Akaryakıt sektöründe veri madenciliği uygulamaları her geçen gün gelişmekte ve yaygınlaşmaktadır. Sektörde kullanılan yöntemler ve yapılan çeşitli analizler sayesinde, akaryakıt hırsızlığı, operasyonel anormallikler, dolum sırasında meydana gelen miktar aşımları ve aşırı dolum sonrası yaşanan taşma gibi kritik konular izlenerek gerekli aksiyonlar alınmaktadır. Bu çalışmada, bir petrol şirketinin verileri kullanılarak daha önceden belirlenmiş dört önemli kritik kategori için sınıflandırma çalışması gerçekleştirilmiştir. Veri setine ön işleme uygulanmış, analize katkısı olmayan değişkenler veri setinden çıkarılmış ve eksik veriler tamamlanarak analiz için uygun bir hale getirilmiştir. Uygulama aşamasında RAPIDMINER (v.9.10) yazılımından yararlanılmıştır. Veri madenciliği sınıflandırma yöntemlerinden ken yakın komşu algoritması, Rastgele Orman Algoritması, Gradient Boosted Algoritması, ADABOOST Algoritması ve Karar Ağacı (J48) Algoritması kullanılarak sınıflandırma işlemleri gerçekleştirilmiş ve modellerin başarısı çeşitli ölçütler kullanılarak değerlendirilmiştir.
Data mining applications in the fuel industry are advancing and becoming more widespread with day by day. Through the methods and various analyses used in the sector, critical issues such as fuel theft, leaks, quantity overruns during refueling, and overflow after overfilling are monitored, and necessary actions are taken. In this study, classification analysis was conducted for four predefined critical categories using the data of a petroleum company. Preprocessing was applied to the dataset, irrelevant variables that did not contribute to the analysis were excluded, and missing data were completed to make the dataset ready for analysis. The RAPIDMINER (v.9.10) software was utilized during the implementation phase. Classification methods in data mining, including the k-nearest neighbors algorithm, Random Forest Algorithm, Gradient Boosted Algorithm, ADABOOST Algorithm, and Decision Tree (J48) Algorithm, were employed to perform classification. The performance of the models was evaluated using various success metrics.
| Primary Language | Turkish |
|---|---|
| Subjects | Software Testing, Verification and Validation |
| Journal Section | Research Article |
| Authors | |
| Submission Date | July 2, 2025 |
| Acceptance Date | March 8, 2026 |
| Publication Date | April 10, 2026 |
| DOI | https://doi.org/10.17482/uumfd.1733355 |
| IZ | https://izlik.org/JA55UD84BX |
| Published in Issue | Year 2026 Volume: 31 Issue: 1 |
Announcements:
30.03.2021-Beginning with our April 2021 (26/1) issue, in accordance with the new criteria of TR-Dizin, the Declaration of Conflict of Interest and the Declaration of Author Contribution forms fulfilled and signed by all authors are required as well as the Copyright form during the initial submission of the manuscript. Furthermore two new sections, i.e. ‘Conflict of Interest’ and ‘Author Contribution’, should be added to the manuscript. Links of those forms that should be submitted with the initial manuscript can be found in our 'Author Guidelines' and 'Submission Procedure' pages. The manuscript template is also updated. For articles reviewed and accepted for publication in our 2021 and ongoing issues and for articles currently under review process, those forms should also be fulfilled, signed and uploaded to the system by authors.