Random Forest Classifier for Employee Attrition Prediction

    This code implements a Random Forest Classifier to predict employee attrition in an HR dataset. The goal is to assess and understand the factors that contribute to employee turnover.

Importing Necessary Libraries:

• pandas: Used for data manipulation and analysis.
• matplotlib.pyplot: For data visualization.
• train_test_split: Splits the dataset into training and testing sets.
• LabelEncoder: Encodes categorical labels into numerical values.
• RandomForestClassifier: Implements the Random Forest classification algorithm.
• confusion_matrix: Measures the performance of the classification model.
• accuracy_score: Computes the accuracy of the model.
• SMOTE: Synthetic Minority Over-sampling Technique for dealing with imbalanced data.

Data Loading:

• df1: Loads the HR dataset from a CSV file.

Data Preprocessing:

• df: Creates a copy of the dataset for manipulation.
• LabelEncoder: Encodes the 'Attrition' column from text labels to numerical values (0 for 'No', 1 for 'Yes').
• pd.get_dummies(): Converts categorical columns into binary (dummy) variables.
• Dropping columns: Removes columns 'EmployeeCount' and 'EmployeeNumber' as they are not relevant for prediction.

Data Splitting and Balancing:

• X: Features (independent variables).
• y: Target variable (attrition).
• SMOTE: Applies oversampling to balance the dataset.
• train_test_split: Splits the dataset into training and testing sets.

Model Training:

• RandomForestClassifier: Initializes a Random Forest classifier with 100 decision trees.

Model Evaluation:

• y_pred: Predicts attrition for the test set.
• accuracy_score: Calculates the accuracy of the model and prints the result.

Accuracy_Score:

• Our model has accuracy score of 92.71%

Summary:

This code performs data preprocessing, model training, and evaluation using a Random Forest Classifier to predict employee attrition. The accuracy score at the end measures the model's performance in predicting attrition based on the given dataset.