Course Description
With predictive analytics, relevant data should be stored for easy retrieval, kept up-to-date, and attributes must be selected contingent on their predictive potential. Explore data reduction and graphic tools for exploratory data analysis.
Learning Objectives
Data Reduction with PCA and Factor Analysis
- start the course
- recognize key data reduction methodologies
- use principal component analysis for feature selection
- use the information theory approach for feature selection
- recognize the key features of using Chi-square
- recognize key features of the wrapper data reduction method
- recognize the key features of factor analysis
Tools for Exploratory Data Analysis (EDA)
- recognize key features of EDA and how quantitative techniques are used to perform EDA
- use bar charts and box-and-whisker plots to perform EDA
- use run charts and scatter plots to perform EDA
- use histograms and stem-and-leaf plots to perform EDA
Practice: Using PCA for Feature Selection
- recognize the direction, form, and strength of a scatter plot