Feature Engineering For Machine Learning

Updated:06/17/2021 by Computer Hope

Feature are using predicting analysis
every analysis have significance
so we need to select feature
There are lot of technic like missing data
we have technic

Examples of classification problems include: Given an example, classify if it is spam or not. Given a handwritten character, classify it as one of the known characters

Types of Methods in Feature Engineering For Machine Learning

1)Imputation removing null value and preparing data

we can put some string or numeric (mean and Value ). Main motive remove null value or either replace null value . EDA and feature engineering are different . represent by feature engineering

2)Handlig outlier

You have some statical method and using visualisation find outler Visualisation are best method to detect . Statical method are there like standard deviation or percentile methods

3)Log transformation

It is common Sque or log transform data or we checking some magnitude


We creating Bining or over fitting

5) One hot encoding data

we use group by methods

5) Z-Score

deviation How is your data is varies z core = (data -mean)/ standard deviation