CN105893552A - Method and device for processing data - Google Patents
Method and device for processing data Download PDFInfo
- Publication number
- CN105893552A CN105893552A CN201610200992.1A CN201610200992A CN105893552A CN 105893552 A CN105893552 A CN 105893552A CN 201610200992 A CN201610200992 A CN 201610200992A CN 105893552 A CN105893552 A CN 105893552A
- Authority
- CN
- China
- Prior art keywords
- client
- answer
- server
- corpus
- instruction
- 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.)
- Granted
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Mathematical Physics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a method and device for processing data. The method comprises the steps that a server receives input data of a client side, the input data is subjected to word division, and a first word set obtained after word division is obtained; words with the information amount smaller than a first preset threshold value are removed from the first word set through the server, and a second word set is obtained; the server calculates a second weight value corresponding to an answer pre-stored in each corpus knowledge unit in a pre-generated corpus according to the second word set, each corpus knowledge unit in the corpus comprises the answer corresponding to a user question and a first weight value corresponding to the answer; the server ranks the answers according to the first weight values and the second weight values corresponding to the answers in the corpus knowledge units in the corpus, and recommended answers are returned to a client side according to the ranking result. According to the method, the manpower of customer service staff can be freed up, and convenience is provided for work of the customer service staff.
Description
Technical field
The present invention relates to data processing field, process in particular to a kind of data
Method and device.
Background technology
The lifting to electric quotient module formula acceptance along with the universal of the Internet and the public, user
Customer service is supported that demand for services is steady growth, is typically now that artificial customer service is to user
The demand proposed processes, and contact staff needs to process substantial amounts of workload, efficiency
The lowest, the experience of user is the most bad.
Summary of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of data process side
Method and device.
First aspect, a kind of data processing method that the embodiment of the present invention provides, application
In customer service field, described method includes: server receives the input data of client,
Described input data are carried out participle, it is thus achieved that the first word set after participle；Described service
Quantity of information is removed from described first word set by device less than the word of the first predetermined threshold value, obtains
Obtain the second word set；Described server, according to described second word set, calculates pre-stored respectively
Answer described in each language material blocks of knowledge is corresponding in the corpus previously generated
Two weighted values, each language material blocks of knowledge in wherein said corpus includes that client carries
Ask the answer of correspondence and the first weighted value that described answer is corresponding；Described server root
According to answer described in language material blocks of knowledge described in described corpus corresponding described first
Described answer is ranked up by weighted value and described second weighted value, and according to sequence
Result returns to described client recommends answer.
Further, return to described client to described client at described server
After recommending answer, described method also includes: described service end receives described client
Removal request instruction, described removal request instruction include described client request delete
Answer corresponding to client questions；When described server inquires in described corpus
During answer corresponding to described client questions in the instruction of described removal request, by described visitor
The first weighted value that family puts question to corresponding answer corresponding reduces.
Further, return to described client to described client at described server
After recommending answer, described method also includes: described service end receives described client
Amendment instruction, described amendment instruction include described client request amendment client
Put question to corresponding answer and amended answer；When described server is at described language material
When storehouse inquires answer corresponding to described client questions in described amendment instruction, will
The answer that described in described corpus, client questions is corresponding replaces with described amended answer
Case, and by the first corresponding for answer corresponding for client questions described in described corpus power
Weight values increases.
Further, return to described client to described client at described server
After recommending answer, described method also includes: described service end receives described client
Interpolation instruction, described add instruction include answer and the institute that client questions is corresponding
State the satellite information that client request is added；When described server is in described corpus
Inquire the answer that the described described client questions added in instruction is corresponding, by described attached
Genus information adds the described language material blocks of knowledge of correspondence to, and by described client questions pair
First weighted value corresponding to the answer answered increases.
