Description
Andrew Russell – CFI Education – Loan Default Prediction with Machine Learning download, Andrew Russell – CFI Education – Loan Default Prediction with Machine Learning review, Andrew Russell – CFI Education – Loan Default Prediction with Machine Learning free
Andrew Russell – CFI Education – Loan Default Prediction with Machine Learning
Loan Default Prediction with Machine Learning
Combine a data set with basic Machine Learning skills to predict which customers are likely to default on their loans.
Overview
Loan Default Prediction with Machine Learning Course Overview
Machine Learning is about making predictions using data. In this course, you’ll learn to use basic Machine Learning skills to predict which customers are likely to default on their loans.
Once your model classifies each loan, you’ll learn to visualize your predictions to see how well the model performed.
Predicting defaults and creditworthiness is hugely valuable to risk management and pricing decisions.
We will cover the entire Machine Learning process in Python, reinforcing concepts from Python fundamentals. You’ll learn how to create predictive classification models, fine-tune and test your process, and how to interpret the results.
Machine Learning is a hot topic in the world of data, particularly data science. At a basic level, Machine Learning is not as complex as it may sound. If you’ve ever done linear regression, you may be surprised to learn that you’ve already taken steps toward this exciting world.
Join Andrew for a comprehensive step-by-step walkthrough of the Machine Learning process.
Loan Default Prediction with Machine Learning Objectives
Upon completing this course, you will be able to:
Explain and discuss the main steps of the Machine Learning cycle
Load and clean data into a python notebook
Use Exploratory Data Analysis to identify variables with likely predictive power
Use Feature Engineering to transform data into a more useful format
Build a logistic regression and random forest prediction model
Evaluate and compare model performance using common evaluation metrics
Who should take this course?
The Machine Learning cycle is one of the most foundational aspects of Data Science. Using this process, we can learn to make predictions using all types of data and variables. Anyone looking to make predictions in a practical Python environment should absolutely be doing this course.
What you’ll learn
Introduction
Course Introduction
What is Machine Learning?
Case Study Overview
Course Materials
Student Downloads
Course Outline
Load & Clean Data
Chapter Introduction – Load and Clean Data
Vehicle Loans Data Set
Import Libraries and Data Set
Exploring Basic Data Parameters
Exploring Our Target Variable
Identifying Missing Data
Dealing with Missing Data
Dates – Exploring Date Columns
Dates – Calculating Age
Dates – Extracting Month Number
Strings – Exploring String Columns
Strings – Extract Numbers from Strings
Strings – Strings Function Exercise
Strings – Strings Function Exercise review
Chapter Summary
Exploratory Data Analysis
Chapter Intro – Exploratory Data Analysis
Categorical, Continuous and Binary Variables
Identifying Features (Columns) of Interest
Dealing with Category or ID Columns
Grouping Data by Categories
Looking at Default Frequency Within Groups
Exercise – EDA Function for Categorical Variables
Exercise – Review
Plotting Continuous Variables
Plotting Continuous Variables by Group
Exercise – EDA Function for Continuous Variables
Exercise – Review
Exploring Binary Variables
Chapter Summary
Feature Engineering
Chapter Introduction – Feature Engineering
Exploring Outliers
Creating Bins
Combining Features into New Columns
Exercise – Creating New Columns
Exercise – Review and Calculate Percentages
Dealing with Null Values
Min and Max Scaling
Chapter Summary
Classification with Logistic Regression
Chapter Introduction – Classification with Logistic Regression
Linear vs. Logistic Regression
Train and Test Split
Import Data and Modify Column Types
Exercise – Select Chosen Variables
Exercise – Review
Exercise – Separate the Target Variable
Exercise – Review
Splitting the Data into Train and Test
Dummy Variables
Variable Encoding (One-Hot-Encoding)
Exercise – Train, Test and Split One-Hot-Encoded Data
Exercise Review and Testing Our Logistic Regression
Chapter Summary
Model Evaluation
Chapter Introduction – Model Evaluation
Student Exercise
Review Logistic Regression Model
Evaluation Metrics Theory
Evaluation – Creating a Confusion Matrix
Evaluation – Precision, Recall and F1 Scores
The ROC Curve
ROC Curve – Extracting Predicted Probabilities
ROC Curve – Plotting the Curve
Advanced Evaluation – Prediction Percentages
Advanced Evaluation – Class Probability Distributions
Advanced Evaluation – Plotting Class Probabilities
Advanced Evaluation – Creating an Evaluation Function
Chapter Summary
Classification with Random Forest
Chapter Introduction – Classification with Random Forest
Decision Tree Theory
Random Forest Theory
Creating a Function for Train and Test Split
Review the Train Test Split Function
Exercise – Create a Simple Random Forest
Reviewing the Random Forest
Identifying Overfitting in Our Results
Hyperparameters
Testing the Impact of Number of Trees
Testing the Impact of Maximum Depth
Chapter Summary
Improving Accuracy
Chapter Introduction – Improving Accuracy
Recap and Theory
Exercise
Balancing Classes Automatically
Manual Class Balancing
Resampling – Upsampling
Training a New Model Based on Resampled Data
Evaluating the Downsampled Model
SMOTE
Chapter Summary
Qualified Assessment
Qualified Assessment
This Course is Part of the Following Programs
Why stop here? Expand your skills and show your expertise with the professional.
Business Intelligence & Data Analyst (BIDA)®
Loan Default Prediction with Machine Learning is part of the Business Intelligence & Data Analyst (BIDA)®, which includes 33 courses.
Skills Learned
Data visualization, data warehousing and transformation, data modeling and analysis
Career Pre
pBusiness intelligence analyst, data scientist, data visualization specialist