Feature selection is one of the most important tasks in machine learning. Learn how to use a simple random search in Python to get good results in less time.
Feature selection has always been a great task in machine learning. According to my experience, I can surely say that feature selection is much more important than model selection itself.
I have already written an article about feature selection. It was an unsupervised way to measure feature importance in a binary classification model, using Pearson’s chi-square test and correlation coefficient.
Generally speaking, an unsupervised approach is often enough for a simple feature selection. However, each model has its own way of “thinking” the features and treat their correlation with the target variable. Moreover, there are models that do not care too much about collinearity (i.e., the correlation between the features) and other models that show very big problems when it occurs (for example, linear models).
Automating artificial intelligence for medical decision-making