Intro to Machine Learning

Overview

This Week Presents Two Topics

1. We continue our study of pre-machine learning analyses, including the extension of the data analysis material presented at the end of Week 1, bar charts and histograms. After preparing the data for analysis, the first analysis summarizes the relevant variables, with statistics and with data visualizations.
2. We begin our study of the conceptual basis of machine learning with our first form of supervised machine learning, regression analysis. However, it will not be until next week that we do a formal machine learning analysis with the Python machine learning framework.

Remember that this course presents all needed information regarding Python for doing the homework and tests. Searching for additional information can be useful, to read or view some explanation with someone else's words, but there is never any need to obtain more information about Python or any other content beyond the instructions and information provided for this course. And, there are always multiple coding possibilities to accomplish any objective, so be prepared to encounter code and explanations beyond what is needed in this course.

Comment on the Following Basic Introduction to Regression Analysis

Regression analysis is the first machine learning analysis procedure, more than a century old but machine learning in the modern sense. The basic regression material presented in this course was presented in your stat course, plus in GSCM 451/571 or BTA 516 that many of you have had. For many, this conceptual material is a review, with the new material being the Python implementation. Nonetheless, an understanding of regression analysis is critical to understanding machine learning, so even if you studied this material not long ago, always a good review. If you forgot this material from your intro stat course and have not had a more recent exposure, make sure to take the time to understand these concepts explained in Sections 1 and 2 of the online posted reading.

Introduction to Machine Learning

Regression Concepts

Regression Python Template

There are two straightforward coding changes since recording the video:

• The function distplot() in the video is now deprecated in favor of the newer displot(), illustrated in the pdf.
• For the scatterplot, relplot() is used in the pdf, and scatterplot() in the video. relplot() is newer with more features.

video [22:48]

• 00:00 Intro
• 01:25 Data
• 04:00 X and y data structures
• 07:00 Data Visualizations
• 09:54 Model Definition
• 12:06 Estimated Model
• 17:09 Model Fit

When the video plays you will see three horizontal bars at the top left. Click on those bars and a table of contents appears where you can skip to any individual chapter. The resulting menu will cover the left side of the video. To close the menu, go to the top-right side fo the menu and find another three horizontal bars. Click to close the menu down. With a light background color the second set of white bars may be difficult to recognize.

Homework

Short-Answer Questions

Analysis Problems If downloading adds a .txt to the file type, then delete the .txt from the name.