Efficient skyline query processing in incomplete graph databases using machine learning techniques
Skyline queries play a critical role in multi-criteria decision-making systems by retrieving non-dominated data points from large datasets. In recent years, the rapid growth of graph-structured data across various domains has introduced challenges in efficiently processing skyline queries over incom...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
IIUM Press
2025
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/122467/2/122467_Efficient%20skyline%20query%20processing%20in%20incomplete.pdf http://irep.iium.edu.my/122467/ https://journals.iium.edu.my/kict/index.php/IJPCC/article/view/595 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Skyline queries play a critical role in multi-criteria decision-making systems by retrieving non-dominated data points from large datasets. In recent years, the rapid growth of graph-structured data across various domains has introduced challenges in efficiently processing skyline queries over incomplete and large-scale graph databases. Processing skyline queries in such massive, incomplete graphs is computationally intensive due to missing values and high-dimensional data. Traditional techniques often fail to scale or effectively handle data imperfections. There is a pressing need for a scalable, intelligent framework that can manage missing data, reduce computational overhead, and improve skyline query efficiency. This study adopts the Design Science Research Methodology (DSRM) to design and implement an optimisation framework that integrates machine learning techniques, including domination score ranking, dimension-based filtering, K-Means clustering and quicksort. These methods collectively reduce the search space and redundant comparisons. Experimental evaluation on real graph datasets demonstrates significant improvements in skyline computation time and accuracy, with clear reductions in pairwise comparisons and improved processing efficiency on large-scale graphs. By leveraging machine learning techniques for sorting, filtering and clustering, the approach reduces computational complexity and enhances scalability. These results show promising directions for applying intelligent query optimization in big data environments |
|---|