Book Image

Data Cleaning and Exploration with Machine Learning

By : Michael Walker

Book Image

Data Cleaning and Exploration with Machine Learning

By: Michael Walker

Overview of this book

Many individuals who know how to run machine learning algorithms do not have a good sense of the statistical assumptions they make and how to match the properties of the data to the algorithm for the best results. As you start with this book, models are carefully chosen to help you grasp the underlying data, including in-feature importance and correlation, and the distribution of features and targets. The first two parts of the book introduce you to techniques for preparing data for ML algorithms, without being bashful about using some ML techniques for data cleaning, including anomaly detection and feature selection. The book then helps you apply that knowledge to a wide variety of ML tasks. You’ll gain an understanding of popular supervised and unsupervised algorithms, how to prepare data for them, and how to evaluate them. Next, you’ll build models and understand the relationships in your data, as well as perform cleaning and exploration tasks with that data. You’ll make quick progress in studying the distribution of variables, identifying anomalies, and examining bivariate relationships, as you focus more on the accuracy of predictions in this book. By the end of this book, you’ll be able to deal with complex data problems using unsupervised ML algorithms like principal component analysis and k-means clustering.

Preface

Who this book is for

What this book covers

To get the most out of this book

Download the example code files

Download the color images

Conventions used

Share Your Thoughts

Section 1 – Data Cleaning and Machine Learning Algorithms

Section 1 – Data Cleaning and Machine Learning Algorithms

Free Chapter

Chapter 1: Examining the Distribution of Features and Targets

Chapter 1: Examining the Distribution of Features and Targets

Technical requirements

Subsetting data

Generating frequencies for categorical features

Generating summary statistics for continuous and discrete features

Identifying extreme values and outliers in univariate analysis

Using histograms, boxplots, and violin plots to examine the distribution of features

Chapter 2: Examining Bivariate and Multivariate Relationships between Features and Targets

Chapter 2: Examining Bivariate and Multivariate Relationships between Features and Targets

Technical requirements

Identifying outliers and extreme values in bivariate relationships

Using scatter plots to view bivariate relationships between continuous features

Using grouped boxplots to view bivariate relationships between continuous and categorical features

Using linear regression to identify data points with significant influence

Using K-nearest neighbors to find outliers

Using Isolation Forest to find outliers

Chapter 3: Identifying and Fixing Missing Values

Chapter 3: Identifying and Fixing Missing Values

Technical requirements

Identifying missing values

Cleaning missing values

Imputing values with regression

Using KNN imputation

Using random forest for imputation

Section 2 – Preprocessing, Feature Selection, and Sampling

Section 2 – Preprocessing, Feature Selection, and Sampling

Chapter 4: Encoding, Transforming, and Scaling Features

Chapter 4: Encoding, Transforming, and Scaling Features

Technical requirements

Creating training datasets and avoiding data leakage

Removing redundant or unhelpful features

Encoding categorical features

Encoding categorical features with medium or high cardinality

Using mathematical transformations

Feature binning

Feature scaling

Chapter 5: Feature Selection

Chapter 5: Feature Selection

Technical requirements

Selecting features for classification models

Selecting features for regression models

Using forward and backward feature selection

Using exhaustive feature selection

Eliminating features recursively in a regression model

Eliminating features recursively in a classification model

Using Boruta for feature selection

Using regularization and other embedded methods

Using principal component analysis

Chapter 6: Preparing for Model Evaluation

Chapter 6: Preparing for Model Evaluation

Technical requirements

Measuring accuracy, sensitivity, specificity, and precision for binary classification

Examining CAP, ROC, and precision-sensitivity curves for binary classification

Evaluating multiclass models

Evaluating regression models

Using K-fold cross-validation

Preprocessing data with pipelines

Section 3 – Modeling Continuous Targets with Supervised Learning

Section 3 – Modeling Continuous Targets with Supervised Learning

Chapter 7: Linear Regression Models

Chapter 7: Linear Regression Models

Technical requirements

Linear regression and gradient descent

Using classical linear regression

Using lasso regression

Using non-linear regression

Regression with gradient descent

Chapter 8: Support Vector Regression

Chapter 8: Support Vector Regression

Technical requirements

Key concepts of SVR

SVR with a linear model

Using kernels for nonlinear SVR

Chapter 9: K-Nearest Neighbors, Decision Tree, Random Forest, and Gradient Boosted Regression

Chapter 9: K-Nearest Neighbors, Decision Tree, Random Forest, and Gradient Boosted Regression

Technical requirements

Key concepts for K-nearest neighbors regression

K-nearest neighbors regression

Key concepts for decision tree and random forest regression

Decision tree and random forest regression

Using gradient boosted regression

Section 4 – Modeling Dichotomous and Multiclass Targets with Supervised Learning

Section 4 – Modeling Dichotomous and Multiclass Targets with Supervised Learning

Chapter 10: Logistic Regression

Chapter 10: Logistic Regression

Technical requirements

Key concepts of logistic regression

Binary classification with logistic regression

Regularization with logistic regression

Multinomial logistic regression

Chapter 11: Decision Trees and Random Forest Classification

Chapter 11: Decision Trees and Random Forest Classification

Technical requirements

Decision tree models

Implementing random forest

Implementing gradient boosting

Chapter 12: K-Nearest Neighbors for Classification

Chapter 12: K-Nearest Neighbors for Classification

Technical requirements

Key concepts of KNN

KNN for binary classification

KNN for multiclass classification

Chapter 13: Support Vector Machine Classification

Chapter 13: Support Vector Machine Classification

Technical requirements

Key concepts for SVC

Linear SVC models

Nonlinear SVM classification models

SVMs for multiclass classification

Chapter 14: Naïve Bayes Classification

Chapter 14: Naïve Bayes Classification

Technical requirements

Naïve Bayes classification models

Naïve Bayes for text classification

Section 5 – Clustering and Dimensionality Reduction with Unsupervised Learning

Section 5 – Clustering and Dimensionality Reduction with Unsupervised Learning

Chapter 15: Principal Component Analysis

Chapter 15: Principal Component Analysis

Technical requirements

Key concepts of PCA

Feature extraction with PCA

Using kernels with PCA

Chapter 16: K-Means and DBSCAN Clustering

Chapter 16: K-Means and DBSCAN Clustering

Technical requirements

The key concepts of k-means and DBSCAN clustering

Implementing k-means clustering

Implementing DBSCAN clustering

Other Books You May Enjoy

Other Books You May Enjoy

Packt is searching for authors like you

Share Your Thoughts

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

Using K-fold cross-validation

So far, we have held back 30% of our data for validation. This is not a bad strategy. It prevents us from peeking ahead to the testing data as we train our model. However, this approach does not take full advantage of all the available data, either for training or for testing. If we use K-fold cross-validation instead, we can use all of our data while also avoiding data leakage. Perhaps that seems too good to be true. But it’s not because of a neat little trick.

K-fold cross-validation trains our model on all but one of the K folds, or parts, leaving one out for testing. This is repeated k times, each time excluding a different fold for testing. Performance metrics are then based on the average scores across the K folds.

Before we start, though, we need to think again about the possibility of data leakage. If we scale all of the data that we will use to train our model and then split it up into folds, we will be using information from all...