singhvks / Employee_Attrition

PREDICTIVE ANALYTICS - LOGISTIC REGRESSION . Predicting employee attrition using HR data

Employee_Attrition

Predicting employee attrition using HR data

Business Requirement

Predict the employee attrition using the HR data.

ML Requirement

Attrition is the target variable represented as a Boolean field. Using features provided in the CSV file predict whether the employee will leave the organisation or not.

Assumptions

1. Since we do not have test data separately, we will use the Train-Test Split from SciKit-Learn for generating the test data (used to compute the performance metrics).
2. Data is fictional and can result in a very low accuracy score- we will then opt for some-other metrics for model evaluation.
3. There is no Time-Series data for the attrition of employees which will result in some deviation.

Contents

1. Correlation Heatmap - to explain the relation in various parameters
2. Project Report - tracking the status and lifecycle of project
3. Employee Attrition Predictor - IPython notebook
4. HR-Employee-Attrition - Data file taken from https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset
5. Visualization plots are added in file Visualization

Algorithms used

1. Logistic Regression
2. Random Forest

Encoder used

1. Label Encoder
2. One Hot Encoder

Note:

I welcome any suggestions or recommendation to make this project better. Also I am working on the visualization part too.