent non-missing value where frequency is weighted by proximity

The missing values are given a new categorical value that codes for missing

Proximity used in Random Forest imputation is based on

How many features the two observations have in common

How often two observations are put into the same leaf node

Euclidean distance

What is the first step for scoring new observations with missing data?

Create copies of the data with the possible values for the target

Use a random forest to predict the missing values

Median imputation on the missing values

What method does a random forest use to make an intial guess for missing values in the training data?

The mean value

The median value

Proximity matrix

Please fill in the comment box below.

Time is Up!

Time's up

This entry was posted on December 16, 2020 by lukas.halim@gmail.com.

← Random Forests Lasso and Ridge Regression →

Build Knowledge

Drill to increase your skill! Wikipedia explains, “The testing effect is the finding that long-term memory is often increased when some of the learning period is devoted to retrieving the to-be-remembered information.” In other words, you remember what you learn better when you answer questions than when you spend the same time watching videos or reading blog posts. Answering open-ended questions (explain a random forest) is generally more effective than answering multiple-choice questions.

Leave a Reply Cancel reply