Conclusion - Information Systems Reengineering, Integration and Normalization - page 372

on each record type of network database (besides the system-owned record types)
and on the non-root segments of the hierarchical database. The result is that these
record types or segment types of the existing nonrelational database can be accessed
like a table.
To convert a relational database to an object-oriented database, we must map the
static data from the relational database to the object-oriented database in schema
translation and data conversion. We then capture the dynamic behavior of each
mapped class by translating each database I/O statement into the operations (meth-
ods) of each class. We have described the schema translation and data conversion
in our methodology. The translation of database programs between the relational
databases and the object-oriented databases is difficult to automate. To convert a re-
lational database into an XML database, we extract an XML view of an EER model,
and load the relational data into an XML document according to the translated XML
schema.

Database System and Expert System Integration System reengineer, broadly
defined as the use of engineering knowledge or artifacts from existing systems to
build new ones, is a technology for improving system quality and productivity.
Much traditional work is focused on the reuse of existing software systems (i.e.,
software programs, files, and databases). Since the use of the knowledge-based sys-
tem is emerging in information systems, many of these systems have been built or
will be built. For knowledge-based systems such as expert systems to make further
contributions to our society, it will be necessary to reuse their knowledge for other
expert systems. The idea of reusing knowledge between expert systems and data-
base systems is an attractive one for similar reasons as the reuse of software. For
example, knowledge from an application for process monitoring may be useful in
an application for training the operators. Furthermore, knowledge must be shared
among different applications.
A reengineering methodology for these systems must capture the information
and the knowledge of the existing systems. Information can be represented by pro-
gramming. Knowledge can be represented by rules. In our methodologies, we have
developed ways to derive and store the knowledge. The rationale behind such a
decision is that a class encapsulates both the static data structure and its feasible
operations (i.e., its dynamic behavior), in its methods. Our reengineering technique
is to map the data structure of the database system into the static data of each class,
and to map the operations of each rule of the expert system into the method of a
corresponding class (i.e., class with the same name).

10.1 Application of Database Conversion Methodologies

The methodologies described in this topic provide an alternative approach for
schema translation in which user input contributes to the process. Direct schema
translation from hierarchical or network into relational cannot guarantee to capture