Further, return to described client to described client at described server
After recommending answer, described method also includes: described service end receives described client
Newly-built instruction, described newly-built instruction includes the described client newly-built client of request
Put question to corresponding answer；Described server creates described newly-built finger in described corpus
The answer that client questions described in order is corresponding, and by answer corresponding for described client questions
The first corresponding weighted value is entered as the 3rd predetermined threshold value.
Further, return to described client to described client at described server
After recommending answer, described method also includes: described service end receives described client
Second input data, it is judged that described second input data is and described recommendation answer phase
With, when described second inputs data and described recommendation answer differs, described service
Described second input data are added in described corpus by device.
Second aspect, a kind of data processing equipment that the embodiment of the present invention provides, application
In customer service field, described device includes: word-dividing mode, receives visitor for server
Described input data are carried out participle by the input data of family end, it is thus achieved that after participle
One word set；First processing module, for described server by pre-less than first for quantity of information
If the word of threshold value removes from described first word set, it is thus achieved that the second word set；Computing module,
For described server according to described second word set, calculate respectively and be pre-stored in pre-Mr.
The second weight that in the corpus become, described in each language material blocks of knowledge, answer is corresponding
Value, each language material blocks of knowledge in wherein said corpus includes that client questions is corresponding
Answer and the first weighted value corresponding to described answer；Order module, for described
Server is corresponding according to answer described in language material blocks of knowledge described in described corpus
Described answer is ranked up by described first weighted value and described second weighted value, and
Recommendation answer is returned to described client according to ranking results.
Further, described device also includes the second processing module, and described second processes
Module, for described order module at described server to described client to described visitor
Family end returns after recommending answer, and described service end receives the deletion of described client please
Instruction, the instruction of described removal request is asked to include that the client that the request of described client is deleted carries
Ask the answer of correspondence；When described server inquires described deletion in described corpus
During answer corresponding to described client questions in request instruction, by described client questions pair
First weighted value corresponding to the answer answered reduces.
Further, described device also includes the 3rd processing module, and the described 3rd processes
Module, for described order module at described server to described client to described visitor
Family end returns after recommending answer, and described service end receives the amendment of described client and refers to
Order, described amendment instruction includes that the client questions of described client request amendment is corresponding
Answer and amended answer；When described server is inquired about in described corpus
During to the answer that the described described client questions revised in instruction is corresponding, by described language material
The answer that client questions described in storehouse is corresponding replaces with described amended answer, and will
First weighted value corresponding to answer that described in described corpus, client questions is corresponding increases
Greatly.
Further, described device also includes fourth processing module, described fourth process
Module, for described order module at described server to described client to described visitor
Family end returns after recommending answer, and described service end receives the interpolation of described client and refers to
Order, described add instruction and includes answer that client questions is corresponding and described client
The satellite information that request is added；When described server inquires institute in described corpus
State the answer that the described client questions added in instruction is corresponding, described satellite information is added
It is added to the described language material blocks of knowledge of correspondence, and by answer corresponding for described client questions
The first corresponding weighted value increases.
Further, described device also includes the 5th processing module, and the described 5th processes
Module, is at described server to described client to institute for described order module
After stating client return recommendation answer, described service end receives the new of described client
Building instruction, described newly-built instruction includes the client questions that the request of described client is newly-built
Corresponding answer；Described server creates in described newly-built instruction in described corpus
The answer that described client questions is corresponding, and answer corresponding for described client questions is corresponding
The first weighted value be entered as the 3rd predetermined threshold value.
Further, described device also includes the 6th processing module, and the described 6th processes
Module, is at described server to described client to institute for described order module
Stating client to return and recommend after answer, described service end receives the of described client
Two input data, it is judged that described second input data are identical with described recommendation answer,
When described second inputs data and described recommendation answer differs, described server will
Described second input data are added in described corpus.
