sonik8494 / feature_engineering_project

Feature Engineering Project

Welcome to the Feature Engineering project. In this project, we will strengthen whatever we have learned so far in feature engineering course.

In all above procedures we have used cleaned data or Pre-processed data.

Real-life data contains missing values, ouliers, and is less optimized.

What we have learned in class so far:

• Missing Value - finding out the missing value from the dataset and Imputation of missing value
• Outlier - Outlier detection and methods to deal with outliers
• Skewness - Skewness treatment using transformation methods
• Encoding - Encoding of categorical data.

Dataset

In this assignment you're going to be dealing with a familiar data set i.e. NY Housing data.

As we all know that this dataset contains 81 variables related to houses listed in NY and their SalePrice. You will be given a subset of this dataset to perform few tasks and the whole (original) dataset for other tasks. This will give you an idea about how techniques are related.

Why this assignment?

• This assignment will help you to get familiar with primary and important steps of Pre-Processing and cleaning a dataset.
• We will learn why use specific techniques over other techniques.
• For this assignment you will be given following python packages:
  • Pandas
  • Numpy
  • sklearn
  • scipy

By completing this project you will be awarded with 250 points.

So, let's get started!