Can I read k-NN for Missing Data Imputation on EtoBox?

k-NN for Missing Data Imputation by Anjitha Divakaran is a document available to read on EtoBox.

What is k-NN for Missing Data Imputation about?

The document discusses different techniques for imputing missing data: mean substitution, median substitution, and standard deviation substitution. It applies these techniques using k-nearest neighbors (k-NN) algorithm to group a dataset into clusters of different sizes. The results show that median and standard deviation substitution have higher accuracy than mean substitution for imputing missing data. Accuracy also improves with larger cluster sizes. Median and standard deviation are thus better single i

Author

Anjitha Divakaran

Language

EN