Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-06-26
2001-02-20
Hong, Stephen S. (Department: 2776)
Data processing: database and file management or data structures
Database design
Data structure types
Reexamination Certificate
active
06192379
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to data modelling and finds particular application in the compilation of data models.
2. Related Art
In modelling an aspect of a particular field in data, a data modelling application typically requires a modelling environment in which a predetermined data manipulation component may be constructed with the purpose of operating on variable input data to produce a data output. A complex data model may comprise many levels of data manipulation, combining data from different levels of input, internal and external, in a web of linkage and calculation, possibly interfacing to other data models.
Data modelling applications may often be satisfactorily implemented using personal computers running proprietary spreadsheet products. Data models may be created in spreadsheet form either as stand-alone models or within a client server architecture operating over a network in which models may be created and used in a shared environment.
Proprietary spreadsheet and other data modelling products support a number of basic features to enable data models of varying size and complexity to be assembled. They provide a highly flexible means of model construction, but the process can be very time consuming and error-prone, without a very comprehensive design to work from. Sometimes a model is called for to provide rough cost estimates, for example; something that can be put together rapidly, perhaps building on past experience or relying on proven assumptions to yield a reliable result. Information on the requirements or assumptions for a data model may be initially sketchy or the accuracy of information on which a model is to be initially based is only within broad tolerances. To build up even a top level model using only the basic facilities of a generally available spreadsheet product may take too long to be useful.
It has been observed that different data modelling applications often use similar blocks of formulae. Even complex data models will often replicate a relatively small number of similar formulae to manipulate data, but corresponding economies in time and effort of construction and in susceptibility to error are often difficult to achieve in practice.
SUMMARY OF THE INVENTION
According to the invention there is provided a computer system comprising data processing means connected to:
memory means having a control program stored therein;
a store;
an input interface;
an output interface; and
a reference store containing a plurality of pre-defined program components each having a plurality of fields, wherein each component has a field containing a label unique to that component and wherein at least one component has a field containing information defining a computation to be performed;
wherein the data processing means are arranged, in operation, under control of said control program:
(a) in response to user selection via the input interface to copy selected ones of said program components into the store;
(b) in response to user commands via the input interface to enter into the store link information determining the transfer of data from a field of one program component stored in the store into a field of another such component;
(c) to perform computation as defined by the program components and link information stored in the store and to output results thereof via the output interface;
(d) in response to the label field of a program component stored in the store, to compare a field of the component with a corresponding field of the program component stored in the reference store and having the same label, and to identify a field shown by said comparison to be different.
The invention seeks to verify a particular type of data model with respect to a reference, the data model preferably having being constructed using a particular type of component part, known for the purposes of this invention as a “piece”, and a particular method of model assembly from such “pieces”.
A piece typically comprises inputs, outputs and pre-defined formulae and is intended to model only one type of entity likely to arise in a particular modelling application. A range of pre-defined pieces, designed and constructed to model a range of associated entities, may be used as the basis for a method of model assembly supporting a wide range of data modelling applications.
Preferably a reference library of pre-defined pieces is assembled to support particular data modelling requirements. A library may contain pieces modelling both generally useful entities and more specialised entities such that data modelling at different levels of detail may be achieved by copying and linking pre-defined pieces from the reference library. The library is intended to provide a single controlled source of pieces to ensure consistency in modelling assumptions and parameters and greater integration of modelling activities across common areas of application. It is intended that a formula or assumption in a model may be traced back to a single corresponding master copy in the library, and that it is clear whether or not the master formulae and assumptions in the library have been updated or otherwise altered since the model was created.
The invention seeks to provide a means to automatically compare particular fields of pieces contained within a data model with corresponding reference versions of those pieces stored in a library. The comparison function highlights those fields compared and found to differ from the reference versions. Such an comparison facility is likely to be of particular value in relation to large and complex data models in which changes or accidental damage to formulae or key parameters in particular fields can be very difficult to identify, even with a highly structured model building approach as described herein.
Preferably, means are included to automatically update a particular field in a data model, as required, with the contents of the corresponding field of the reference version of the piece in the library, depending upon the result of the comparison.
Particular benefit has been derived from the comparison of those fields of a model piece containing formulae with the corresponding reference versions of those formulae in the library piece. Parameter data held in a model piece may also be verified with respect to the corresponding reference piece.
REFERENCES:
patent: 5303146 (1994-04-01), Ammirato et al.
patent: 5309296 (1994-05-01), Grandall et al.
patent: 5603021 (1997-02-01), Spencer et al.
patent: 5604854 (1997-02-01), Glassey
patent: 5633998 (1997-05-01), Schlafly
patent: 5634124 (1997-05-01), Khoyi et al.
patent: 5883623 (1999-03-01), Cseri
British Telecommunications public limited company
Hong Stephen S.
Nixon & Vanderhye P.C.
LandOfFree
Data model compiler selectively using a reference store of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Data model compiler selectively using a reference store of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Data model compiler selectively using a reference store of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2611762