EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An AI-Driven Intelligent Decision-Making Framework for Strategic Management, Digital Commerce, Governance, and Adaptive Leadership in Data-Driven Organizations

Dr. R. Mahalakshmi ., Somya Tyagi, Dr. CA Jayesh Manjrekar, Prof. Dr Ajay Bhushan Prasad, Dr. A. Jagdish Mohan Rao, Mrs. K. Bhavadharani .

Computer and Decision Making An International Journal · 01.11.2026

DOI: 10.1922/comdem.v3i6s.18880 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making An International Journal, online erschienen am 01.11.2026. Verfasst von Dr. R. Mahalakshmi . und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This work provides an AI-powered intelligent decision-making framework for data-driven organizations, combining predictive analytics, explainable machine learning, feature engineering based on behaviour, threshold optimization and decision-intelligence segmentation. The UCI Online Shoppers Purchasing Intention Dataset is used in the empirical evaluation. Once 125 exact duplicate sessions were removed there were 12,205 records (10,297 non-purchase and 1,908 purchase sessions). To the original predictors 6 deterministic behavioural variables were added, yielding 23 model inputs. Stratified cross validation was performed to compare the performance of three classifiers—Logistic Regression, Gaussian Naive Bayes and a Nearest-Centroid probabilistic classifier—where the precision-recall area under the curve (PR-AUC) was the main criterion for model selection. Logistic Regression was the best performing method with a cross validation PR-AUC of 0.6441 and held out PR-AUC of 0.6218 and ROC-AUC of 0.8893. The model achieved an accuracy of 86.48%, balanced accuracy of 78.23%, precision of 55.73%, recall of 66.23%, specificity of 90.24% and an F1-score of 0.6053 at a decision threshold of 0.56 which was derived from the training. Propensity segmentation yielded a strong monotonic gradient with the High segment accounting for about one-quarter of the test sample and 76.18% of all purchases and a conversion rate of 47.55%. PageValues were the most important predictive factor discovered by permutation analysis. The results demonstrate how information from a transparent predictive model can be applied to actionable prioritization for digital commerce without losing sight of human oversight, auditability, and interpretation to ensure governance.

Zitieren

APA

., D. R. M., Tyagi, S., Manjrekar, D. C. J., Prasad, P. D. A. B., Rao, D. A. J. M., & ., M. K. B. (2026). An AI-Driven Intelligent Decision-Making Framework for Strategic Management, Digital Commerce, Governance, and Adaptive Leadership in Data-Driven Organizations. Computer and Decision Making An International Journal. https://doi.org/10.1922/comdem.v3i6s.18880

Vancouver

. DRM, Tyagi S, Manjrekar DCJ, Prasad PDAB, Rao DAJM, . MKB. An AI-Driven Intelligent Decision-Making Framework for Strategic Management, Digital Commerce, Governance, and Adaptive Leadership in Data-Driven Organizations. Computer and Decision Making An International Journal. 2026. doi:10.1922/comdem.v3i6s.18880

BibTeX
@article{2026,
  author = {., Dr. R. Mahalakshmi and Tyagi, Somya and Manjrekar, Dr. CA Jayesh and Prasad, Prof. Dr Ajay Bhushan and Rao, Dr. A. Jagdish Mohan and ., Mrs. K. Bhavadharani},
  title = {An AI-Driven Intelligent Decision-Making Framework for Strategic Management, Digital Commerce, Governance, and Adaptive Leadership in Data-Driven Organizations},
  journal = {Computer and Decision Making An International Journal},
  year = {2026},
  doi = {10.1922/comdem.v3i6s.18880}
}
Kostenlos starten →