CONVERTING NUMERIC VALUES TO STRINGS FOR OPTIMIZED DATABASE STORAGE

Patent №

US 6,907,435

Granted

2005-06-14

Filed 2001

Owner

MICROSOFT CORPORATION

AI components

1

nlp

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

09794867

The present invention provides a conversion method for converting numeric multi-dimensional database dimension members to textual representations. The conversion is specifically designed to enable the textual representations of the numeric dimension members to be properly sorted and converted back into numeric representations with accuracy. All numeric and date data is transformed during conversion, such that the textual representation of the numeric and date data is properly sorted as a text string, in the same order as it would be sorted as a number. The present invention also provides a means for accurate reverse-conversion (i.e., from textual representation to numeric representation) by first converting into text the information necessary to support the precision required by the relevant convention of the numeric representation. The present invention also provides a convention for determining the data type of the textual representations of values of various data types. The textual representation of mixed types is configured so that the sort order of different data types and the text string for numeric types will sort by value independent of the numeric data-types.

Natural languageG06F 16/258Y10S 707/917Y10S 707/957Y10S 707/99932Y10S 707/99937Y10S 707/99954

AI classification

Natural language0.99
Knowledge representation0.01
AI hardware0.01
Planning0.00
Vision0.00
Speech0.00
Machine learning0.00
Evolutionary computation0.00

Ownership

MICROSOFT CORPORATION

assignment · 115820579

Assignors

COULSON, MICHAEL J., WORTENDYKE, DAVID, GREALISH, KEVIN

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

© 2026 NYSGPT2525 LLC