CN103514217B - Method and system for processing associated prompts of retrieval condition of retrieval application - Google Patents
Method and system for processing associated prompts of retrieval condition of retrieval application Download PDFInfo
- Publication number
- CN103514217B CN103514217B CN201210221725.4A CN201210221725A CN103514217B CN 103514217 B CN103514217 B CN 103514217B CN 201210221725 A CN201210221725 A CN 201210221725A CN 103514217 B CN103514217 B CN 103514217B
- Authority
- CN
- China
- Prior art keywords
- word
- word portion
- index
- cue
- column
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method and system for processing associated prompts of a retrieval condition of a retrieval application. According to the method and system for processing the associated prompts of the retrieval condition of the retrieval application, due to the fact that a word portion database corresponding to a prompting word library and an index of the word portion database are additionally built, a computer system can find that a target word portion has an index basis from the word portion database in an index query mode, involved matching comparison in the whole prompt processing process can be accomplished through retrieval searching processing, the searching processing mode that comparison matching searching is carried out on prompting word storage units one by one in the prompting word library is omitted, even when the data size of prompting words contained in the prompting word library is huge, the computer system can provide the associated prompting words of the retrieval condition for users in a prompting mode, the associated prompting words are obtained in the index searching processing mode through the few times of comparison, efficiency of the computer system for processing the associated prompts of the retrieval condition of the retrieval application is improved in essence, the response speed of the computer system is increased, and better use experience is provided for the users.
Description
Technical field
The present invention relates to computer communication network technology and retrieval applied technical field are and in particular to a kind of retrieve in application
The associated cue processing method of search condition and system.
Background technology
Along with the continuous development and progress in computerized information epoch, no matter search engine is to apply neck in Internal retrieval
Domain is still obtained in the retrieval application of enterprises and is widely applied, and search engine is except having according to user input
Search condition return associated result basic function outside, some intelligent functions of search engine are also constantly gushing
Existing, it is exactly for example one of search engine main intelligent functions in retrieval application to the associated cue function of search condition.
Associated cue function to search condition in retrieval application, its function is, in user input search condition, if
Criteria character string as search condition matches the partial character of in retrieval application standby cue, then will match
Each cue is prompted to user as associated cue word, is selected as new search condition for user；Since so, for example, use
User, during the expected place name of input, a phrase or an English word etc. are as search condition, is worked as in family
During partial character in place name, phrase or English word expected from input, system just can automatically can by this partial character
Can the place name of coupling, phrase or English word associated cue to user, thus allowing user with regard to energy before not yet fully inputting
Enough choose expected search condition faster, carry out needing search operaqtion.Certainly, the realization of this function needs to pre-build one
Individual prompting dictionary, for standby cue in memory scan application.
Index technology has become as one of retrieval application major technique, and computer communication network technology and retrieval should
Both known about with those skilled in the art, if the number that a data storage cell is recorded will be searched from a data base
According to compared to for each data storage cell is compared lookup one by one in data base, data being obtained by index search
Comparison number of times needed for retrieval result will be far fewer than the former, and therefore index search can obtain data inspection faster from data base
Hitch fruit, generally all can design in therefore conventional in the prior art computer system and automatically set up rope for existing data volume
Draw, obtain the response speed of data research result with lift system from data with existing storehouse.For example to retrieval in retrieval application
In the associated cue function of condition, due to needing to pre-build a prompting dictionary, therefore also can according to conventional system design
Automatically set up the index of prompting dictionary, search, from prompting dictionary, the processing speed extracting cue to be lifted；However, it is conventional
System design in do not set up the related other index functions of prompting dictionary, when the search condition of user input is cue
In storehouse during the partial character of standby cue, to find, from prompting dictionary, the cue matching with described partial character
As associated cue word, it is typically only capable to for described partial character to add formation fuzzy matching entry after asterisk wildcard, by this fuzzy
Join entry and compared matched and searched one by one with each cue memory element in prompting dictionary, the prompting comprising in prompting dictionary
In the case that word data volume is huge, is determined by way of so comparing matched and searched one by one and to match with described partial character
Cue, it needs the comparison number of times of experience many, takes and also grows although successfully determining the cue matching very much
Afterwards, the extraction process of cue is rapidly completed by index search, but user is obtained the entirety of associated cue result
Response speed is then significantly affected because of the time-consuming of matched and searched process, thus greatly reduce user closing to search condition
The experience sense of connection prompt facility.
As can be seen here, how to lift the response speed of search condition associated cue function further, have become as lifting inspection
One importance of rope application service quality, however poorer to this correlational study and report in prior art.
Content of the invention
For deficiency above-mentioned present in prior art, it is an object of the invention to provide retrieving bar in a kind of retrieval application
The associated cue processing method of part, with lifted computer system to retrieval application in search condition associated cue treatment effeciency and
Response speed, brings more preferable experience to user.
For achieving the above object, present invention employs following technological means：
In a kind of retrieval application, the associated cue processing method of search condition is it is characterised in that pre-build prompting dictionary
Corresponding word portion data base, obtaining number of characters in prompting dictionary by word portion deconsolidation process is that plural cue is corresponding
Each word portion is simultaneously recorded in institute predicate portion data base, sets up the index of institute predicate portion data base；Each word portion is right by it
The character string that in the cue answered, partial character is constituted；In user input search condition, obtain as search condition
Criteria character string and calculate its index value, by each of index search of word portion data base to word portion data base with described
Criteria character string has the word portion of same index value respectively as conjunctive word portion, and each corresponding cue in conjunctive word portion is divided
It is not prompted to user as associated cue word.
In the associated cue processing method of search condition in above-mentioned retrieval application, as a kind of preferred version, described " logical
Crossing word portion deconsolidation process and obtaining number of characters in prompting dictionary is plural cue each word portion corresponding " among, to carrying
Show in dictionary that any one number of characters carries out word portion deconsolidation process for plural cue and is specially：Obtain the word of this cue
Symbol number Nx, then Nx >=2, make the value of default character properties n be respectively the integer from 1 to Nx-1；Every for character properties n
One value, takes the character string that front n character in this cue is constituted as the corresponding word portion of this cue；By
This obtains the corresponding Nx-1 word portion of this cue, and the word portion deconsolidation process of this cue is completed.
