Employee Quit Predictor : Project Overview

• Created a model to predict if an employer would leave a company based off some data provided to us by the HR department of the company

• Exploratory Data Analysis was applied to discover insights in the data

• Got a Test Accuracy of 99%

• Used a Decision Tree as model, since problem was a Classification

---

Model Performance

The model performed well with the following with the following metrics but had 53 False Positives and 37 False Negatives

Metric	Score (%)
Accuracy	98.07
Precision	95.22
Recall	96.74

---

Code and Resources Used

• Python Version: 3.8
• Packages: pandas, numpy, matplotlib, seaborn, sklearn
• Dataset: HR Analytics

---

Features

• satisfaction_level
• last_evaluation
• number_project
• average_montly_hours
• time_spend_company
• Work_accident
• promotion_last_5years
• salary_code

Target

• left
  • 0 - no
  • 1 - yes

---

Model Building

I split the test and train set 70% and 30% respectively

I used Decision Tree as the algorithm for the model

---

Confusion Matrix