The Human Resource Management challenge of predicting employee turnover using CNN-BiLSTM Based Model

There is a lot of concern among many organizations and corporations around employee turnover. This is a critical issue since it threatens the longevity of business strategies, company cultures, and jobs. Human resources departments are increasingly concentrating on employee turnover as a means to better understand the underlying factors and causes. This proposed aims to address that gap by creating a new method for predicting employee turnover using an improved random forest algorithm. Strict adherence to this technique is crucial to attain the greatest potential outcomes throughout model training, feature selection, and preprocessing. Among data preparation, one of the numerous stages of data mining is crucial; this section defines the seven most significant elements. A feature selection strategy based on Intensive Optimized Principal Component Analysis (PCA) is suggested. A CNN-BiLSTM was used to train the model. The proposed method outperforms more conventional approaches like CNN and LSTM. The accuracy was enhanced by 92.41% after implementing the technique.