In the associated cue processing method of search condition in above-mentioned retrieval application, as a kind of preferred version, described " build
The index of vertical institute predicate portion data base " is specially：Each word portion in word portion data base with identical characters number is lined up as one
Individual word portion index column, the number of characters in word portion single in word portion index column is identified ID as the row of this word portion index column, and calculates
The index value in each word portion in word portion index column, sets up the corresponding column index of word portion index column；Thus, by word portion data base
All word portions line up as multiple word portions index column, and set up and obtain each corresponding column index of word portion index column, each word portion rope
Draw the set i.e. index as word portion data base arranging corresponding column index.
In the associated cue processing method of search condition in above-mentioned retrieval application, as a kind of preferred version, in user
During input search condition, described " obtain as the criteria character string of search condition and calculate its index value, by word portion data base
Index search there is the word portion of same index value respectively as pass to each of word portion data base with described criteria character string
Connection word portion, the cue corresponding to using each conjunctive word portion is prompted to user as associated cue word " concrete process
Comprise the steps：
c1）Obtain the criteria character string as search condition, and calculate the index value of described criteria character string；
c2）Judge in the row mark ID of each word portion index column with the presence or absence of equal with the number of characters of described criteria character string
Row mark ID；If existing, execution step c3；If not existing, judge currently not obtaining associated cue word；
c3）Identify a word portion index column of ID using having the row equal with the number of characters of described criteria character string as pass
Connection object indexing row, are searched by the corresponding column index of affiliated partner index column and judge whether deposit in described affiliated partner index column
In the word portion with described criteria character string with same index value；If existing, execution step c4；If not existing, judge to work as
Front do not obtain associated cue word；
c4）Each of described affiliated partner index column and described criteria character string had the word part of same index value
Not as conjunctive word portion, and then the cue corresponding to using each conjunctive word portion is prompted to use as associated cue word
Family.
In the associated cue processing method of search condition in above-mentioned retrieval application, as a kind of preferred version, institute's predicate
In the index of portion data base, the index value in each word portion is the cryptographic Hash in corresponding words portion；The index value of described criteria character string is should
The cryptographic Hash of criteria character string.
Correspondingly, present invention also offers a kind of associated cue being capable of search condition in above-mentioned retrieval application is processed
The associated cue processing system of method；For this reason, present invention employs following technological means：
The associated cue processing system of search condition in a kind of retrieval application, this system includes pointing out dictionary, word portion data
Storehouse is set up module, word portion processing module, is indexed and set up module and associated cue processing module；Described prompting dictionary carries for record
Show word；Institute's predicate portion Database module is used for setting up prompting dictionary corresponding word portion data base；Institute's predicate portion processing module
It is plural cue each word portion corresponding and remembers for number of characters in prompting dictionary is obtained by word portion deconsolidation process
Record in institute predicate portion data base；Each word portion is a character being made up of partial character in its corresponding cue
String；Described index sets up module for setting up the index of institute predicate portion data base；Described associated cue processing module be used for
During the input search condition of family, obtain the criteria character string as search condition and calculate its index value, by word portion data base's
Index search has the word portion of same index value respectively as associating to each of word portion data base and described criteria character string
Word portion, the cue corresponding to using each conjunctive word portion is prompted to user as associated cue word.
In the associated cue processing system of search condition in above-mentioned retrieval application, as a kind of preferred version, institute's predicate
Portion's processing module includes word portion deconsolidation process unit and word portion record processing unit；Institute's predicate portion deconsolidation process unit is used for passing through
It is plural cue each word portion corresponding that word portion deconsolidation process obtains number of characters in prompting dictionary；Wherein, to prompting
In dictionary, any one number of characters carries out word portion deconsolidation process for plural cue and is specially：Obtain the character of this cue
Number Nx, then Nx >=2, make the value of default character properties n be respectively the integer from 1 to Nx-1；Each for character properties n
Individual value, takes the character string that front n character in this cue is constituted as the corresponding word portion of this cue；Thus
Obtain the corresponding Nx-1 word portion of this cue, the word portion deconsolidation process of this cue is completed；Predicate portion of institute record processes single
Unit will be prompted to number of characters in dictionary and is recorded in institute predicate portion data base for plural cue each word portion corresponding.
In the associated cue processing system of search condition in above-mentioned retrieval application, as a kind of preferred version, index is built
Formwork erection block include word portion line up processing unit, row mark processing unit and column index set up unit；Predicate portion of institute lines up to process list
Unit is for lining up each word portion in word portion data base with identical characters number for a word portion index column；Thus, by word portion
All word portions in data base line up as multiple word portions index column；Described row mark processing unit is used for respectively by each word portion
In index column, the number of characters in single word portion identifies ID as the row of corresponding words portion index column；Described column index set up unit by based on
Calculate the index value in each word portion in each word portion index column, set up each corresponding column index of word portion index column respectively.
In the associated cue processing system of search condition in above-mentioned retrieval application, as a kind of preferred version, described pass
Connection prompting processing includes search condition index value computing unit, affiliated partner index column judging unit, the judgement of conjunctive word portion
Unit and associated cue word Tip element；Described search condition index value computing unit is used for obtaining the condition as search condition
Character string, and calculate the index value of described criteria character string；Described affiliated partner index column judging unit is used for judging each word
The row of portion's index column identify in ID and identify ID with the presence or absence of the row equal with the number of characters of described criteria character string；If existing,
Triggering runs conjunctive word portion judging unit；If not existing, judge currently not obtaining associated cue word；Described conjunctive word portion judges
Unit is used for identifying a word portion index column of ID as association using having the row equal with the number of characters of described criteria character string
Object indexing arranges, and is searched by the corresponding column index of affiliated partner index column and judges to whether there is in described affiliated partner index column
There is the word portion of same index value with described criteria character string；If existing, triggering runs associated cue word Tip element；If no
Exist, then judge currently not obtaining associated cue word；Described associated cue word Tip element is used for indexing described affiliated partner
Each of row have the word portion of same index value respectively as conjunctive word portion with described criteria character string, and then each is closed
Cue corresponding to connection word portion is prompted to user respectively as associated cue word.
