メインコンテンツへスキップ

コース

Rによる次元削減

基礎スキルレベル

更新日 2024/12

Rで次元削減を学び、あなたのデータとモデルに最適な特徴選択・特徴抽出を習得しましょう。

コースを無料で開始

RMachine Learning

4時間

16 ビデオ

56 演習

4,600 XP

2,744

修了証明書

何千もの企業の従業員が支持

チームのトレーニングを担当していますか？

Businessをお試しください

コース説明

前提条件

Modeling with tidymodels in R

1

Foundations of Dimensionality Reduction

Prepare to simplify large data sets! You will learn about information, how to assess feature importance, and practice identifying low-information features. By the end of the chapter, you will understand the difference between feature selection and feature extraction—the two approaches to dimensionality reduction.

Introduction to dimensionality reduction

Dimensionality and feature information

Mutual information features

Information and feature importance

Calculating root entropy

Calculating child entropies

Calculating information gain of color

The Importance of Dimensionality Reduction in Data and Model Building

Calculate possible combinations

Curse of dimensionality, overfitting, and bias

チャプターを開始

2

Feature Selection for Feature Importance

Learn how to identify information-rich and information-poor features missing value ratios, variance, and correlation. Then you'll discover how to build tidymodel recipes to select features using these information indicators.

Feature selection vs. feature extraction

Create a zero-variance filter

Create a missing values filter

Feature selection with the combined filter

Selecting based on missing values

Create a missing value ratio filter

Apply a missing value ratio filter

Create a missing values recipe

Selecting based on variance

Create a low-variance filter

Create a low-variance recipe

Selecting based on correlation with other features

Identify highly correlated features

Select correlated feature to remove

Create a high-correlation recipe

チャプターを開始

3

Feature Selection for Model Performance

Chapter three introduces the difference between unsupervised and supervised feature selection approaches. You'll review how to use tidymodels workflows to build models. Then, you'll perform supervised feature selection using lasso regression and random forest models.

Supervised feature selection

Supervised vs. unsupervised feature selection

Decision tree feature selection type

Model Building and Evaluation with tidymodels

Split out the train and test sets

Create a recipe-model workflow

Fit, explore, and evaluate the model

Lasso Regression

Scale the data for lasso regression

Explore lasso regression penalty values

Tune the penalty hyperparameter

Fit the best model

Random forest models

Create full random forest model

Reduce data using feature importances

Create reduced random forest

チャプターを開始

4

Feature Extraction and Model Performance

In this final chapter, you'll gain a strong intuition of feature extraction by understanding how principal components extract and combine the most important information from different features. Then learn about and apply three types of feature extraction — principal component analysis (PCA), t-SNE, and UMAP. Discover how you can use these feature extraction methods as a preprocessing step in the tidymodels model-building process.

Foundations of feature extraction - principal components

Understanding principal components

Naming principal components

Principal Component Analysis (PCA)

PCA: variance explained

Mapping features to principal components

PCA in tidymodels

t-Distributed Stochastic Neighborhood Embedding (t-SNE)

Separating house prices with PCA

Separating house prices with t-SNE

Uniform Manifold Approximation and Projection (UMAP)

Separating house prices with UMAP

UMAP reduction in a decision tree model

Evaluate the UMAP decision tree model

チャプターを開始

Rによる次元削減

コース完了

修了証明書を取得

この修了書をLinkedInや履歴書、CVに追加しましょう
ソーシャルメディアや人事評価で共有しましょう今すぐ登録

19百万人を超える学習者と共にRによる次元削減を始めましょう！

DataCamp for Mobileでデータスキルを磨きましょう

モバイルコースと毎日の 5 分間のコーディングチャレンジで、外出先でも進歩できます。