EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Predictive Analytics for Understanding Consumer Purchase Behaviour and Optimizing Personalized Marketing Strategies

Dr. S. Md Shakir Ali, Dr Ashish M Puranik, Dr. Panduarng Sudamrao Dabhade, Dr. Deepika Bandil, Dr. Bahman Moghimi, Dr khushali Oza

Computer and Decision Making: An International Journal · 12.11.2026 · 4 (1) · 79-87

DOI: 10.1922/comdem.v4i1.19079 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer and Decision Making: An International Journal, online erschienen am 12.11.2026. Verfasst von Dr. S. Md Shakir Ali und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In this study, the authors explored how this achieved using predictive analytics and machine-learning methods to gain insights that will inform personalized marketing approaches. The data comprised 3,900 customer records with 19 customer behavioural, transactional, and demographic variables, which were analysed using descriptive statistics, exploratory statistics, K-Means clustering, regression, classification and feature-importance methods. The purchasing amount and the purchasing category were analyzed as key behavioural outcomes, and the customer characteristics and transactional attributes were analyzed as predictors. The findings revealed significant differences in the expenditures made, the products previously purchased, ratings of reviews, frequency of purchases, product preferences, and seasonal buying patterns. K-Means clustering was used to determine six customer segments with different spending and engagement patterns, including high spenders, high-frequency customers and low-value customers. XGBoost performance was best among regression models and negative R² values suggested that its predictive power was restricted about the individual purchase amount. Modest performance was also obtained in the case of classification models, implying that product-category preferences were not well explained by the variables available. Review rating, color, age, and previous purchases were relatively influential predictors, as determined by feature-importance analysis. Overall, the results suggest that the dataset is more effective for consumer profiling and segmentation than for high-accuracy prediction but is leveraged for targeted promotions, loyalty strategies, product recommendations, and customer-retention planning.

Zitieren

APA

Ali, D. S. M. S., Puranik, D. A. M., Dabhade, D. P. S., Bandil, D. D., Moghimi, D. B., & Oza, D. K. (2026). Predictive Analytics for Understanding Consumer Purchase Behaviour and Optimizing Personalized Marketing Strategies. Computer and Decision Making: An International Journal, 4(1), 79-87. https://doi.org/10.1922/comdem.v4i1.19079

Vancouver

Ali DSMS, Puranik DAM, Dabhade DPS, Bandil DD, Moghimi DB, Oza DK. Predictive Analytics for Understanding Consumer Purchase Behaviour and Optimizing Personalized Marketing Strategies. Computer and Decision Making: An International Journal. 2026;4(1):79-87. doi:10.1922/comdem.v4i1.19079

BibTeX
@article{Ali2026,
  author = {Ali, Dr. S. Md Shakir and Puranik, Dr Ashish M and Dabhade, Dr. Panduarng Sudamrao and Bandil, Dr. Deepika and Moghimi, Dr. Bahman and Oza, Dr khushali},
  title = {Predictive Analytics for Understanding Consumer Purchase Behaviour and Optimizing Personalized Marketing Strategies},
  journal = {Computer and Decision Making: An International Journal},
  year = {2026},
  volume = {4},
  number = {1},
  pages = {79--87},
  doi = {10.1922/comdem.v4i1.19079}
}
Kostenlos starten →