Compared to prior art, the present invention has the advantages that：
1st, the associated cue processing method of search condition and system in present invention retrieval application, set up cue by increasing
The index of storehouse corresponding word portion data base and word portion data base so that computer system by way of search index from word portion
Inquire target word portion in data base and possess index basis, ensure that computer system inquires from word portion data base first
The treatment effeciency in target word portion and response speed；And the word portion recording in word portion data base, exactly right in prompting dictionary by it
The partial character answering cue is constituted, therefore, if the criteria character string as search condition of user input and word portion number
Consistent according to some or certain several word portions in storehouse, computer system just can pass through the index search of word portion data base rapidly
Coupling obtains consistent with described criteria character string some or certain several word portions as conjunctive word portion, further according to each pass
The connection word portion associated cue word corresponding with the quick acquisition of corresponding relation of cue in prompting dictionary is prompted to user, thus puts
Take off the lookup processing mode that each cue memory element in prompting dictionary is compared one by one with matched and searched；Even if prompting
The cue data volume that dictionary comprises is huge, and computer system also can be by index search processing mode after few in number
The associated cue word contrasting for several times and obtaining search condition is prompted to user, thus improving computer system in itself to inspection
The associated cue treatment effeciency of search condition and response speed in Suo Yingyong.
2nd, in the associated cue processing method and system of search condition in present invention retrieval application, additionally provide a kind of excellent
The word portion deconsolidation process scheme of choosing, carries out word portion deconsolidation process by this word portion deconsolidation process scheme to cue, by number of characters
Cue for Nx can accordingly split Nx-1 word portion of acquisition, and this Nx-1 word portion covers n before the Nx-1 kind of cue
The sequence input condition of individual character is so that the word portion that word portion data place comprises contains user's inspection the most universal as much as possible
The various situations of rope condition entry custom cover accuracy rate it is ensured that word portion data base possesses higher association, contribute to into one
Step improves the utilization rate of word portion data base；And, this preferred word portion deconsolidation process scheme has relatively regular deconsolidation process rule
Rule, easily can realize computer system automatization word portion deconsolidation process by computer programming.
3rd, in the associated cue processing method and system of search condition in present invention retrieval application, additionally provide a kind of excellent
The word portion database index of choosing sets up scheme, and also with have impact on the concrete mistake that the later stage processes to conjunctive word portion index search
Journey；Compared with whole word portion data base is directly set up with overall index, this preferred word portion database index sets up scheme by word
All word portions in portion data base line up as multiple word portions index column, and set up and obtain each corresponding row rope of word portion index column
Draw；In the index search processing procedure to conjunctive word portion for the later stage, the affiliated partner index column determining after being positioned by row is only
One of several word portion index column that word portion data base comprises, the word portion quantity that this affiliated partner index column is comprised must
So comprise word portion quantity less than whole word portion data place, therefore with that whole word portion data base is directly set up with overall index is laggard
Line index inquiry is compared, and this corresponding column index of affiliated partner index column is indexed inquire about experienced contrast number of times necessarily
Can be less, thus contributing to further being lifted the Whole Response speed of associated cue.
Brief description
Fig. 1 is the FB(flow block) of the associated cue processing method of search condition in the embodiment of the present invention one；
Fig. 2 is to any one number of characters in prompting dictionary in the embodiment of the present invention one using preferred term portion deconsolidation process scheme
Carry out the FB(flow block) of word portion deconsolidation process for plural cue；
Fig. 3 is the FB(flow block) that in the embodiment of the present invention one, preferred term portion database index sets up scheme；
Fig. 4 be the embodiment of the present invention one in, in the case that Fig. 3 institute predicate portion database index sets up scheme, for
The search condition of family input executes the FB(flow block) that associated cue processes operation；
Fig. 5 is the system framework figure of the associated cue processing system of search condition in present invention retrieval application.
Specific embodiment
In the prior art, why computer system is to the entirety retrieving the associated cue function of search condition in application
Response speed is slower, main reason is that, computer system is contrasting the association of matched and searched search condition from prompting dictionary
During cue, do not have to break away to prompting dictionary in each cue memory element compared matched and searched one by one
Lookup processing mode so as to contrast matching times drastically increase in the case that the prompting cue data volume that comprises of dictionary is huge
Plus, treatment effeciency is low, and time-consuming, has had a strong impact on computer system and the associated cue of search condition in retrieval application is integrally rung
Answer speed.For this problem, the present invention proposes a kind of associated cue processing method of search condition in retrieval application, and existing
Search condition associated cue function treatment mode in technology is different, and the inventive method pre-builds prompting dictionary corresponding word portion
Data base, by number of characters in word portion deconsolidation process acquisition prompting dictionary for plural cue each word portion corresponding simultaneously
It is recorded in institute predicate portion data base, set up the index of institute predicate portion data base；Each word portion is by its corresponding cue
The character string that middle partial character is constituted；In user input search condition, obtain the criteria character as search condition
Go here and there and calculate its index value, by each of index search of word portion data base to word portion data base and described criteria character string
The word portion with same index value respectively as conjunctive word portion, using each corresponding cue in conjunctive word portion as association
Cue is prompted to user.Can see, in the associated cue processing method of search condition in present invention retrieval application, by increasing
Plus set up the index of prompting dictionary corresponding word portion data base and word portion data base so that computer system passes through search index
Mode inquire from word portion data base target word portion possess index basis, ensure that computer system from word portion number first
According to the treatment effeciency inquiring target word portion in storehouse and response speed；And in word portion data base record word portion, exactly by its
In prompting dictionary, the partial character of corresponding cue is constituted, therefore, if the condition word as search condition of user input
Symbol string is consistent with some or certain several word portions in word portion data base, and computer system just can be by the rope of word portion data base
Draw and search rapidly coupling and obtain consistent with described criteria character string some or certain several word portions as conjunctive word portion, then
According to the associated cue word prompting corresponding with the quick acquisition of corresponding relation of cue in prompting dictionary of each conjunctive word portion
To user, thus break away from the lookup that matched and searched is compared one by one to each cue memory element in prompting dictionary and processed
Mode；Even if the cue data volume that prompting dictionary comprises is huge, computer system also can pass through index search processing mode
It is prompted to user after few in number associated cue words contrasting for several times and obtaining search condition, thus improving in itself
Computer system is to the associated cue treatment effeciency of search condition and response speed in retrieval application.In present invention retrieval application
In the associated cue processing method of search condition, why carry out for plural cue for number of characters in prompting dictionary
Word portion deconsolidation process, reason is, for the associated cue function of search condition in retrieval application, an only character
Cue does not need to be associated pointing out again, and therefore there is no need it is processed with acquisition word portion, thus contributing to reducing word portion number
According to unnecessary data record in storehouse, reduce the redundancy of word portion data base, improve the utilization rate of word portion data base.Certainly, do not having
In the case of having specified otherwise, those skilled in the art understand, in step before and after same technical method, each computation index
The index value-based algorithm that value is adopted is same index value-based algorithm；The associated cue processing method of the present invention is also such.
