Question

What is x fearure and y feature in data mining?

Answer 1

Answer:

(1) avoid inclusion of duplicate or correlated features (2) consider only those features that are well seperated from each other. Reducing the dimensionality of the input training feature set would reduce the computational burden of the classifier thereby improving its classification accuracy. For e.g., if x & y are two features under consideration, the difference of their means must be large (which represents that x & y are well seperated from each other) and at the same time their individual standard deviations must be low (which reprents the less spread of each feature). Other than PCA, LDA, SVD, etc. some distance based feature selection methods can be adopted.

Explanation:

hope it helps u

:)