A Machine Learning Model for Predicting Workplace Spirituality Levels Based on Bhagavad Gita Stress Management Parameters

Spirituality in the workplace is essential in
facilitating the welfare and motivation of employees as well as
harmony in the organization. Nevertheless, the majority of the
current methods are based on subjective evaluation and do not
have quantitative predictive practices. To overcome this
weakness, it proposes a machine learning model, KarmaNet, to
estimate Workplace Spirituality Levels (WSL) based on
Bhagavad Gita-based parameters of stress management, including Equanimity, Karma Orientation, Attachment
Control, Desire Regulation, and Anger Modulation. A special
dataset of 820 records of employees was trained and
preprocessed on the KarmaNet neural framework. The test
accuracy of the model was 92.15%, the validation accuracy was
93.42%, and the Spiritual Harmony Index (SHI) was 0.918,
which was better than the conventional algorithms such as
SVM, Random Forest, and KNN. Findings have shown that
Equanimity and Karma Orientation are the most effective
features in the prediction of spirituality. The proposed
framework represents a new combination of the philosophy of
spirituality and artificial intelligence to increase well-being in
the workplace and forecast human resource analytics.