HIGH-PERFORMANCE EXTENSIBLE DOCUMENT TRANSFORMATION

Patent №

US 6,941,511

Granted

2005-09-06

Filed 2000

Owner

IBM CORPORATION

AI components

2

nlp · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

09653080

The present invention provides a method, system, and computer program product for applying transformations to extensible documents, enabling reductions in the processing time required to transform arbitrarily-structured documents having particular well-defined elements. Signatures for structured document types are defined, along with one or more transformations to be performed upon documents of that type. The transformations are specified using syntax elements referred to as maps. A map specifies an operation code for the transformation to be performed, and describes the input and output of the associated transformation. A special map processing engine locates an appropriate transformation object to a particular input document at run-time, and applies the transformation operation according to the map definition. This technique is preferably used for a set of predetermined core transformations, with other transformations being processed using stylesheet engines of the prior art. The input documents may be encoded in the Extensible Markup Language (XML), or in other structured notations. The techniques of the present invention are particularly well suited to use in high-volume and throughput-sensitive environments such as that encountered by business-to-business transaction servers.

Natural languageAI hardwareG06F 16/258G06F 40/123G06F 40/143G06F 40/154

AI classification

AI hardware1.00
Natural language0.78
Knowledge representation0.29
Planning0.01
Vision0.01
Speech0.00
Evolutionary computation0.00
Machine learning0.00

Ownership

IBM CORPORATION

assignment · 111230979

Assignors

HIND, JOHN R., LI, YONGCHENG, TAN, YIH-SHIN

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC