IBM-HR-Analytics

This repository explores the factors that affect employee performance and attrition. Different classifiers were tested to predict both employee attrition and performance.

Description of the dataset can be found at this link: https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset.

Requirements

1. Tensorflow==2.0
2. Keras==2.4.3
3. Seaborn==0.11.0
4. Scikit-learn==0.23.1
5. matplotlib
6. pandas
7. imblearn 

Summary of worklfow

1. Clean and Normalize Data
2. Conduct exploratory data analysis on dataset
3. Feature Engineering and selection to select best features to be used in classifiers  
4. Create different classifiers to predict attrition and performance based on different features.
   The following classifiers were compared: i) Logistic Regression
                                            ii) Decision Trees
                                            iii) Neural Networks
 
5. Compare performance of different classifiers.