In the associated cue processing method of search condition in present invention retrieval application, many factors are to associated cue
Association covers accuracy rate and associated cue response speed also has close ties.On the one hand, carry out word portion to cue to tear open
Office reason obtains the concrete fractionation mode in word portion, can be according to the practical business service environment of actual retrieval application depending on；Example
As, the word portion obtaining is split to cue, can be to be made up of the random partial character extracting in its corresponding cue, permissible
It is to be made up of one or several characters in its corresponding cue with designated character feature（For example it is appointed as capitalizing English words
Symbol）, can also be and be made up of one or several characters of specified location in its corresponding cue（Such as specified location is forward,
Placed in the middle or character rearward）, etc.；And the word portion data base that different word portions deconsolidation process mode is formed, the word portion that it comprises
The covering accuracy rate that associates and the search condition input habit of user between is not quite similar, can be by selecting the word of more optimization
Portion's deconsolidation process scheme covers accuracy rate come the association to lift word portion data base.On the other hand, the index of word portion data base is set up
Design, is also can be determined, different technologies personnel institute according to the practical business service environment of actual retrieval application
Grasp and the Index Design scheme of preference may be had nothing in common with each other；But different Index Design schemes, its index search processing procedure
In contrast number of times be then not quite similar, can by select more optimization Index Design scheme reduce as far as possible at index search
Contrast number of times during reason, and then further lift associated cue response speed.These factors cover accurately to association
The close ties of rate and associated cue response speed so that the associated cue processing method of the present invention exist deeper into retrieval
Application service quality room for promotion.Below just by embodiment, that may be present to the present invention deeper into technique effect do into
One step discussion and explanation.
Embodiment one：
For example in the retrieval of China's domestic zone weather forecast business service is applied, user needs first to retrieve its needs and looks into
Ask the provinces, municipalities and autonomous regions place name of weather forecast and place name at county level, if with the domestic each provinces, municipalities and autonomous regions place name of China and
Place name at county level sets up prompting dictionary as standby place name cue, then the cue quantity in prompting dictionary is up to tens thousand of
Individual.The concrete form of prompting dictionary can be expressed as shown in table 1：
Table 1：
Represent because being only used as form, the prompting dictionary shown in table 1 only enumerates a limited number of place name cue, other ground
Name cue omits and represents.
If adopting associated cue mode of the prior art, because computer system is set up only for prompting dictionary itself
There is index, if the search condition of user input is temporarily criteria character string " lake ", computer system cannot be directly from cue
Criteria character string " lake " therefore can only be added that asterisk wildcard " % " forms fuzzy afterwards by index search criteria character string " lake " in storehouse
Join entry " lake % ", this fuzzy matching entry is compared one by one with tens thousand of cue memory element in prompting dictionary and mates
Search, contrast matching times are up to tens thousand of times, finally just can determine that criteria character string " lake " matches " Hubei in prompting dictionary
The associated cue word such as province ", " Hunan Province ", is then prompted to user and selects.User is from input search condition " lake " to acquisition association
Cue " Hubei Province ", " Hunan Province " etc., the response time of period is likely to be breached second level, or even several seconds, response speed and user
Experience effect is all excessively poor.
It is directed to this, provide in the retrieval application that the present embodiment is serviced to this weather forecasts services using the inventive method
Search condition associated cue processing function is improved, and changes its associated cue processing method to search condition.The present embodiment
The associated cue processing method being adopted, its flow process is as shown in figure 1, concrete process step is as follows：
S1）Pre-build prompting dictionary corresponding word portion data base.
This step is used for being one data space of word portion data base's assignment of allocation in computer systems, sets up simultaneously
Storehouse corresponding relation between word portion data base and prompting dictionary.
S2）Obtaining number of characters in prompting dictionary by word portion deconsolidation process is plural cue each word corresponding
Portion is simultaneously recorded in institute predicate portion data base；Each word portion is being made up of partial character in its corresponding cue
Character string.
This step degree of being used for cue carries out word portion deconsolidation process and obtains corresponding word portion, and each word portion obtaining is recorded
In word portion data base, establish the corresponding points relationship between the corresponding cue in each word portion simultaneously.In view of lifting word
Covering accuracy rate is associated, in the present embodiment between word portion that portion's data place comprises and the search condition input habit of user
In, step S2 employs a kind of preferred word portion deconsolidation process embodiment, this preferred word portion deconsolidation process embodiment it
In, for plural cue, word portion deconsolidation process flow process is carried out as shown in Fig. 2 having to any one number of characters in prompting dictionary
Body is：
s21）Obtain the number of characters Nx of this cue, then Nx >=2, make default character properties n value be respectively from 1 to
The integer of Nx-1；
s22）For each value of character properties n, take the character string that front n character in this cue is constituted
As the corresponding word portion of this cue；It is derived from the corresponding Nx-1 word portion of this cue, the word portion to this cue
Deconsolidation process completes.
By upper predicate portion deconsolidation process scheme, for plural cue, word portion is all carried out to number of characters in prompting dictionary
After deconsolidation process, then will be prompted to number of characters in dictionary and be recorded in institute's predicate for plural cue each word portion corresponding
In portion data base, step S2 then processes and completes.
Because user is when inputting place name, input habit the most universal is each character sequentially inputting place name；Cause
This, the word portion deconsolidation process scheme preferably employing in the present embodiment, acquisition can accordingly be split by the cue for Nx for the number of characters
Nx-1 word portion, and this Nx-1 word portion covers the sequence input condition of n character before the Nx-1 kind of cue.With regard to this enforcement
For example, for example, for the place name cue " Heilongjiang Province " in prompting dictionary, its number of characters is 4, and split acquisition corresponds to
3 word portions be respectively " black ", " black dragon " and " Heilungkiang "；For the place name cue " Hubei Province " in prompting dictionary, its word
Symbol number is 3, and the corresponding 2 word portions of split acquisition are respectively " lake " and " Hubei "；For the place name prompting in prompting dictionary
Word " Hunan Province ", its number of characters is 3, and the corresponding 2 word portions of split acquisition are respectively " lake " and " Hunan "；For cue
Place name cue " Ningxia Hui Autonomous Region " in storehouse, its number of characters is 7, and the corresponding 6 word portions of split acquisition are respectively
" peaceful ", " Ningxia ", " Hui Nation In Ningxia ", " Hui Nationality in Ningxia Hui Nationality Autonomy ", " Hui Nationality in Ningxia Hui Nationality Autonomy is certainly " and " Hui Nationality in Ningxia Hui Nationality Autonomy is autonomous "；Etc..As can be seen here, should
Preferably word portion deconsolidation process scheme is so as to contain user as much as possible the most universal word portion that word portion data place comprises
The various situations of search condition input habit cover accuracy rate it is ensured that word portion data base possesses higher association, contribute to into
One step improves the utilization rate of word portion data base；And, this preferred word portion deconsolidation process scheme has relatively regular deconsolidation process
Rule, easily can realize computer system automatization word portion deconsolidation process by computer programming.
