Feature engineering is a machine learning technique that uses data to create new variables not present in the training set. It involves techniques like outlier detection and removal, one hot encoding, log transforms, dimensionality reduction using PCA, handling missing values, and scaling. Outliers are unusual data points that differ significantly from other samples and can occur due to errors or be legitimate variations, while percentiles describe the value a given percentage of values are lower than.

1. Feature Engineering

2. Feature Engineering
Feature engineering is a machine
learning technique that leverages data to
create new variables that aren't in the
training set.
Math, Statistics and Domain Knowledge

3. Important area in the field of machine learning and data
analysis
Feature engineering techniques,
1) Outlier detection and removal
2) One hot encoding
3) Log transform
4) Dimensionality reduction using principal component
analysis (a.k.a. PCA)
5) Handling missing values
6) Scaling

6. Outliers
Outliers are unusual data points that differ
significantly from rest of the samples.
They can occur due to an error in data
collection process or they are just legitimate
data points and represent natural variation.
Eg: age more than 100

7. Percentile
Percentiles are used in statistics to give you a
number that describes the value that a given
percent of the values are lower than.

8. Outliers