Compared with prior art, a kind of data processing method that the embodiment of the present invention provides
And device, by the input data of client being carried out participle and removing quantity of information minimizing
Word after, according to the word set that described size of message is higher, previously generate from being pre-stored in
Corpus obtains recommendation by process in the described answer of each language material blocks of knowledge answer
Case, and by feedback mechanism, enrich constantly and improve corpus so that customer service people
Member has only to importation key word and just can generate from the corpus previously generated properly
Answer, and without inputting complete sentence, the manpower of contact staff can be discharged,
Work for contact staff provides convenient.
For making the above-mentioned purpose of the present invention, feature and advantage to become apparent,
Preferred embodiment cited below particularly, and coordinate appended accompanying drawing, it is described in detail below.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below will be to reality
Execute the required accompanying drawing used in example to be briefly described, it will be appreciated that the following drawings
Illustrate only certain embodiments of the present invention, be therefore not construed as the limit to scope
Fixed, for those of ordinary skill in the art, before not paying creative work
Put, it is also possible to obtain other relevant accompanying drawings according to these accompanying drawings.
Fig. 1 is the server of embodiment of the present invention offer and showing that client interacts
It is intended to.
Fig. 2 is the block diagram of the server that the embodiment of the present invention provides.
Fig. 3 is the flow chart of a kind of data processing method that the embodiment of the present invention provides.
Fig. 4 is the flow process of a kind of data processing method that another embodiment of the present invention provides
Figure.
Fig. 5 is the functional module of a kind of data processing equipment that the embodiment of the present invention provides
Schematic diagram.
Fig. 6 is the function of a kind of data processing equipment that another embodiment of the present invention provides
Module diagram.
Detailed description of the invention
Below in conjunction with accompanying drawing in the embodiment of the present invention, to the skill in the embodiment of the present invention
Art scheme is clearly and completely described, it is clear that described embodiment is only
The a part of embodiment of the present invention rather than whole embodiments.Generally at accompanying drawing herein
Described in and the assembly of the embodiment of the present invention that illustrates can come with various different configurations
Arrange and design.Therefore, below to the embodiments of the invention provided in the accompanying drawings
Describe the scope of claimed invention of being not intended to limit in detail, but only table
Show the selected embodiment of the present invention.Based on embodiments of the invention, people in the art
The every other embodiment that member is obtained on the premise of not making creative work,
Broadly fall into the scope of protection of the invention.
It should also be noted that similar label and letter represent similar terms in following accompanying drawing,
Therefore, the most a certain Xiang Yi accompanying drawing is defined, then in accompanying drawing subsequently not
Need it to be defined further and explains.Meanwhile, in describing the invention,
Term " first ", " second " etc. be only used for distinguish describe, and it is not intended that instruction or
Hint relative importance.
As it is shown in figure 1, be server 200 and the client of embodiment of the present invention offer
100 schematic diagrams interacted.Described server 200 is by network 300 and one
Or multiple client 100 is communicatively coupled, to carry out data communication or mutual.Institute
Stating server 200 can be the webserver, database server etc..Described client
End 100 can be PC (personal computer, PC), panel computer,
Smart mobile phone, personal digital assistant (personal digital assistant, PDA) etc..
As in figure 2 it is shown, be the block diagram of described server 200.Described service
Device 200 includes data processing equipment, memorizer 220, storage control 230, processes
Device 240.
Described memorizer 220, storage control 230, each element of processor 240 are mutual
Between be electrically connected with directly or indirectly, to realize the transmission of data or mutual.Such as,
These elements can realize electricity by one or more communication bus or holding wire each other
Property connect.Described data processing equipment includes that at least one can be with software or firmware
(firmware) form is stored in described memorizer or is solidificated in described server
Software function module in the operating system (operating system, OS) of 200.
Described processor 240 is for performing the executable module of storage, example in memorizer 220
The software function module included such as described data processing equipment or computer program.