S3）Set up the index of institute predicate portion data base；
This step is used for for its special index of word portion Database so that computer system passes through search index
Mode inquires target word portion from word portion data base and possesses index basis.And scheme set up in the index of word portion data base, and
Directly affects the later stage to the contrast number of times in conjunctive word portion index search processing procedure；Index in view of reducing the later stage as far as possible
Search the contrast number of times of processing procedure, further lift associated cue response speed, in the present embodiment, step S3 adopts
A kind of preferred word portion database index constitution and implementation scheme, this preferred word portion database index sets up the flow process of scheme such as
Shown in Fig. 3, specially：
s31）Each word portion in word portion data base with identical characters number is lined up as a word portion index column, by word portion
In index column, the number of characters in single word portion identifies ID as the row of this word portion index column, and calculates each word portion in word portion index column
Index value, set up the corresponding column index of word portion index column；
s32）By execution step s21 repeatedly, all word portions in word portion data base are lined up as multiple word portions index column,
And set up and obtain each corresponding column index of word portion index column, the set of each corresponding column index of word portion index column is i.e. as word
The index of portion data base.
For the present embodiment, thus obtained word portion index column line up situation and its each word portion respectively with prompting
In dictionary, the corresponding points relationship of each place name cue is represented by table 2：
Table 2：
Represent because being only used as form, the situation of lining up of the word portion index column shown in table 2 only enumerates in word portion index column and has
Several word portions of limit, other word portions omit and represent；In table 2, symbol " RID " represents the row mark ID of word portion index column, symbol " ＆ "
Represent the corresponding points relationship between word portion and the place name cue in word portion index column.This preferred word portion database index is built
Cube case, the advantage of one side is, it has relatively regular index and sets up rule, can be easily real by computer programming
Existing computer system automatic Building lithol draws；On the other hand, compared with whole word portion data base is directly set up with overall index, should
Preferably word portion database index is set up scheme and is lined up as multiple word portions index column by all word portions in word portion data base, and builds
Stand and obtain each corresponding column index of word portion index column so that the later stage can pass through in conjunctive word portion index search processing procedure
Row positioning reduces contrast number of times further, thus further lifting associated cue response speed, the effect of this one side will
Follow-up process step is more specifically embodied.
S4）In user input search condition, obtain the criteria character string as search condition and calculate its index value, lead to
Cross the word that the index search of word portion data base has same index value to each of word portion data base with described criteria character string
Portion, respectively as conjunctive word portion, each corresponding cue in conjunctive word portion is prompted to user as associated cue word.
This step is used for processing operation for the search condition execution associated cue of user input.Concrete place in step S4
Reason process will set up the difference of scheme with word portion database index employed in step S3 and different；For example, if being based on
In step S3 in the present embodiment, preferred word portion database index sets up scheme, the handling process in this step S4 as shown in figure 4,
It is specially：
s41）Obtain the criteria character string as search condition, and calculate the index value of described criteria character string；
s42）Judge that the row of each word portion index column identify in ID with the presence or absence of the number of characters phase with described criteria character string
Deng row mark ID；If existing, execution step s43；If not existing, judge currently not obtaining associated cue word；
s43）Using have row equal with the number of characters of described criteria character string identify the word portion index column of ID as
Whether affiliated partner index column, searched by the corresponding column index of affiliated partner index column and judge in described affiliated partner index column
There is the word portion with described criteria character string with same index value；If existing, execution step s44；If not existing, judge
Currently do not obtain associated cue word；
s44）Each of described affiliated partner index column and described criteria character string had the word portion of same index value
Respectively as conjunctive word portion, and then the cue corresponding to using each conjunctive word portion is prompted to use as associated cue word
Family.
For the present embodiment, for example, if the search condition of user input is temporarily criteria character string " lake ", computer
System, by obtaining the criteria character string " lake " as search condition, calculates its index value, and learns criteria character by judgement
String " lake " number of characters is 1, then the value of row mark ID is equal to 1（I.e. RID=1）Word portion index column as affiliated partner index
Row, that is, be equivalent to and carried out row positioning in word portion data base；Then, then by the corresponding column index of affiliated partner index column look into
Find at least two, the word portion " lake " with same index value, by it respectively as conjunctive word portion, this time index search is processed
During, due to affiliated partner index column be only several word portion index column of comprising of word portion data base one of, this pass
Connection object indexing the comprised word portion quantity of row is certainly less than whole word portion data place and comprises word portion quantity, therefore with to whole
Word portion data base is indexed inquiry after directly setting up overall index and compares, and this corresponding column index of affiliated partner index column is entered
The experienced contrast number of times of line index inquiry will necessarily be less, thus the entirety contributing to further being lifted associated cue is rung
Answer speed；Finally, by each conjunctive word portion " lake " each corresponding points relationship between cue, at least two are obtained
Associated cue word, respectively " Hubei " and " Hunan ", thus the associated cue word such as " Hubei ", " Hunan " is prompted to user's choosing
Select.Using the associated cue processing method of the present invention, because the coupling contrast being related in whole prompting processing procedure is all by inspection
Rope is searched and is processed and complete, and has broken away from and each cue memory element in prompting dictionary is compared matched and searched one by one and looks into
Look for processing mode, therefore user arrives acquisition associated cue word " Hubei Province ", " Hunan Province " etc., period from input search condition " lake "
Response time only need to tens microseconds, or even a few microsecond, response speed is ideal, then experiences input for a user
After search condition, its associated cue word is just presented immediately, has very good Consumer's Experience effect.
