Data Manipulation and PCA (Principal Component Analysis )

Data Manipulation and PCA

Last updated 6/2018

English

What you'll learn

By the end of this course , a student will be able to do the following:
Stet a working directory , Import a txt or csv file, eliminate duplicate rows in the data, detect rows containing missing values, eliminate rows containing missing values, replace missing values by the mean, replace missing values by a specified information, use the apply function , do some arithmetic on columns , detect strongly correlated variable (some nice plots for visualization ), compute the correlation matrix , the eigenvalue and eigenvector vector, select the number of components the compute the components

1 section • 12 lectures • 1h 9m total length

Introduction0:46
How to download and install R7:54
How to install a package and import a library3:44
How to import a data (Formats : csv, txt) and how to set a working directory10:47
Eliminate duplicate rows2:44
Identify duplicate rows and keep one observation per unique entry, showing how removing duplicates affects the collection.
Missing values detection and treatment9:20
Identify missing values in a dataset, learn when to remove or impute them, visualize missingness, and apply mean or median imputation to numeric features for data prepared for PCA.
Data visualization (Detection of Strongly correlated variables)6:15
select a subset of the data based on specified criteria7:06
Learn how to select a subset of the data based on specified criteria, such as year greater than 2000 and age less than 30, and compute the proportions for analysis.
Operation on columns, Variables and standardization , how to use the apply() f11:16
Selecting the number of principal components1:45
Select the number of principal components in PCA by examining the curve, noting a big drop at five, and diagnose each component with a correlation matrix.
Computation of the correlation matrix, eigenvalues and vectors3:03
Computation of components5:06
Explore data manipulation and principal component analysis, computing principal components from a standardized correlation matrix using eigenvectors, projecting data into reduced five-component space.

In this course, we learn the following:

How to Stet a working directory

How to Import a txt or csv file

How to eliminate duplicate rows in the data

How to detect rows containing missing values

How to eliminate rows containing missing values

How to replace missing values

How to select a subset of the data based on specifics criteria

How to do arithmetic on columns

How detect strongly correlated variable (some nice plots for visualization )

How to compute the correlation matrix , the eigenvalue and eigenvector

How select the number of components

How to compute the components

If you are working on a large data set or trying to get some informations about your data , then this course is a right fit for you.