Leveraging Machine Learning for Enhanced Employee Productivity Insights

Abstract

Employee productivity is vital for modern corporate success, ashighly productive employees perform better and have lower attrition rates. Productivity enhances creativity, customer satisfaction, and cost efficiency. However, measuring the complex factors driving innovation can be challenging, as traditional methods often fail to capture the connections between human traits, workplace dynamics, and organizational culture.To address this, we developed a methodology using two open-source Kaggle datasets—one predicting employee frustration and the other predicting performance. Data preprocessing included numerical conversion, feature scaling, and transforming variables into dummy variables to improve machine learning model performance. Several models, including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and Multi-Layer Perceptron (MLP), were trained and tested. We used 80% of the data for training and 20% for testing, with 5-fold cross-validation to prevent overfitting. Hyperparameter tuning was applied to optimize the models, which were evaluated using precision, recall, F1-score, and accuracy. SVM and DT achieved 100% accuracy on both datasets, with SVM showing superior performance due to its effectiveness in handling high-dimensional classification problems.This structured approach offers valuable insights into the relationships between variables and outcomes, promoting consistency and re producibility in machine learning projects.

Keywords: Hyperparameter-Tuning, One Hot Encoding, 5-fold Cross Validation, Support Vector Machine, PerformanceScoreID.