CN101287228A - Phoneticizing error correcting technique and device applying to query by short message service of mobile phone - Google Patents
Phoneticizing error correcting technique and device applying to query by short message service of mobile phone Download PDFInfo
- Publication number
- CN101287228A CN101287228A CNA2008101126069A CN200810112606A CN101287228A CN 101287228 A CN101287228 A CN 101287228A CN A2008101126069 A CNA2008101126069 A CN A2008101126069A CN 200810112606 A CN200810112606 A CN 200810112606A CN 101287228 A CN101287228 A CN 101287228A
- Authority
- CN
- China
- Prior art keywords
- entity
- mobile phone
- query
- technology
- error correction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention provides a spelling correction technology for mobile phone message query and a device thereof. The steps of the spelling correction technology are as follows: step 1: the homophony and approximant entities of an entity are searched and the homophony or approximant wrongly written or mispronounced characters are matched and corrected; step 2: the fuzzy syllable entity of the entity is searched and the fuzzy syllable is matched and corrected; step 3: the expansion terms in similar forms to the entity are searched and sigla are matched and corrected. The technology and the device of the invention have the advantages that: according to the input characteristics of a mobile phone, the spelling correction technology and the abbreviation identification technology are utilized to identify geographical names, thus solving the problem that a client does not know the detailed spelling of geographical names or the geographical names include uncommon characters, and the client needs not to remember the complete geographical names, which is more convenient and more suitable for the habits of the client. Through the technology and the device of the invention, the query sentences that are keyed in by clients with natural languages become words that can be understood by the system, and the geographical entity words mastered by the system constitute a joint query, which is convenient for the further treatment of the later geographical navigation system.
Description
Technical field
The invention belongs to mobile phone expanded function technical field, particularly a kind of phonetic error correcting technique and device that is applied to mobile phone short message enquiry.
Background technology
The application service on note at present can only be a simple customize services etc. owing to do not possess natural language processing technique, and the user need learn to send customizing messages such as code and just can simply use.And for the search and the application of this user's request complexity in path, this mode can not meet the demands far away, also can cause bad user experience because of complex operating steps.
Cellphone subscriber's characteristics are the error rate problem of higher of the input that brings of cellphone inputting method, the input method major part is the phonetic input on the mobile phone at present, and in the input method that is nothing like aspect phrase quantity, the ease for use on the computer, this just cause a lot of users in input because operate miss or for fast, use wrong word or speech like the sound.As often can find of this sort note " to the Dongzhimen how to get to not? " up to (knowing)In the application of local search and guidance, it is more that this situation occurs, because most place name, road name be not in the dictionary of input method, the user often replaces like phrase with sound commonly used for convenience, under many circumstances, and user even also do not know the correct literary style of certain place name, road name, just know pronunciation, and in the road name, place name more rarely used word being arranged, the user does not know how to import, and can only replace like word or likeness in form word with sound.As " HaiLong Building ", the user may be entered as " oceanic rise mansion "." Wuyuan " user may be entered as " having no chance " or " Chinese blister beetle source ".On system level, traditional message search system adopts based on keyword or based on the search technique of instructing, and brought very big inconvenience to the user, and said process has well solved this problem on application.
Summary of the invention
The objective of the invention is to, the text that the cellphone subscriber is imported by natural language text carries out the phonetic error correction.
To achieve these goals, the invention provides a kind of phonetic error correcting technique that is applied to mobile phone short message enquiry, comprise: step 1: search the unisonance of entity, nearly sound entity, unisonance or nearly sound wrong word are mated error correction; Step 2: search the fuzzy sound entity of entity, fuzzy sound is mated error correction; Step 3: search the likeness in form expansion word of entity, abbreviation is mated error correction.
The present invention also provides a kind of phonetic error correction device that is applied to mobile phone short message enquiry, comprises: search the unisonance of entity, nearly sound entity, unisonance or nearly sound wrong word are mated the unisonance correction module of error correction; Search the fuzzy sound entity of entity, fuzzy sound is mated the fuzzy sound correction module of error correction; Search the likeness in form expansion word of entity, abbreviation is mated the abbreviation correction module of error correction.
The beneficial effect of technical scheme provided by the invention is: at the characteristics of mobile phone input, the technology of phonetic error correction and the technology of abbreviation identification are used in the identification of place name, solved the problem of user when not knowing that concrete literary style of place name or place name comprise rarely used word, and make things convenient for the user can remember the complete name of place name, more meet user's custom.By the present invention, the query statement of user's natural language input, it is accessible to have become system, by the conjunctive query that the geographical entity speech that system grasped constitutes, is convenient to the further processing of the Geographic Navigation system of back.
