Unit 1- Regression with Tidymodels

Download PDF

100%

Try now

Unit One - Regression with Tidymodels

Study Objectives

After completing this unit, students should be able to:

> Apply packages within the tidymodels ecosystem

> Perform linear regression with tidymodels

> Perform logistic regression with tidymodels

> Fit a model using the parsnip package

> Perform model evaluation with the yardstick package

Overview of Predictive and Prescriptive Analytics

Big Data Analytics facilitates the analyses of complex data and the generation of insights which hitherto were impossible using traditional data processing software.

Types of Big Data Analytics:

Descriptive analytics

Diagnostic Analytics

Predictive Analytics

Prescriptive Analytics

Types of Big Data Analytics

Supervised Methods

Supervised Machine Learning

Supervised Machine Learning Workflow

Modelling with Tidymodels Package

Supervised Machine Learning and Tidymodels

Classification Models with Tidymodels

Linear Regression Model

Data Resampling [Data Partitioning into Training and Testing Data]

Data Resampling

Linear Regression with Tidymodels

One. Specify model type

Two. Specify the engine

Three. Specify the mode

Model Fitting - Linear Regression

Estimating Model Parameters

Making Predictions

Adding Predictions to the Test Data

Evaluating Model Performance - Yardstick Package

Root Mean Square Error

Evaluating Model Performance - Yardstick Package

Streamlining Model Fitting - Last Fit Function

Streamlining Model Fitting - Collecting Metrics

Streamlining Model Fitting -Collecting Predictions

LOGISTIC REGRESSION

Use Cases of Logistic Regression

Logistic Regression in R -Using Tidymodels Package

Logistic Regression

Logistic Regression -Model Specification

Logistic Regression -Model Fitting

Logistic Regression -Predicting Outcome Categories

Assessing Model Fit

Assessing Model Fit - Confusion Matrix

Visualizing Model Performance

Visualizing Model Performance -Exploring Performance Across Thresholds

Visualizing Model Performance.

Summarizing the ROC Curve.

Final Thoughts.

Unit two - Feature Engineering.

Feature Engineering.

Feature Engineering (with Recipes Package).

Determine variable data types.

Feature Engineering (with Recipes Package) - Data Pre-processing Steps.

Feature Engineering (with Recipes Package) - Applying Recipes to New Data.

Transforming Nominal Predictors.

Unit three -Decision and Regression Trees Modeling.

Decision Trees

Decision Tree - Divide and Conquer Process

Splitting Data into "Pure" Regions

How to Determine Best Splits

Computation of GINI Index - Splitting Based on GINI

Advantages of Decision Trees

Disadvantages of Decision Trees

Training Decision Trees in R

Classification and Regression Trees - rpart package

Hyperparameters in Decision Trees

Cost Complexity Parameter

Maxdepth

Cost-Complexity Parameter (CP)

Regression Trees

Common Metrics for Regression

The average absolute distance between the actual (or observed values) and predicted values.

Root Mean Square Error (RMSE)

RMSE equals

Unit Four Random Forest

Bagging Trees

Steps to Bagging

Bagging Outcome

Bootstrap Example

Flaws of Bootstrapping

Random Forest

Random Forest Algorithm in R

Advantages of Random Forest Package

Unit five - K-Nearest Neighbour (kNN) Algorithm

K Nearest Neighbour (kNN)

How kNN Works

The kNN Algorithm - Distance

Illustrative Application of KNN - Self Driving Car

Measuring Similarity

Applying KNN in R

What is the 'k' in kNN

What is the 'k' in kNN ... continued

Bigger 'k' is not always better ...

The kNN Algorithm - pragmatically selecting k

Preprocessing Data for Nearest Neighbors - Dummy Encoding

Preparing Data for Nearest Neighbors - Normalization

Requirements for kNN in R

Dummy Encoding

Naïve Bayes

What is Naïve Bayes?

Why Naïve?

Use Cases of Naïve Bayes

Document Categorization

How the Naïve Bayes Algorithm Works

Bayes Theorem

Bayes Theorem - Terminology

Bayes Theorem

Naïve Bayes Algorithm

Illustrative Example - Naïve Bayes

Final Thoughts ...

Unit Seven - Support Vector Machines (SVM)

Support Vector Machine

System

Support Vectors

Support Vector Machines

The Kernel Trick

For Example:

SVM final thoughts

Unit Eight - Neural Networks

Neural Networks

Structure of Neural Networks

Feedforward and Feedback Artificial Neural Network

Feedforward Artificial Neural Networks

Feedback Artificial Neural Networks

How the Neural Network Works

Activation Functions - Identity Function

Activation Functions - Binary Step Function

Activation Functions - Sigmoid Function

Activation Functions - ReLU Function

Advantages of Neural Networks

Disadvantages of Neural Networks

Overview

The document provides an educational overview of regression analysis employing tidymodels in R, focusing on model building, training, and evaluation, particularly for supervised learning methods. It includes insights into applying various tidymodels packages for effective data analysis.

Key Points

1Students will learn to apply tidymodels for linear and logistic regression
2The importance of feature engineering and data resampling in model fitting is emphasized
3Evaluation metrics such as RMSE and R-squared are explained for assessing model performance
4Supervised machine learning methods are highlighted with practical examples
5The last_fit() function streamlines the modeling process in tidymodels.

Details

Authors: Nana Boateng
Category: Technology and Engineering

PDF
Streaming VLM: Real-Time Understanding for Infinite Video Streams
This document presents Streaming VLM, a vision-language model designed to understand infinite video streams in real-time, addressing critical challenges in memory usage and latency. It introduces methods for efficient processing and evaluation benchmarks for performance assessment.
PDF
Artificial Intelligence as a general-purpose technology: an historical perspective
This paper examines the impact of general-purpose technologies like artificial intelligence on productivity, drawing historical parallels with previous technological revolutions to explore potential future implications for economic growth.
PDF
Structured fabrics with tunable mechanical properties
This article discusses the development of structured fabrics with tunable mechanical properties, focusing on their design, experimental validation, and potential applications in wearable technology and adaptive materials.
PDF
Specification Engineering: The New Skill After Prompt Engineering
This document discusses the emerging skill of specification engineering, which goes beyond traditional prompt engineering to define the goals, constraints, and validation processes for AI-assisted tasks. It highlights the importance of clear specifications for improved outcomes in various AI applications.
PDF
AD95 Information Assurance and IT Security
This document outlines the information assurance and IT security policies at Penn State University, detailing responsibilities, guiding principles, classification of information, and mandatory adherence to security standards.