KMAN Publication Inc. (KMANPUB) International Journal of Innovation Management and Organizational Behavior (IJIMOB) 3041-8992 6 Serial Number 24 2026 01 01 A Machine‑Learning Approach to Employee Change Innovation: Roles of Adaptability and Job Crafting A Machine‑Learning Approach to Employee Change Innovation: Roles of Adaptability and Job Crafting 1 11 10.61838/kman.ijimob.5256 EN Journal Article 2025 07 15 <table> <tbody> <tr> <td> <p><strong>Objective:</strong> The objective of this study is to employ an advanced machine-learning approach to precisely predict employee change innovation by computationally delineating the complex, non-linear predictive roles of psychological adaptability and multidimensional job crafting behaviors.</p> <p><strong>Methods and Materials:</strong> A cross-sectional quantitative design was utilized, surveying full-time professionals in Turkey experiencing organizational change, selected via purposeful and snowball sampling. Data were collected using validated self-report digital questionnaires evaluating adaptability, job crafting, and change innovation (Cronbach’s ). The dataset was split into an training set ( ) and a testing set ( ). An advanced machine learning pipeline evaluated Support Vector Regression (SVR), Random Forest (RF), and Gradient Boosting Machine (GBM) algorithms, optimizing hyperparameters via 10-fold cross-validation.</p> <p><strong>Findings:</strong> The GBM model demonstrated the highest predictive accuracy on the unseen test set ( , , ), outperforming both RF ( ) and SVR ( ). Feature importance analysis extracted from the GBM revealed that the job crafting dimension of “increasing structural job resources” was the paramount predictor ( ), followed closely by the adaptability dimension of “confidence” ( ). “Increasing challenging job demands” ( ) and “control” ( ) were also substantial drivers. Conversely, traditional demographic variables such as organizational tenure ( ) and age ( ) provided minimal predictive utility.</p> <p><strong>Conclusion:</strong> Proactive resource-seeking and psychological confidence are the fundamental micro-level drivers of employee innovation during organizational transitions, highlighting the superior capacity of machine learning to map complex, non-linear behavioral dynamics.</p> </td> </tr> </tbody> </table> https://journals.kmanpub.com/index.php/ijimob/article/download/5256/9546