System and method for detecting and correcting mismatched Chinese character

• US 8,725,497 B2
• Filed: 10/05/2011
• Issued: 05/13/2014
• Est. Priority Date: 10/05/2011
• Status: Active Grant

• Current Assignee: Daniel M. Wang
• Original Assignee: Daniel M. Wang
• Inventors: Wang, Daniel M.
• Primary Examiner(s): GUERRA-ERAZO, EDGAR X
• Application Number:
  US13/253,662
• Publication Number:
  US 20130090916A1
• Time in Patent Office: 951 Days
• Field of Search: 704/3, 704/4, 704/9, 704/10, 704/251, 704/270, 704/275
• US Class Current: 704/9
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation

Abstract

A system and method for detecting and correcting mismatched Chinese characters in a phrase. The system comprises a database for the look-up of characters and Chinese phrases, a module to compare the input phrases with the look-up data retrieved from the database and a module to correct the mismatched characters. The database contains correct phrases as well as attributes associated with each character, such as pronunciation and radical composition. The modules inputs a Chinese phrase that has at least two characters and compares it with the data retrieved from the database to determine if there are incorrect characters. The spell checking method includes two groups of steps: mismatched character detection and mismatched character correction. Whether there is any mismatched character to be corrected is determined by the edit distance, the phrase length and comparisons of the pronunciation and radical composition of the mismatched characters.

Claims

1. A non-transitory computer-readable medium on which is stored a computer program for detecting and correcting mismatched Chinese characters in a phrase, the computer program comprising computer instructions that when executed by a computing device performs the steps of:

(S1) inputting an input phrase;

(S2) a mismatched character detection module comparing the input phrase with Chinese phrases stored in a database;

(S3) in step (S2), if there is an exact match, the mismatched character detection module determining the input phrase as constructed correctly;

(S4) in step (S2), if there is no exact match and an edit distance is greater than two, the mismatched character detection module determining the input phrase as an unknown phrase;

(S5) in step (S2), if there is no exact match, the edit distance is two and the phrase length is two, the mismatched character detection module determining the input phrase as an unknown phrase;

(S6) in step (S2), if there is no exact match, the edit distance is one, or the edit distance is two and the phrase length is greater than two, the mismatched character detection module looking up a database to retrieve the associated pronunciation and radical composition for each of the mismatched Chinese characters in the input phrase and sending the retrieved information to an mismatched character correction module;

(S7) the mismatched character correction module performing an adjacency match by swapping two adjacent and mismatched characters in the input phrase and see if the input phrase would match any of the partially matched phrases, if a match being found, the mismatched character correction module adding the partially matched phrase to a completely matched phrase list;

(S8) the mismatched character correction module comparing the pronunciation of the mismatched character(s) with that of the character(s) in the same position of the partially matched phrase, if the pronunciation is identical, the mismatched character correction module adding the partially matched phrase to the completely matched phrase list;

(S9) the mismatched character correction module comparing the pronunciation of the mismatched character(s) with that of the character(s) in the same position of the partially matched phrase, if the pronunciation is similar, the mismatched character correction module adding the partially matched phrase to the completely matched phrase list;

(S10) the mismatched character correction module compare the radical composition of the mismatched character(s) with that of the character(s) in the same position of the partially matched phrase, if there is a radical match, the mismatched character correction module adding the partially matched phrase to the completely matched phrase list.

2. The non-transitory computer-readable medium as claimed in claim 1, wherein the edit distance is used as a measurement by the mismatched character detection module to determine if there are mismatched characters.

3. The non-transitory computer-readable medium as claimed in claim 1, wherein if the edit distance is equal to one or two but the phrase length is greater than two, the mismatched character detection module performs a database look up to retrieve the pronunciation and radical composition for the characters mismatched.

4. The non-transitory computer-readable medium as claimed in claim 1, wherein the mismatched character detection module sends to the mismatched character correction module the mismatched characters, pronunciations and radical composition of the characters, and a list of partially matched phrases that have the edit distance equal to or less than two for correcting mismatched characters.

5. The non-transitory computer-readable medium as claimed in claim 1, wherein the mismatched characters can be corrected if there is only one phrase in the completely matched phrase list.

6. The non-transitory computer-readable medium as claimed in claim 1, wherein the mismatched characters can be corrected if there are multiple phrases in the completely matched phrase list and a best matched phrase can be found through the following steps performed in descending order:

if only one of the completely matched phrases contains the adjacency match, the best-matched phrase is found;

if only one of the completely matched phrases has an identical pronunciation match, the best-matched phrase is found;

if only one of the completely matched phrases has both similar pronunciation and radical matches, the best-matched phrase is found;

if only one of the completely matched phrases has a similar pronunciation match, the best-matched phrase is found; and

if only one of the completely matched phrases has a radical match, the best-matched phrase is found.