TEXT PREDICTION

• US 20120259615A1
• Filed: 04/06/2011
• Published: 10/11/2012
• Est. Priority Date: 04/06/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Stevens, David A., Morin, Frederic
• Granted Patent: US 8,914,275 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

One or more techniques and/or systems are provided for suggesting a word and/or phrase to a user based at least upon a prefix of one or more characters that the user has inputted. Words in a database are respectively assigned a unique identifier. Generally, the unique identifiers are assigned sequentially and contiguously, beginning with a first word alphabetically and ending with a last word alphabetically. When a user inputted prefix is received, a range of unique identifiers corresponding to words respectively having a prefix that matches the user inputted prefix are identified. Typically, the range of unique identifiers corresponds to substantially all of the words that begin with the given prefix and does not correspond to words that do not begin with the given prefix. The unique identifiers may then be compared to a probability database to identify which words have a higher probability of being selected by the user.

Claims

1. A method, comprising:

receiving a first character set comprising one or more characters; and

identifying a first range of one or more unique identifiers respectively corresponding to a word having a prefix that matches the first character set.

2. The method of claim 1, comprising, before identifying a range of one or more unique identifiers, identifying one or more words having a prefix that matches the first character set, the identified first range of one or more unique identifiers respectively corresponding to a respective identified word.

3. The method of claim 1, comprising:

determining a respective probability for respective words corresponding to a unique identifier within the identified first range, the probability relating to a likelihood that the word matches a word a user is intending to input.

4. The method of claim 1, respective unique identifiers being an integer.

5. The method of claim 4, the first range being contiguous.

6. The method of claim 5, respective identifiers respectively corresponding to words in a lexicon of words.

7. The method of claim 1, comprising:

receiving a second character set comprising one or more characters;

identifying a second range of one or more unique identifiers respectively corresponding to a word having a prefix that matches the second character set; and

identifying one or more phrases comprising two or more words using the first range of one or more unique identifiers and the second range of one or more unique identifiers.

8. The method of claim 7, comprising:

determining a probability for respective identified phrases, the probability relating to a likelihood that the identified phrase matches a phrase a user is intending to input.

9. The method of claim 1, comprising:

receiving one or more additional characters that increases the size of the first character set; and

narrowing the first range of one or more unique identifiers in response to the received one or more additional characters.

10. A system, comprising:

a lookup component configured to identify a range of one or more unique identifiers respectively corresponding to a word comprising a prefix that matches a received character set comprising one or more characters.

11. The system of claim 10, no two words corresponding with a same unique identifier.

12. The system of claim 10, respective unique identifiers being integers.

13. The system of claim 12, the range of unique identifiers being contiguous.

14. The system of claim 13, comprising a database configured to store a lexicon, the lexicon comprising a plurality of words and a plurality of unique identifiers, the words arranged alphabetically and the unique identifiers assigned to the arranged words at least one of sequentially and contiguously.

15. The system of claim 10, comprising a database configured to store probability information for respective unique identifiers, the probability information relating to a likelihood that a word associated with the unique identifier is a word that matches a word a user is intending to input.

16. The system of claim 10, the lookup component configured to identify one or more phrases based at least in part upon two unique identifiers, a first unique identifier associated with a prefix of a first word and a second unique identifier associated with a prefix of a second word.

17. The system of claim 10, the lookup component configured to narrow the range of one or more unique identifiers in response to an additional character in the character set.

18. A method comprising:

receiving a character set comprising at least one completed word and a prefix of a second word;

identifying a unique identifier corresponding to the completed word;

identifying a range of one or more unique identifiers respectively corresponding to a word having a prefix that matches the prefix of the second word; and

using the unique identifier corresponding to the completed word and the range of one or more unique identifiers to identify one or more phrases that a user may be attempting to input.

19. The method of claim 18, comprising determining a respective probability for respective phrases, where the probability relates to a likelihood that the phrase matches a phrase a user is attempting to input.

20. The method of claim 18, comprising:

arranging a plurality of words alphabetically; and

assigning respective words a unique identifier, the unique identifiers being integers, the unique identifiers assigned sequentially and contiguously to the arranged words.