Employees Retention Predictive Churn Model

Objective

To develop a predictive model that forecasts the likelihood of employee turnover in a company using machine learning. This model aims to assist HR departments in implementing retention strategies effectively. The model prediction accuracy is 93%

Challenge:

Using Looker studio and design the dashboard

First time working with a looker studio require some level of understanding and research time

Tools & Technologies: Python (PyCaret), Google BigQuery, Google Colab

Datasets: Dataset - 1 ; Dataset - 2

Coding file with step by step instruction: Here

Link to the report in Looker Studio: Here

Static Image:

Work Flows:

Data Acquisition: Connected to Google BigQuery to retrieve employee data. The dataset includes various features like satisfaction level, last evaluation, number of projects, average monthly hours, time spent in the company, etc.
Data Exploration: Performed initial exploration of the dataset using Python and SQL, examining the structure and content of the data to understand the features available for modeling.
Model Development: Using Python library PyCaret for model selection, feature selection and model building
Data Visualization: Report building in Looker Studio

Additional information for KPIs:

Department Number: Represent the number of department in our data (10)

Satisfaction Level: Represent the average satisfaction of our employees (0.5)

Total Years: Represent the average years spend at the company of our employees (3.39)

Last Evaluation: Represent the average evaluation (0.47)

Page updated

Google Sites

Report abuse