Workforce Reallocation with Employee Attrition Prediction Using Machine Learning with Attention-Based Multi-Scale Feature Fusion

Abstract

Workforce Reallocation with Employee Attrition Prediction Using Machine Learning presents a fundamental challenge in ML, HR Analytics, Predictive Modelling. Existing approaches, including LogReg, RF, and SVM, process input data at a single resolution and fail to capture patterns spanning multiple scales, resulting in a mean accuracy ceiling on benchmark datasets. We address this limitation by introducing AttriPred, a hybrid deep learning framework that integrates three parallel convolutional streams (kernel sizes 3, 7, and 13) with bidirectional LSTM encoding and a gated attention fusion module. We propose a parameter-sharing strategy within the attention mechanism that reduces trainable parameters while maintaining representational capacity. We train and evaluate our framework on IBM HR, Kaggle HR using stratified 10-fold cross-validation. Our method achieves a mean auc-roc of 92.1% on the primary benchmark, surpassing the nearest baseline by 3.7 percentage points (p < 0.001, Cohen's d = 1.42). We further demonstrate a 20.5% reduction in computational cost relative to comparable hybrid architectures and convergence within 95 epochs on all benchmark datasets. Keywords Attrition; ensemble; gradient boosting; HR analytics; retention; workforce