Wherein, memorizer 220 may be, but not limited to, random access memory
(Random Access Memory, RAM), read only memory (Read Only
Memory, ROM), programmable read only memory (Programmable Read-Only
Memory, PROM), erasable read-only memory (Erasable Programmable
Read-Only Memory, EPROM), electricallyerasable ROM (EEROM) (Electric
Erasable Programmable Read-Only Memory, EEPROM) etc..Its
In, memorizer 220 is used for storing program, and described processor 240 is receiving execution
After instruction, performing described program, aforementioned embodiment of the present invention any embodiment discloses
Flow through the method performed by the server of Cheng Dingyi to can apply in processor, or
Realized by processor.
Processor 240 is probably a kind of IC chip, has the process energy of signal
Power.Above-mentioned processor can be general processor, including central processing unit
(Central Processing Unit is called for short CPU), network processing unit
(Network Processor is called for short NP) etc.；Can also is that digital signal processor
(DSP), special IC (ASIC), ready-made programmable gate array (FPGA) or
Other PLDs, discrete gate or transistor logic, discrete hardware
Assembly.Can realize or perform the disclosed each method in the embodiment of the present invention, step
Rapid and logic diagram.General processor can be microprocessor or this processor also may be used
To be the processor etc. of any routine.
Fig. 3 shows the stream of a kind of data processing method that one embodiment of the invention provides
Cheng Tu, refers to Fig. 3, and what the present embodiment described is the handling process of server, institute
The method of stating includes:
Step S301, server receives the input data of client, to described input
Data carry out participle, it is thus achieved that the first word set after participle.
Wherein, the embodiment that input data carry out participle has multiple, the most permissible
Participle based on dictionary, participle based on dictionary is used to need first to preset a participle word
" if, the morning " these words are such as left in dictionary for word segmentation, so by allusion quotation in advance
After described input data are cut into individual character, be combined into word go in dictionary search,
Having mixed and just picked out a word, do not match is just cut into individual character.Permissible
Being understood by, implementation is not limited to mode described above.
It is understood that the input data that server receives can be single word,
Can also be multiple word, it is also possible to be incomplete sentence.
Step S302, described server by quantity of information less than the first predetermined threshold value word from
Described first word set is removed, it is thus achieved that the second word set.
Wherein, the quantity of information of each word such as can calculate according to shannon formula,
Quantity of information after calculating can also be less than the word of the first predetermined threshold value by server in advance
It is placed in a preset table, according to the difference of customer service platform, can be in the server
Placing different preset table, preset table includes that the quantity of information drawn after calculating is little
In the word of the first predetermined threshold value, such as in Taobao's platform, the word such as " close ", " relative ",
For another example in most of scenes, " you ", " I ", " ", the word such as "Yes".It is appreciated that
, deposit the quantity of information mode less than the word of the first predetermined threshold value, except preset table
Outward, the most multiple implementation, it is not limited to described embodiment.
Server is by described through traversal for the first word set got through step S301
Preset table, removes the word protected in preset table from the first word set, just can obtain
Two word sets.
It is understood that quantity of information is less than the first predetermined threshold value by described server
Word removes from described first word set, it is thus achieved that the embodiment of the second word set can have many
Kind, and it is not limited to mode described above.
Step S303, described server, according to described second word set, calculates respectively and prestores
Storage answer described in each language material blocks of knowledge in the corpus previously generated is corresponding
Second weighted value, each language material blocks of knowledge in wherein said corpus includes client
Put question to the answer of correspondence and the first weighted value that described answer is corresponding.
Wherein, described corpus is to build in advance and store in the server.Server
Collect the dialogue of contact staff and client the most in a large number, can in advance contact staff be returned
The answer answered according to the method identical with described in step S301 and step S302,
The second word set is extracted, by corresponding for each answer second after each answer is carried out participle
Word set leaves in the language material blocks of knowledge of its correspondence respectively.It is understood that institute
State the first weighted value that described in each language material blocks of knowledge, answer is corresponding can set flexibly
Put.To being pre-stored in the corpus previously generated described in each language material blocks of knowledge
The first upper initial value of weighted value tax that answer is corresponding, the when of building of corpus, each
First weighted value corresponding to answer can be entered as identical value, but server also may be used
With the demand according to user, described first weighted value is modified.
