Feature Selection

Embedded approaches: data mining algorithm itself decides which attributes to use. e.g. decision tree classifier
Filter approaches: before data mining algorithm is run; independent of the data mining task. e.g. pairwise correlation is as low as possible
Wrapper: use data mining algorithm as black box

Feature Selection Architecture

a measure for evaluating a subset
a strategy that controls the generation of a new subset of features
a stopping criterion
a validation procedure

mRMR (minimum Redundancy Maximum Relevance Feature Selection)

Univariate Selection

One variable + One target

http://blog.datadive.net/selecting-good-features-part-i-univariate-selection/

Reasons of feature selection:

Reducing the number of features, to reduce overfitting and improve the generalization of models.

To gain a better understanding of the features and their relationship to the response variables.

Pearson Correlation

A value between -1 and 1

-1: perfect negative correlation
+1: perfect positive correlation
0: no linear correlation

Use pearsonr in Scipy: https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.pearsonr.html

from scipy.stats import pearsonr

pearsonr(x, y)

Pros:

fast to calculate
returned value [-1, 1] instead of [0, 1], extra negative/positive info

Cons:

only sensitive to linear relationship.

Maximal Information Coefficient

https://en.wikipedia.org/wiki/Maximal_information_coefficient

Searches for optimal binning and turns mutual information score into a metric that lies in range [0;1].
Linear or non-linear.

Distance correlation

While for Pearson correlation, the correlation value 0 does not imply independence, distance correlation of 0 does imply that there is no dependence between the two variables.

Linear Model And Regularization

http://blog.datadive.net/selecting-good-features-part-ii-linear-models-and-regularization/

Random Forest

http://blog.datadive.net/selecting-good-features-part-iii-random-forests/