Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2000-07-07
2002-03-12
Mizrahi, Diane (Department: 2171)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000
Reexamination Certificate
active
06356897
ABSTRACT:
BACKGROUND—FIELD OF INVENTION
This invention relates to a unique indexing system, record structure, and linking methodology for bringing a high level of organization and flexibility to assemblies of information stored on electronic-based media.
BACKGROUND—DESCRIPTION OF PRIOR ART
Today, advances in computer technology have provided a plurality of temporary and permanent storage devices to store, organize, access, and manipulate a variety of data. Such devices include ferromagnetic rotating disks, magneto-optical disks, optical disks, dynamic random access memory, and static random access memory herein below referred to as electronic-based media. Historically, most data stored on electronic-based media has been organized and accessed using a variety of systems such as hierarchical models, network models, and list models. All provide a set of capabilities which recommend them for the solution of specific information processing problems. One of the most common list models is the relational table with data normalization and linking between records stored in separate tables to create representations of real world processes such as accounting and inventory systems.
However, the very advantages created by the present relational model has created a new set of problems. The growing storage capacity and processing power of computers to run the relational data model has lead to an accelerating appetite for feeding ever-larger volumes of ever-more detailed information into them in an effort to gain additional control over management processes and compete more effectively in the market place. The increased volume and detail of the information gathered has, in turn, lead to a growing requirement for a more flexible and powerful method of ordering, linking, and accessing the data in order to facilitate the extraction of knowledge from the underlying information by rapidly reorganizing it, adding new types of information, and defining new types of relations between informational items.
A review of prior art reveals that much attention has been paid to optimizing present database structures for speed and functionality within the strict limits of said hierarchical, networking, and relational models. Bachman (U.S. Pat. No. 4,631,664) utilizes a network data model to pair data elements together and to define their relationship. Thurman et al. (5,253,361) utilizes a composite index storing arrays of identifiers to locate pages of linked records. Leenstra, Sr. et al. (U.S. Pat. No. 5,303,367) introduces the concept of storing all data elements in a single tabular structure and then linking data together in a data array structure. Perez (U.S. Pat. No. 5,592,666) introduces the concept of a multidimensional array of database pointers to locate data nodes in an hiearchical data structure. Doktor (U.S. Pat. No. 5,604,899) defines a data relationships processor that allows a user to create additional, direct relational links between two tables in a data set without having to restructure the data dictionary. Bennett et al. (U.S. Pat. No. 5,615,367) provide a system that provides an intelligent guess as to which common fields are to be linked during the design of a data set using a visual interface. Sharif-Askary et al. introduce the concept of creating a knowledge-base that defines all possible entity relationships in a data set. Hohensee and Sampson (U.S. Pat. No. 5,717,922) introduce the use of a logical link element defined for a group of uniquely identified data objects that defines the type of data in the data objects, the type of links, as well as source and target documents where a link is made.
A review of the literature on database theory and implementation reveals a focus on relational links and joins in data queries between tables of records with or without the use of stored procedures. For example, in An Introduction to Database Systems (Sixth Edition, Addison-Wesley Publishing Company 1995), C. J. Date details the various possible relations between tables of records through primary and secondary indexes. He also introduces the concept of defining a specialized relationship between the two data tables via a link to a third data table with two linking fields (pages 68-70). In A Guide to Data Integrity Rules (PC Magazine, August 1994 pages 387-390) S. Ricciardi provides a comprehensive review of maintaining referential integrity in a relational data model. This includes the important concept of separately defining domains of allowable data types and value ranges for storing in predefined key fields of a table to assure that each record is uniquely identified within said table.
A review of high and low level database programming languages such as Object PAL/Paradox for Windows (licensed to Corel Corp. by Borland International) and Object Pascal/Delphi (Borland International) and database engines such as the Borland Database Engine and Microsoft's Access reveals a standard set of data models, relational procedures, and query language structures. The concept of pairing two unique indicum to provide a link between pairs of data—called a map or an association—can only be found in compiled data containers used in many programs today. This system generates unique identification numbers at run time, assigns them to data stored temporarily in memory and processes this data using said mapping. The most common usage is with arrays and string lists where the array or string value and its ordinal position in the array or list are the paired values. The linking relationship is fixed at compilation time and lasts only as long as the program is running.
These and other prior art reviewed fail to provide a comprehensive, bi-directional, hierarchical, non-hierarchical, recursive, and indirect means to link together any two or a plurality of informational items in an assembly of informational items stored on electronic-based media. In addition, each prior art solution ignores the special requirements for creating knowledge from an assembly of informational items where the relationships between the informational items do not fall within the strict relational model. For example, as presently designed, relational databases and their underlying data dictionaries are highly restrictive requiring the relationship be based on a specific set of one or a plurality of fields in a pair of linked tables. Said model only allows linking at the record level and does not provide a means to link individual fields within said records or to simultaneously link a plurality of fields and/or records located in more than two tables simultaneously in an hierarchical or nonhierarchical structure.
Furthermore, only those linking relationships between tables of records defined during the design phase of the data dictionary are allowed during day-to-day operations. This means that if there is a need to create a new linkage between a different set of records, the entire database must undergo a redesign which is costly in time and money. These design changes must be made by technical personnel who often do not understand the functioning system represented by the data model nor what the end user wants or needs.
Furthermore, the present relational model does not provide a means for introducing new data tables into the database without requiring a major redesign of the underlying data structure, the software program which runs it, and the user interface that allows access to the stored information.
Furthermore, the present relational model does not provide a means for defining and storing index arrays and procedures that join together multiple records from a plurality of tables into a linked data cluster.
Furthermore, the present relational model does not provide a means for predefining links outside said data tables and allowing the interlinking between a plurality of records to be controlled during run time by user changes in stored field values or by predefined algorithms that are initiated by said values and predefined links.
Each of these capabilities and many others not discussed above are key requirements for developing a useful kno
LandOfFree
Associative database model for electronic-based... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Associative database model for electronic-based..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Associative database model for electronic-based... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2841367