EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Integrated Anomaly Detection in Automated Cloud Data Engineering Frameworks

Prof. Dr. Utku Kose

Scientific Journal of Artificial Intelligence and Blockchain Technologies · 30.04.2026 · 3 (2)

DOI: 10.63345/sjaibt.v3.i2.102 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific Journal of Artificial Intelligence and Blockchain Technologies, online erschienen am 30.04.2026. Verfasst von Prof. Dr. Utku Kose. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper gives an example of an automated data engineering solution that focuses on large-scale cloud analytics and attempts to solve issues such as high data velocity, variable workloads, and numerous sources of heterogeneous data. The proposed model aims to achieve maximum performance and improved reliability, while saving on manual steps by utilizing a single pipeline combining automated ingestion control, adaptive data transformation, and resource optimization in a closed-loop manner. The model has been tested to provide a data ingestion throughput of 6.7 units in comparison to traditional cloud analytics platforms, and thus improving analytics throughput by up to 28 percent, and reducing end to end pipeline latency to 360 units. In terms of tested variable workloads, the proposed model showed an average of 94 percent for scalability, and 9 units for elasticity defined as the immediate response to adaptive demand by the model. The model showed data processing consistency of 97.9 percent and accuracy of 97.2 percent through confidence-based filtering, and early validation and schema alignment, in the end reducing automation overhead to 11 percent, demonstrating a high degree of operational autonomy. Enhanced system reliability and availability improved to 99.5 percent and fault recovery time was recorded at 16 units boosting system robustness. In improved resource utilization, the model achieved 91 percent balancing optimal arrangement of computing resources relative to workload, and overall reducing operational resource expenditure. In general, the findings support that the proposed framework provides a reliable, cost-effective, and scalable foundation for next-generation cloud analytics systems in dynamic environments.

Zitieren

APA

Kose, P. D. U. (2026). Integrated Anomaly Detection in Automated Cloud Data Engineering Frameworks. Scientific Journal of Artificial Intelligence and Blockchain Technologies, 3(2). https://doi.org/10.63345/sjaibt.v3.i2.102

Vancouver

Kose PDU. Integrated Anomaly Detection in Automated Cloud Data Engineering Frameworks. Scientific Journal of Artificial Intelligence and Blockchain Technologies. 2026;3(2). doi:10.63345/sjaibt.v3.i2.102

BibTeX
@article{Kose2026,
  author = {Kose, Prof. Dr. Utku},
  title = {Integrated Anomaly Detection in Automated Cloud Data Engineering Frameworks},
  journal = {Scientific Journal of Artificial Intelligence and Blockchain Technologies},
  year = {2026},
  volume = {3},
  number = {2},
  doi = {10.63345/sjaibt.v3.i2.102}
}
Kostenlos starten →