In this chapter you will learn about the problems addressed by HR analytics, as well as will explore a sample HR dataset that will further be analyzed. You will describe and visualize some of the key variables, transform and manipulate the dataset to make it ready for analytics.

Introduction and overview

Finding categorical variables

Observing categoricals

Transforming categorical variables

Encoding categories

Getting dummies

Dummy trap

Descriptive statistics

Correlations in the employee data

Percentage of employees who churn

Introduction to HR Analytics

This chapter introduces one of the most popular classification techniques: the Decision Tree. You will use it to develop an algorithm that predicts employee turnover.

De gegevens splitsen

Target en features scheiden

Medewerkersdata splitsen

Introductie tot Decision Tree-classificatie

Gini-index berekenen

De boom splitsen

Medewerkersverloop voorspellen met beslissingsbomen

De boom afstemmen op werknemersdata

De nauwkeurigheid van je voorspelling controleren

Interpretatie van de beslisboom

De boom exporteren

Resultaten interpreteren

Predicting employee turnover

Here, you will learn how to evaluate a model and understand how "good" it is. You will compare different trees to choose the best among them.

Tuning employee turnover classifier

Pruning the tree

Limiting the sample size

Evaluating the model

Interpreting accuracy metrics

Calculating accuracy metrics: precision

Calculating accuracy metrics: recall

Targeting both leavers and stayers

Calculating the ROC/AUC score

Class imbalance

Balancing classes

Comparison of Employee attrition models

Evaluating the turnover prediction model

In this final chapter, you will learn how to use cross-validation to avoid overfitting the training data. You will also learn how to know which features are impactful, and which are negligible. Finally, you will use these newly acquired skills to build a better performing Decision Tree!

Hyperparameter tuning

Cross-validation using sklearn

Setting up GridSearch parameters

Implementing GridSearch

Important features for predicting attrition

Interpreting importance

Sorting important features

Selecting important features

Develop and test the best model

Final thoughts

Choosing the best turnover prediction model

Employee turnover data

Van alle bedrijfsdomeinen is HR nog het minst ontwricht. Met de nieuwste ontwikkelingen in tools en technologieën voor gegevensverzameling en -analyse kun je echter in alle onderdelen, dus ook in HR, datagedreven beslissingen nemen. Deze cursus geeft je een stevige basis om met werknemersgegevens te werken en een voorspellend model te ontwikkelen om personeelsverloop te analyseren.

Intermediate Python

Leer hoe je Python gebruikt voor personeelsdata en een voorspellend model voor personeelsverloop.

HR-analytics: verloop van medewerkers voorspellen in Python

In deze cursus leer je hoe je machine learning kunt gebruiken in HR.

Resultaten interpreteren

HR-analytics: verloop van medewerkers voorspellen in Python

Praktische interactieve oefening