Selecting the best way to forecast income in the banking industry using data mining methods, a case study
OPSEARCH, cilt.62, sa.3, ss.1383-1422, 2025 (ESCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 62 Sayı: 3
- Basım Tarihi: 2025
- Doi Numarası: 10.1007/s12597-024-00852-3
- Dergi Adı: OPSEARCH
- Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, ABI/INFORM, INSPEC, zbMATH
- Sayfa Sayıları: ss.1383-1422
- Anahtar Kelimeler: Clustering, Customer revenue, Data science, Forecasting, Prediction
- Ankara Yıldırım Beyazıt Üniversitesi Adresli: Hayır
Özet
One of the most important topics in data science is the prediction topic. The prediction accuracy is often sought after in many articles, aiming to establish trust in the prediction results. Customer revenue prediction is deemed crucial in any industry, as awareness of future income allows for strategic development across various industry sectors. Traditionally, profits and losses were assessed retrospectively following a business event. However, today, future revenue can be predicted using data science concepts. Future revenue prediction can be achieved through the application of different types of data mining algorithms, each possessing its level of accuracy. In this article, the forecasting of bank customers’ revenue will be examined using various algorithms. To this end, different forecasting methods such as regression, decision tree, random forest, cart, etc., will be implemented, both with and without clustering, to identify the optimal approach for predicting customer revenue.