Certainly, in present invention retrieval application, the associated cue processing method of search condition not only limits to that to be applied to weather pre-
In the retrieval application of report business service, for any service traffic class, as long as wherein needing to use retrieval application, such as
Bibliographic data base phrase vocabulary retrieval, the key message retrieval of the English word and search of English dictionary, internet, enterprises
Retrieving contacts etc., then in its retrieval application, the associated cue of search condition processes and all can adopt the inventive method realization.
As for calculating the specific algorithm that index value is adopted in the inventive method, can be according to the specific service service environment of retrieval application
From index value-based algorithm commonly used in the prior art.Consideration as more excellent scheme is it is proposed that calculate the present invention using hash algorithm
The index value being related in method, that is, in the index of institute predicate portion data base, the index value in each word portion is preferably the Kazakhstan in corresponding words portion
Uncommon value；The index value of described criteria character string is preferably the cryptographic Hash of this criteria character string；Why using hash algorithm as excellent
Choosing, is because hash algorithm by the binary value of random length character string（Character string is all presented as that two enter in computer systems
Value processed）The value being all mapped as the less binary value of regular length, i.e. cryptographic Hash, and cryptographic Hash for identical characters string is unique
Property very good, value length is also comparatively short, and computing good stability, computer processing time are short, for lifting associated cue of the present invention
The treatment effeciency of processing method and response speed also have certain help.
Embodiment two：
Corresponding, present invention also offers a kind of associated cue processing system retrieving search condition in application, in order to reality
The associated cue processing method of search condition in existing present invention retrieval application.At the associated cue of search condition in this retrieval application
Reason system is as shown in figure 5, main include prompting dictionary 10, word portion Database module 20, word portion processing module 30, index and build
Formwork erection block 40 and associated cue processing module 50；Prompting dictionary 10 is used for recording cue；Word portion Database module 20 is used
Point out dictionary corresponding word portion data base in setting up；Word portion processing module 30 is used for obtaining prompting dictionary by word portion deconsolidation process
Middle number of characters for plural cue each word portion corresponding and is recorded in institute predicate portion data base, and each word portion is
A character string being made up of partial character in its corresponding cue；Index is set up module 40 and is used for setting up institute's predicate portion number
Index according to storehouse；Associated cue processing module 50 is used for, in user input search condition, obtaining the condition as search condition
Character string simultaneously calculates its index value, by each of index search of word portion data base to word portion data base and described condition word
Symbol string has the word portion of same index value respectively as conjunctive word portion, and the cue corresponding to each conjunctive word portion is made respectively
It is prompted to user for associated cue word.These modules all can be utilized the computer programming skill of its grasp by those skilled in the art
Art is achieved by programming.
Preferably, word portion processing module 30 mainly can process list by word portion deconsolidation process unit and word portion record
Unit is constituted；It is plural carrying that word portion deconsolidation process unit is used for obtaining number of characters in prompting dictionary by word portion deconsolidation process
Show word each word portion corresponding；Wherein, carry out word portion to any one number of characters in prompting dictionary for plural cue to tear open
Office reason is specially：Obtain the number of characters Nx of this cue, then Nx >=2, make the value of default character properties n be respectively from 1
Integer to Nx-1；For each value of character properties n, take the character that front n character in this cue is constituted
String is as the corresponding word portion of this cue；It is derived from the corresponding Nx-1 word portion of this cue, the word to this cue
Portion's deconsolidation process completes；Word portion record processing unit will be prompted in dictionary number of characters be plural cue corresponding each
Word portion is recorded in institute predicate portion data base.Programmed using this preferred version and build word portion processing module 30, you can in terms of passing through
Calculation machine self-operating realizes preferred word portion deconsolidation process scheme described in embodiment one it is ensured that word portion data base possesses higher pass
Connection covers accuracy rate, to help the utilization rate improving word portion data base further.
Preferably, index is set up module 40 and mainly can be lined up processing unit, be arranged mark processing unit by word portion
Set up unit with column index to constitute；Word portion lines up processing unit for having each word of identical characters number in word portion data base
Portion lines up for a word portion index column, thus, all word portions in word portion data base to be lined up as multiple word portions index column；Row mark
Know processing unit be used for respectively using the number of characters in single word portion in each word portion index column as corresponding words portion index column row
Mark ID；Column index sets up unit for calculating the index value in each word portion in each word portion index column, sets up each respectively
The corresponding column index of individual word portion index column.Module 40 is set up using this preferred version programming index building, you can with by calculating
Machine self-operating is realized preferred word portion database index described in embodiment one and is set up scheme.
Further, set up the preferred version of module 40 based on above-mentioned index, correspondingly, associated cue processing module 50 is then led
Will be by search condition index value computing unit, affiliated partner index column judging unit, conjunctive word portion judging unit and associated cue
Word Tip element is constituted；Search condition index value computing unit is used for obtaining the criteria character string as search condition, and calculates
The index value of described criteria character string；Affiliated partner index column judging unit is used for judging the row mark ID of each word portion index column
In identify ID with the presence or absence of row equal with the number of characters of described criteria character string, if existing, triggering runs conjunctive word portion and sentences
Disconnected unit, if not existing, judges currently not obtaining associated cue word；Conjunctive word portion judging unit will be for having and described bar
The row that the number of characters of part character string is equal identify a word portion index column of ID as affiliated partner index column, by affiliated partner
The corresponding column index of index column is searched to judge to whether there is in described affiliated partner index column has phase with described criteria character string
With the word portion of index value, if existing, triggering runs associated cue word Tip element, if not existing, judges currently not obtaining
Associated cue word；Associated cue word Tip element is used for each of described affiliated partner index column and described criteria character string
The word portion with same index value is respectively as conjunctive word portion, and then the cue corresponding to each conjunctive word portion is made respectively
It is prompted to user for associated cue word.Programmed using this preferred version and build associated cue processing module 50, in conjunction with above-mentioned index
Set up the preferred version of module 40 so that reducing further to positioning by row in conjunctive word portion index search processing procedure
Contrast number of times, thus further lift the Whole Response speed of associated cue.
