Chuck Eastman, D. Stott Parker, and Tay-Sheng Jeng
The purpose of this work is to develop automatic methods of semantic integrity maintenance, in support of concurrent engineering. Semantic integrity relations in any final engineering design are built up incrementally, through the use of different computer applications. Here, the structure of these integrity relations are formalized for representation within a database. When changes to a design have to be made, they can invalidate integrity relations in other parts of the design. Formal methods are defined for identifying what data nad integrity relations are invalidated by any change. Methods for making changes that minimize re-design are described and formalized. Opportunities for using semantic integrity to assess progress on a design are reviewed.appeared in: Research in Engineering Design, 9:3, 125--145, November 1997.
Ping-Yu Hsu and D. Stott Parker
appeared in
Proc. Intnl. Conf. on Data Engineering (DE'95),
Taiwan, R.O.C., March 1995.
Expanded version submitted to ACM TODS also available from the authors.
A generalized quantifier is a particular kind of operator on sets. Coming under increasing attention recently by linguists and logicians, they correspond to many useful natural language phrases, including phrases like: three, Chamberlin's three, more than three, fewer than three, at most three, all but three, no more than three, not more than half the, at least two and not more than three, no student's, most male and all female, etc.Reasoning about quantifiers is a source of recurring problems for most SQL users, and leads to both confusion and incorrect expression of queries. By adopting a more modern and natural model of quantification these problems can be alleviated. In this paper we show how generalized quantifiers can be used to improve the SQL interface.
Ping-Yu Hsu
Ph.D. Dissertation, UCLA Computer Science Department, September 1995.
In this dissertation, we study the incorporation of context into databases. Contexts are viewed as ``databases'', have names, and can be treated as values. As databases, contexts contain both schema and data associated with the schema. Treating contexts as databases allows each context to contain independent information and to be updated independently. With names, a context can be referred to by other contexts. Such a design allows users to specify relations between contexts. As values, the contents or identifiers of contexts can be returned as query values and can be used to form queries. No special syntax is required to query contexts.With these features, we show that databases which can store sets of contexts can handle information that cannot be properly stored by conventional databases. Three directions for incorporating context into databases are shown: contexts as name spaces, contexts as nodes in a specially-designed semantic index system, and contexts as operands of generalized quantifiers. Each of these directions produces a significant contribution:
- Treating name spaces as contexts inspires the design of a system that is effective for storing evolving and irregular information. The system can also be applied to WWW and provide a possible way to perform distributed set-oriented search on HTML pages.
- Treating nodes as contexts in semantic index systems leads to the design of a query language that allows users to consult any subset of the index systems. Such systems can guide neophyte users search while allowing experienced users to search as if there were no index.
- Treating generalized quantifiers as predicates of contexts allows us to implement a useful subset of generalized quantifiers. The implementation discussed here not only works with contextual DBMS, but can also work with ordinary DBMS.