Description of drawings
Fig. 1 is a natural language processing technique flow chart of the present invention;
Fig. 2 is a phonetic error correcting technique flow chart of the present invention;
Fig. 3 is a phonetic error correction device structural representation of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, embodiment of the present invention is described further in detail below in conjunction with accompanying drawing.
The invention provides a kind of phonetic error correcting technique that is applied to mobile phone short message enquiry, be based on another mobile phone short message inquiry error correction method of natural language processing technique.Fig. 1 is a natural language processing technique flow chart of the present invention.At first, explain described natural language processing technique, its processing procedure is: the cellphone subscriber imports natural language text query statement (step S101), as " from the airport to the oceanic rise mansion how to get to? " word-dividing mode is handled, by the everyday words dictionary natural language text is divided into everyday words (step S102), this sentence be split into " from/airport/to how/sea/grand/mansion// walk/?/".Then, text is sent to the part-of-speech tagging module, this module is by part of speech dictionary and feature lexicon, everyday words is marked part of speech and feature (step S103), be noted as " airport " " general place name ", " to " be noted as verb, by such step, we are appreciated that the structure of sentence, as SVO etc.; Utilize the auxiliary classification of syntactic feature and everyday words feature to query statement.Again by question sentence field identification module, by domain features dictionary and field way to put questions feature lexicon, the natural language text that will belong to " transport information " is distributed to the Entity recognition module, this step need be in conjunction with the interrogative feature, as whether comprising " where ", " how ", the verb feature, as " walking " " to " " going ", and everyday words domain features, as " general place name ", " name commonly used " etc., understand the simple semanteme in the query statement, according to semantic feature to text classify (step S104).The text that will belong to " transport information " sends to the Entity recognition module, by field related entities dictionary, identifies possible domain entities (step S105)." airport " in the problems referred to above, " oceanic rise mansion " are identified.Afterwards, in the entity matching module, carry out the entity coupling, utilize POI entity dictionary, identify correct entity speech and may be the speech string (step S106) of entity, through this step, all place names and possible place name entity all are identified, as " Wangfujing ", " East 4th Ring Road ".
Fig. 2 is a phonetic error correcting technique flow chart of the present invention.Then, to may being that the speech string of entity carries out the phonetic error correction.Because cellphone inputting method usually is simple spelling input method, unisonance or nearly sound wrong word appear easily, as " Zhong Guan village ", " oceanic rise mansion " etc., the unisonance entity that we utilize the unisonance correction module to search the possibility entity carries out error correction (step S201).This step is output as the entity matching result through error correction, and above-mentioned " oceanic rise mansion " is converted into " HaiLong Building ".Consider various places accent characteristics simultaneously, by fuzzy sound correction module, added error correction, again as " f " and (step S202) such as " h " based on fuzzy sound.Again, add likeness in form abbreviation entity matching result by the abbreviation correction module, the speech that is about to the abbreviation likeness in form is matched to correct entity speech (step S203).Above-mentioned " airport " is mapped to " Capital Airport ".At last, all coupling entities are output.
As shown in Figure 3, the present invention also provides a kind of phonetic error correction device that is applied to mobile phone short message enquiry, comprises: unisonance correction module 1, and to search and may mate for the unisonance entity of entity, output is through the correct entity of coupling; Fuzzy sound correction module 2 is searched and may be mated for the fuzzy sound entity of entity, and output is through the correct entity of coupling; Abbreviation correction module 3, the speech that abbreviation is similar to is matched to correct entity speech.
Being representative instance of the present invention only below, is not to be used for limiting practical range of the present invention.Be that all equalizations of being done according to the present patent application claim change and modification, be all claim of the present invention and cover.
Claims (2)
1, a kind of phonetic error correcting technique that is applied to mobile phone short message enquiry is characterized in that, comprises:
Step 1: search the unisonance of entity, nearly sound entity, unisonance or nearly sound wrong word are mated error correction;
Step 2: search the fuzzy sound entity of entity, fuzzy sound is mated error correction;
Step 3: search the likeness in form expansion word of entity, abbreviation is mated error correction.