The illustration of two embodiments of summary is it can be seen that the present invention retrieves the association of search condition in application
Prompting processing method and system, all by increasing the rope setting up prompting dictionary corresponding word portion data base and word portion data base
Draw so that computer system inquires target word portion by way of search index from word portion data base possesses index base
The coupling contrast being related in plinth, and then whole prompting processing procedure is all processed by retrieval and completes, and has broken away to prompting
In dictionary, each cue memory element is compared the lookup processing mode of matched and searched one by one, though what prompting dictionary comprised
Cue data volume is huge, and computer system also can be by index search processing mode after contrast for several times few in number
The associated cue word obtaining search condition is prompted to user, thus improve computer system in itself examining in retrieval application
The associated cue treatment effeciency of rope condition and response speed；And also can by the word portion deconsolidation process that optimizes further and
The database index foundation of word portion is processed, and the association of further lifting word portion data base covers accuracy rate and associated cue response
Speed, brings more preferable experience to user, from the loyalty to retrieval application product for the lifting user.
Finally illustrate, above example only in order to technical scheme to be described and unrestricted, although with reference to real
Apply example the present invention has been described in detail, it will be understood by those within the art that, can be to the technical side of the present invention
Case is modified or equivalent, the objective without deviating from technical solution of the present invention and scope, and it all should be covered in the present invention
Right in the middle of.
Claims (5)
1. in a kind of retrieval application, the associated cue processing method of search condition points out dictionary pair it is characterised in that pre-building
The word portion data base answering, obtaining number of characters in prompting dictionary by word portion deconsolidation process is that plural cue is corresponding each
Individual word portion is simultaneously recorded in institute predicate portion data base, sets up the index of institute predicate portion data base；Each word portion is by its correspondence
Cue in a character string being constituted of partial character；In user input search condition, obtain as search condition
Criteria character string simultaneously calculates its index value, by each of index search of word portion data base to word portion data base and described bar
Part character string has the word portion of same index value respectively as conjunctive word portion, by the corresponding cue in each conjunctive word portion difference
It is prompted to user as associated cue word；
Described " setting up the index of institute predicate portion data base " is specially：To there is each word of identical characters number in word portion data base
Portion lines up as a word portion index column, using the number of characters in word portion single in word portion index column as this word portion index column row mark
ID, and calculate the index value in each word portion in word portion index column, set up the corresponding column index of word portion index column；Thus, by word portion
All word portions in data base line up as multiple word portions index column, and set up and obtain each corresponding column index of word portion index column,
The set i.e. index as word portion data base of each corresponding column index of word portion index column；
In user input search condition, described " obtain as the criteria character string of search condition and calculate its index value, pass through
The index search of word portion data base has the word portion of same index value to each of word portion data base with described criteria character string
Respectively as conjunctive word portion, the cue corresponding to using each conjunctive word portion is prompted to user as associated cue word "
Concrete process comprise the steps：
c1）Obtain the criteria character string as search condition, and calculate the index value of described criteria character string；
c2）Judge that the row of each word portion index column identify in ID with the presence or absence of the row equal with the number of characters of described criteria character string
Mark ID；If existing, execution step c3；If not existing, judge currently not obtaining associated cue word；
c3）Will be right as associating for a word portion index column with the row mark ID equal with the number of characters of described criteria character string
As index column, searched by the corresponding column index of affiliated partner index column judge in described affiliated partner index column with the presence or absence of with
Described criteria character string has the word portion of same index value；If existing, execution step c4；If not existing, judge currently not
Obtain associated cue word；
c4）The word portion that each of described affiliated partner index column is had same index value with described criteria character string makees respectively
For conjunctive word portion, and then the cue corresponding to using each conjunctive word portion is prompted to user as associated cue word.
2. in retrieval application according to claim 1 the associated cue processing method of search condition it is characterised in that described
Among " obtaining number of characters in prompting dictionary by word portion deconsolidation process is plural cue each word portion corresponding ", right
Point out any one number of characters in dictionary to carry out word portion deconsolidation process for plural cue to be specially：Obtain this cue
Number of characters Nx, then Nx >=2, make the value of default character properties n be respectively the integer from 1 to Nx-1；For character properties n's
Each value, takes the character string that front n character in this cue is constituted as the corresponding word portion of this cue；
It is derived from the corresponding Nx-1 word portion of this cue, the word portion deconsolidation process of this cue is completed.
3. in retrieval application according to claim 1 the associated cue processing method of search condition it is characterised in that described
In the index of word portion data base, the index value in each word portion is the cryptographic Hash in corresponding words portion；The index value of described criteria character string is
The cryptographic Hash of this criteria character string.
4. a kind of retrieval application in search condition associated cue processing system it is characterised in that this system include point out dictionary,
Module and associated cue processing module set up in word portion Database module, word portion processing module, index；
Described prompting dictionary is used for recording cue；
Institute's predicate portion Database module is used for setting up prompting dictionary corresponding word portion data base；
It is plural prompting that institute's predicate portion processing module is used for obtaining number of characters in prompting dictionary by word portion deconsolidation process
Word each word portion corresponding is simultaneously recorded in institute predicate portion data base；Each word portion is by partial words in its corresponding cue
Accord with a constituted character string；
Described index sets up module for setting up the index of institute predicate portion data base；Index set up module include word portion line up process
Unit, row mark processing unit and column index set up unit；Predicate portion of institute lines up processing unit for having in word portion data base
Each word portion having identical characters number lines up as a word portion index column；Thus, all word portions in word portion data base are lined up
For multiple word portions index column；Described row mark processing unit is used for respectively by the character in single word portion in each word portion index column
Number identifies ID as the row of corresponding words portion index column；Described column index is set up unit and is used for calculating in each word portion index column respectively
The index value in individual word portion, sets up each corresponding column index of word portion index column respectively；Described associated cue processing module is used for
In user input search condition, obtain the criteria character string as search condition and calculate its index value, by word portion data
The index search in storehouse to each of word portion data base and described criteria character string have the word portion of same index value respectively as
Conjunctive word portion, the cue corresponding to using each conjunctive word portion is prompted to user as associated cue word；Described association
Prompting processing includes search condition index value computing unit, affiliated partner index column judging unit, conjunctive word portion judgement list
Unit and associated cue word Tip element；Described search condition index value computing unit is used for obtaining the condition word as search condition
Symbol string, and calculate the index value of described criteria character string；Described affiliated partner index column judging unit is used for judging each word portion
The row of index column identify in ID and identify ID with the presence or absence of the row equal with the number of characters of described criteria character string；If existing, touch
Shipping row conjunctive word portion judging unit；If not existing, judge currently not obtaining associated cue word；Described conjunctive word portion judges single
Unit is right as associating for the word portion index column that will have the row mark ID equal with the number of characters of described criteria character string
As index column, searched by the corresponding column index of affiliated partner index column judge in described affiliated partner index column with the presence or absence of with
Described criteria character string has the word portion of same index value；If existing, triggering runs associated cue word Tip element；If not depositing
Then judging currently not obtaining associated cue word；Described associated cue word Tip element is used for described affiliated partner index column
Each of there is the word portion of same index value respectively as conjunctive word portion with described criteria character string, and then each is associated
Cue corresponding to word portion is prompted to user respectively as associated cue word.
