Semistructured database design.
Semistructured data are especially applied in web-applications. Due to the growing number of advanced web-applications and the often huge amount of applied semistructured data, there is a strong need for well designed semistructured databases. A well designed semistructured database should avoid update anomalies and support a good query performance. It is known, from relational databases, that reaching this combined goal is not quite simple. The authors of the book under review have used some material that they have published at international conferences since 2001. This means, their approach has already been discussed and reviewed by specialists in the field. The authors themselves appropriately address their approach as follows: ``A process that initially takes a hard line approach against redundant data, and then relaxes the approach for gains in query performance. The book is written in English. It is recommended to both practitioners and researchers/stu\-dents, who have to deal with the design and storage organization of semistructured data. It will make reading the book easier if the reader is familiar with (relational) database design and has knowledge of XML. The contents: 1. The introduction explains the motivation and goal of the book. 2. ``Data models for semistructured data evaluates known data models for modeling the nested structure of semistructured data (OEM, DOM, DataGuides) and such ones for data management (S3-graphs, CM Hypergraphs, EER and XGrammar, XML-trees, ORA-SS). The evaluation is accompanied by applying the same example to all models. 3. ``ORA-SS elaborates on the details of ORA-SS. The four diagrams ``schema diagram, instance diagram, functional dependency diagram and inheritance hierarchy diagram are illustrated. The major strengths of ORA-SS are that it can model concepts central to the semi-structural data model, such as references, ordering on object classes and attributes, and document instances, participation constraints on all object classes participating in a relationship type, the inheritance hierarchy between object classes, and the ability to distinguish between attributes of relationship types and attributes of object classes. 4. ``Schema extraction describes an algorithm to extract a schema from an XML data instance. The algorithm can deal with the gaps in semantic information during the extraction process. It will indicate which schema information cannot be derived and ask questions to fill the gap. The extracted schema can be translated to DTD, XML Schema and RELAXNG. 5. ``Normalization helps to eliminate redundant data, defines a normal form based on the ORA-SS data model to enable the mapping of an ORA-SS schema diagram to a normal form ORA-SS schema diagram and compares the proposed normal form with existing ones. 6. ``Views presents an approach to design and query semistructured views based on the ORA-SS conceptual model. Furthermore, four transformation operations are described to create XML views. The approach ensures that the view design preserves the semantics in the underlying source data. 7. ``Physical database design investigates the various types of redundancy for semistructured data repositories. The goal is to be able to evaluate which types of redundancy can be allowed in a special case to increase query performance. It is shown how the cost of redundancies can be computed. Guidelines are presented which include normalization and relaxation of normalization rules. A mapping from ORA-SS schema diagram to the nested relational model is described finally to ensure efficient and consistent storage. 8. ``Conclusions gives a summary about the steps to design a repository for semistructured data and the motivation why ORA-SS is a good foundation for modeling semistructured data.
- Designing semistructured databases: A conceptual approach
- scientific article; zbMATH DE number 108406
- scientific article; zbMATH DE number 3904003
- Database modeling and design: logical design.
- scientific article; zbMATH DE number 964539
- scientific article; zbMATH DE number 108405
- scientific article; zbMATH DE number 41736
- XML data manipulation in conventional and temporal XML databases: a survey
- Search optimization in semistructured databases using hierarchy of document schemas
- On the equivalence between FDs in XML and FDs in relations
- scientific article; zbMATH DE number 108406 (Why is no real title available?)
- NF-SS: A normal form for semistructured schema
- Ozone: Integrating structured and semistructured data
- Designing semistructured databases: A conceptual approach
- Database Lexicography
- scientific article; zbMATH DE number 1792116 (Why is no real title available?)
- scientific article; zbMATH DE number 2241906 (Why is no real title available?)
- The Web Resource Space Model
- Theorem prover approach to semistructured data design
This page was built for publication: Semistructured database design.
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4652394)
