Semantically meaningful sample databases for constraint verification : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computer Science at Massey University, Manawatu, New Zealand
| dc.confidential | Embargo : No | |
| dc.contributor.advisor | Koehler, Henning | |
| dc.contributor.author | Ye, Wendy (Weimin) | |
| dc.date.accessioned | 2026-06-09T22:26:38Z | |
| dc.date.issued | 2026 | |
| dc.description.abstract | To ensure high-quality relational database design, theoretical developments are crucial, particularly in measuring constraints. Keys, functional dependencies (FDs), and inclusion dependencies (INDs) play pivotal roles and are practical for conveying data semantics in relational databases. However, ensuring the validity of constraints can be challenging when specified manually or mined from existing database instances. Example databases serve as valuable tools for inspecting the properties of proposed designs. Known in the literature as Armstrong databases, these databases precisely adhere to a specified set of dependencies and their logical consequences, excluding any additional dependencies not implied by them. Research efforts have focused on proving the existence and exploring the structure of Armstrong databases. The size of Armstrong databases is equally important; larger databases can be impractical as overlooked constraints become harder to detect. Moreover, when NULL values are involved, constructing Armstrong databases becomes more complex, requiring additional steps. Therefore, constructing a compact Armstrong database is challenging but essential for practical utility. Our study aims to improve existing methods by investigating and constructing Armstrong databases from scratch. This includes adapting existing real-world databases into Armstrong databases to enhance their utility and effectiveness in practical applications. By focusing on reducing the size of Armstrong databases, we aim to streamline the process of constraint validation and database design optimization. This research contributes to advancing the field by addressing the practical challenges of database constraint management and fostering the development of more efficient and reliable relational database designs. | |
| dc.identifier.uri | https://mro.massey.ac.nz/handle/10179/74545 | |
| dc.publisher | Massey University | |
| dc.rights | © The Author | |
| dc.subject | Compact Armstrong Database Construction | |
| dc.subject | Relational databases | |
| dc.subject | Design | |
| dc.subject | Management | |
| dc.subject | Databases | |
| dc.subject | Semantics | |
| dc.subject | Semantic computing | |
| dc.subject.anzsrc | 460503 Data models, storage and indexing | |
| dc.title | Semantically meaningful sample databases for constraint verification : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Computer Science at Massey University, Manawatu, New Zealand | |
| thesis.degree.discipline | Computer Science | |
| thesis.degree.name | Doctor of Philosophy | |
| thesis.description.doctoral-citation-abridged | Ms Wendy Ye investigated methods for constructing compact Armstrong tables that faithfully represent relational database constraints. Her research developed practical techniques for reducing table size while preserving semantic correctness, improving the usefulness of Armstrong tables for database design, constraint validation, and data quality assurance. | |
| thesis.description.doctoral-citation-long | Ms Wendy Ye investigated the construction of compact Armstrong tables for relational database constraints, including keys, functional dependencies, and databases containing NULL values. Armstrong tables provide a practical means of validating and communicating data constraints in relational database design. Ms YE developed methods to reduce table size while preserving semantic correctness and extended Armstrong table construction to handle incomplete information through possible and certain key semantics. Her research demonstrated effective techniques for generating small, informative, and constraint-faithful Armstrong tables, improving their applicability to database design, data quality assurance, and automated reasoning. | |
| thesis.description.name-pronounciation | Wendy YEH |