Full-form lexicon with tagged data and methods of constructing and using the same
AssigneeMicrosoft Corporation
InventorsKevin Powell, Andrea Jessee, Douglas Potter
Life of this patent
Estimating the term…
Abstract
A lexicon stored on a computer readable medium and used by language processing systems. The lexicon can store word information in a plurality of data fields associated with each entered word. The data fields can include information on spelling and grammar, parts of speech, steps that the entered word can be transformed into another word, a word description, and a segmentation for a compound word. Information that cannot be stored in the lexicon can be stored in an intermediate indexes table. Associated methods of constructing, updating and using the lexicon are introduced.
Claims40
A computer readable medium having a lexicon for storing word information and adapted for use by a language processing system, the lexicon comprising a plurality of entries, each entry corresponding to a word entered in the lexicon, wherein each entry comprises:
- a first data field comprising spelling information for an entered word;
- a second field comprising part of speech information associated with the entered word;
- and
- a third field comprising lemma delta information associated with the entered word.
A lexicon stored on a computer readable medium, the lexicon comprising information for entered words, wherein for each entered word, corresponding word information is stored in data fields, the data fields comprising:
- a spelling and dynamic segmentation field related to the entered word;
- a part of speech field related to the entered word;
- a lemma delta field related to the entered word;
- a description field for the entered word;
- and
- a static segmentation mask field for the entered word.
A method of constructing a lexicon comprising information about words, for each word, the method comprising steps of:
- storing spelling and dynamic segmentation information;
- storing part of speech information;
- and
- storing lemma delta information.
Dependencies are inferred from the claim text.