Method for encoding regular expressions in a lexigon
US6757647B1 · kind B1 · utility
Assignee
Inventors
Key dates
| Filing date | Jul 30, 1998 |
| Grant date | Jun 29, 2004 |
| Priority date | — |
| Expiry date | Jul 30, 2018 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06F40/12
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Systems and methods are described for concisely encoding into a lexicon (or dictionary) and decoding from the lexicon regular expressions that can represent certain huge word lists that might otherwise be considered unmanageably large. Sets of words (character sequences or ‘strings’) that share certain commonalities such as a set of numbers, which share common digits, may be condensed into digital lexicons by representing the set with a regular expression. The regular expression is a string that includes meta-character, where each meta-character is a place-marker that represents a set of at least two normal characters. When accessing or searching the lexicon, the regular expressions are dynamically expanded, as needed, to the underlying, original word list. The methods disclosed are applicable to many lexicon driven language based systems such as spelling verification systems, handwriting recognition systems, speech recognition systems and the like.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.