What the study found
An adapted DBSCAN clustering method was used to early identify configuration items (CIs) in an enterprise information system. The study reports that the method formed cluster sets describing monolithic, modular, and service-oriented architectures.
Why the authors say this matters
The authors conclude that automating this kind of identification could help synthesize descriptions of information system architecture. They say this would improve the quality of information system development by identifying a smaller set of architectural entities than the full set of elementary functions.
What the researchers tested
The researchers adapted DBSCAN, a density-based clustering algorithm, to identify configuration items early in the functional task “Formation and maintenance of an individual plan of a scientific and pedagogical employee at a department.” They used 10 functions and 12 database entities as initial configuration items and compared DBSCAN with Divisive Analysis, Agglomerative Nesting, Chameleon, and k-means using the criteria “Cumbersome solution” and “Identification of separated CIs.”
What worked and what didn't
DBSCAN produced one cluster for monolithic and modular architectures and two clusters for service-oriented architecture, and it detected separated configuration items. According to the comparison, these results were the best for the selected group of clustering methods and algorithms on the stated criteria.
What to keep in mind
The abstract does not describe detailed limitations beyond the specific task and data used. The findings are reported for one functional task and the selected comparison methods, so the scope in the abstract is limited.
Key points
- An adapted DBSCAN clustering method was used for early identification of configuration items in an enterprise information system.
- The method produced cluster sets corresponding to monolithic, modular, and service-oriented architectures.
- The study compared DBSCAN with Divisive Analysis, Agglomerative Nesting, Chameleon, and k-means.
- DBSCAN was reported to detect separated configuration items.
- The abstract says the results were best on the criteria “Cumbersome solution” and “Identification of separated CIs.”
Disclosure
- Research title:
- DBSCAN identified configuration items early in an IT system
- Authors:
- Adrian Ye. Kozhanov, Maksym Ievlanov
- Institutions:
- Kharkiv National University of Radio Electronics
- Publication date:
- 2026-02-27
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.