5. in retrieval application according to claim 4 the associated cue processing system of search condition it is characterised in that described
Word portion processing module includes word portion deconsolidation process unit and word portion record processing unit；
It is plural that institute's predicate portion deconsolidation process unit is used for obtaining number of characters in prompting dictionary by word portion deconsolidation process
Cue each word portion corresponding；Wherein, for plural cue, word portion is carried out to any one number of characters in prompting dictionary
Deconsolidation process is specially：Obtain the number of characters Nx of this cue, then Nx >=2, make default character properties n value be respectively from
1 integer arriving Nx-1；For each value of character properties n, take the character that front n character in this cue is constituted
String is as the corresponding word portion of this cue；It is derived from the corresponding Nx-1 word portion of this cue, the word to this cue
Portion's deconsolidation process completes；
It is that plural cue each word portion corresponding is remembered that predicate portion of institute record processing unit will be prompted to number of characters in dictionary
Record in institute predicate portion data base.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210221725.4A CN103514217B (en) | 2012-06-30 | 2012-06-30 | Method and system for processing associated prompts of retrieval condition of retrieval application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210221725.4A CN103514217B (en) | 2012-06-30 | 2012-06-30 | Method and system for processing associated prompts of retrieval condition of retrieval application |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103514217A CN103514217A (en) | 2014-01-15 |
| CN103514217B true CN103514217B (en) | 2017-02-08 |
Family
ID=49896950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210221725.4A Active CN103514217B (en) | 2012-06-30 | 2012-06-30 | Method and system for processing associated prompts of retrieval condition of retrieval application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103514217B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107169046A (en) * | 2017-04-25 | 2017-09-15 | 广东网金控股股份有限公司 | A kind of database index lookup method, device and user terminal |
| CN107844526B (en) * | 2017-10-12 | 2022-04-01 | 广州艾媒数聚信息咨询股份有限公司 | Knowledge base-based vocabulary relation chain analysis method, system and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1940922A (en) * | 2005-09-30 | 2007-04-04 | 腾讯科技（深圳）有限公司 | Method and system for improving information search speed |
| CN101662541A (en) * | 2009-08-24 | 2010-03-03 | 宇龙计算机通信科技(深圳)有限公司 | Prompting method, system and mobile terminal of related information of contact persons at mobile terminal |
| WO2011024064A2 (en) * | 2009-08-31 | 2011-03-03 | Exalead S.A. | Trusted query system and method |
| CN102411568A (en) * | 2010-09-20 | 2012-04-11 | 苏州同程旅游网络科技有限公司 | Chinese word segmentation method based on travel industry feature word stock |
-
2012
- 2012-06-30 CN CN201210221725.4A patent/CN103514217B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1940922A (en) * | 2005-09-30 | 2007-04-04 | 腾讯科技（深圳）有限公司 | Method and system for improving information search speed |
| CN101662541A (en) * | 2009-08-24 | 2010-03-03 | 宇龙计算机通信科技(深圳)有限公司 | Prompting method, system and mobile terminal of related information of contact persons at mobile terminal |
| WO2011024064A2 (en) * | 2009-08-31 | 2011-03-03 | Exalead S.A. | Trusted query system and method |
| CN102411568A (en) * | 2010-09-20 | 2012-04-11 | 苏州同程旅游网络科技有限公司 | Chinese word segmentation method based on travel industry feature word stock |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103514217A (en) | 2014-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609052A (en) | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle | |
| CN104216895B (en) | A kind of method and device for generating POI data | |
| CN101950284B (en) | Chinese word segmentation method and system | |
| CN103514236B (en) | Search condition error correcting prompt processing method based on phonetic in retrieval application | |
| CN104375992B (en) | A kind of method and apparatus of address matching | |
| CN101398835B (en) | Service selecting system and method, and service enquiring system and method based on natural language | |
| CN105917327A (en) | System and method for inputting text into electronic devices | |
| CN106156082B (en) | A kind of ontology alignment schemes and device | |
| CN100578539C (en) | Automatic question-answering method and system | |
| CN102982076A (en) | Multi-dimensionality content labeling method based on semanteme label database | |
| CN102955833A (en) | Correspondence address identifying and standardizing method | |
| CN103345496B (en) | multimedia information retrieval method and system | |
| CN101398810A (en) | Self-adapting service choice device and method thereof, enquiry system and method thereof | |
| CN104346444B (en) | A kind of the best site selection method based on the anti-spatial key inquiry of road network | |
| CN105956206A (en) | Video retrieval method based on keyword tree and video retrieval system based on keyword tree | |
| CN105976048A (en) | Power transmission network extension planning method based on improved artificial bee colony algorithm | |
| CN109977309B (en) | Combined interest point query method based on multiple keywords and user preferences | |
| CN106844544A (en) | A kind of contract terms Risk Identification Method and system | |
| CN104391908B (en) | Multiple key indexing means based on local sensitivity Hash on a kind of figure | |
| CN106874422A (en) | A kind of figure querying method of facing relation type database | |
| CN102955832A (en) | Correspondence address identifying and standardizing system | |
| CN107085583A (en) | A kind of electronic document management method and device based on content | |
| CN107832047A (en) | A kind of non-api function argument based on LSTM recommends method | |
| CN107102993A (en) | A kind of user's demand analysis method and device | |
| CN107239512A (en) | The microblogging comment spam recognition methods of relational network figure is commented in a kind of combination |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200605 Address after: Room 502-1, floor 5, building 2, courtyard 10, KEGU 1st Street, economic development zone, Daxing District, Beijing 100081 Patentee after: Singularity Xinyuan International Technology Development (Beijing) Co.,Ltd. Address before: The 401121 northern New District of Chongqing municipality Mount Huangshan Road 5 south of Mercury Technology Building 1 floor office No. 3 Patentee before: A-MEDIA COMMUNICATION TECH Co.,Ltd. |
|
| TR01 | Transfer of patent right |