2, a kind of phonetic error correction device that is applied to mobile phone short message enquiry is characterized in that, comprises:
Search the unisonance of entity, nearly sound entity, unisonance or nearly sound wrong word are mated the unisonance correction module of error correction;
Search the fuzzy sound entity of entity, fuzzy sound is mated the fuzzy sound correction module of error correction;
Search the likeness in form expansion word of entity, abbreviation is mated the abbreviation correction module of error correction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101126069A CN101287228A (en) | 2008-05-26 | 2008-05-26 | Phoneticizing error correcting technique and device applying to query by short message service of mobile phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008101126069A CN101287228A (en) | 2008-05-26 | 2008-05-26 | Phoneticizing error correcting technique and device applying to query by short message service of mobile phone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101287228A true CN101287228A (en) | 2008-10-15 |
Family
ID=40059146
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008101126069A Pending CN101287228A (en) | 2008-05-26 | 2008-05-26 | Phoneticizing error correcting technique and device applying to query by short message service of mobile phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101287228A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750351A (en) * | 2012-06-11 | 2012-10-24 | 迪尔码国际营销服务(北京)有限公司 | Matching method of address information based on rules |
| CN103914455A (en) * | 2012-12-30 | 2014-07-09 | 高德软件有限公司 | Method and device for retrieving interest points |
| CN105760359A (en) * | 2014-11-21 | 2016-07-13 | 财团法人工业技术研究院 | Question processing system and method thereof |
| CN110457695A (en) * | 2019-07-30 | 2019-11-15 | 海南省火蓝数据有限公司 | A kind of online text error correction method and system |
-
2008
- 2008-05-26 CN CNA2008101126069A patent/CN101287228A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750351A (en) * | 2012-06-11 | 2012-10-24 | 迪尔码国际营销服务(北京)有限公司 | Matching method of address information based on rules |
| CN103914455A (en) * | 2012-12-30 | 2014-07-09 | 高德软件有限公司 | Method and device for retrieving interest points |
| CN103914455B (en) * | 2012-12-30 | 2017-10-24 | 高德软件有限公司 | A kind of interest point search method and device |
| CN105760359A (en) * | 2014-11-21 | 2016-07-13 | 财团法人工业技术研究院 | Question processing system and method thereof |
| CN110457695A (en) * | 2019-07-30 | 2019-11-15 | 海南省火蓝数据有限公司 | A kind of online text error correction method and system |
| CN110457695B (en) * | 2019-07-30 | 2023-05-12 | 安徽火蓝数据有限公司 | Online text error correction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101287229A (en) | Natural language processing technique and device applying to query by short message service of mobile phone | |
| US10073843B1 (en) | Method and apparatus for cross-lingual communication | |
| US11817101B2 (en) | Speech recognition using phoneme matching | |
| US8290775B2 (en) | Pronunciation correction of text-to-speech systems between different spoken languages | |
| US20220092278A1 (en) | Lexicon development via shared translation database | |
| EP3032532B1 (en) | Disambiguating heteronyms in speech synthesis | |
| US20190087455A1 (en) | System and method for natural language processing | |
| CN102084417B (en) | System and methods for maintaining speech-to-speech translation in the field | |
| US8365070B2 (en) | Spelling correction system and method for misspelled input | |
| De Melo | Lexvo. org: Language-related information for the linguistic linked data cloud | |
| US7742922B2 (en) | Speech interface for search engines | |
| US9058322B2 (en) | Apparatus and method for providing two-way automatic interpretation and translation service | |
| US20090326945A1 (en) | Methods, apparatuses, and computer program products for providing a mixed language entry speech dictation system | |
| US20080215519A1 (en) | Method and data processing system for the controlled query of structured saved information | |
| US20150081294A1 (en) | Speech recognition for user specific language | |
| WO2006106415A1 (en) | Method, device, and computer program product for multi-lingual speech recognition | |
| JP4740837B2 (en) | Statistical language modeling method, system and recording medium for speech recognition | |
| KR20070058953A (en) | Method and apparatus for generating a response sentence in dialogue system | |
| CN110942767B (en) | Recognition labeling and optimization method and device for ASR language model | |
| CN101287228A (en) | Phoneticizing error correcting technique and device applying to query by short message service of mobile phone | |
| Yang et al. | Vocabulary expansion through automatic abbreviation generation for Chinese voice search | |
| US8401855B2 (en) | System and method for generating data for complex statistical modeling for use in dialog systems | |
| US20200372110A1 (en) | Method of creating a demographic based personalized pronunciation dictionary | |
| Liu et al. | CityBrowser II: A multimodal restaurant guide in Mandarin | |
| JP2010257085A (en) | Retrieval device, retrieval method, and retrieval program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| DD01 | Delivery of document by public notice |
Addressee: Roadinfo Systems Co., Ltd. Document name: Notification of before Expiration of Request of Examination as to Substance |
|
| DD01 | Delivery of document by public notice |
Addressee: Wang Weifeng Document name: Notification that Application Deemed to be Withdrawn |
|
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20081015 |