Server according to input data acquisition the second word set, respectively be pre-stored in pre-
The second word that in the corpus first generated, described in each language material blocks of knowledge, answer is corresponding
Concentration is mated, and such as includes according to the second word set of input data acquisition
(A, B, C), it is assumed that be pre-stored in the first language material knowledge list in the corpus previously generated
Second word set corresponding to answer described in unit includes (A, E, D), the second language material knowledge list
Second word set corresponding to answer described in unit includes (B, C, D), then the first language material knowledge
Second weighted value corresponding to answer described in unit is 1, in the second language material blocks of knowledge
Second weighted value corresponding to described answer is 2.
It is understood that the embodiment calculating described second weighted value does not limit to
In described embodiment.
Step S304, described server is according to language material knowledge list described in described corpus
Described first weighted value that answer described in unit is corresponding and described second weighted value are to institute
State answer to be ranked up, and return recommendation answer according to ranking results to described client.
It is understood that described server is known according to language material described in described corpus
Know described first weighted value and described second weighted value that described in unit, answer is corresponding
The embodiment being ranked up described answer can have multiple, can be by described first
Value after weighted value and described second weighted value are added is ranked up, it is also possible to by it
It is ranked up after being multiplied, it is not limited to described embodiment.
For example, it is assumed that the second weight that described in the first language material blocks of knowledge, answer is corresponding
Value be 1 and first weighted value be 1, after the first weighted value and the second weighted value are added
Obtaining the weight that described in the first language material blocks of knowledge, answer is corresponding is 2, the second language material
Second weighted value corresponding to answer described in blocks of knowledge be 2 and first weight be 1,
In like manner obtaining the weight that described in the second language material blocks of knowledge, answer is corresponding is 3, it is assumed that
Needing the recommendation answer number returned to client is 1, is therefore known by the second language material
Know answer described in unit as recommending answer to return to client.
And, it is understood that recommend the individual of answer to what described client returned
It can also be multiple that number can be 1.
The data processing method of embodiment of the present invention offer and device, by by client
Input data carry out participle and remove quantity of information reduce word after, according to described message
Measure higher word set, from being pre-stored in each language material knowledge the corpus previously generated
By processing acquisition recommendation answer in the described answer of unit so that contact staff only needs
Want importation key word just can generate from the corpus previously generated suitably to answer
Case, and without inputting complete sentence, the manpower of contact staff can be discharged, for visitor
The work taking personnel provides convenient.
Fig. 4 shows a kind of data processing method that another embodiment of the present invention provides
Flow chart, refers to Fig. 4, and what the present embodiment described is the handling process of server,
Described method includes:
Step S401, server receives the input data of client, to described input number
According to carrying out participle, it is thus achieved that the first word set after participle；
Step S402, described server by quantity of information less than the first predetermined threshold value word from
Described first word set is removed, it is thus achieved that the second word set；
Step S403, described server, according to described second word set, calculates respectively and prestores
Storage answer described in each language material blocks of knowledge in the corpus previously generated is corresponding
Second weighted value, each language material blocks of knowledge in wherein said corpus includes client
Put question to the answer of correspondence and the first weighted value that described answer is corresponding；
Step S404, described server is according to language material knowledge list described in described corpus
Described first weighted value that answer described in unit is corresponding and described second weighted value are to institute
State answer to be ranked up, and return recommendation answer according to ranking results to described client.
In the present embodiment, the implementation of described step S401～S404 with respectively with
Implementation in step S301～S304 is identical, does not repeats them here.
Step S405, described service end receives the removal request instruction of described client,
The instruction of described removal request includes that the client questions that the request of described client is deleted is corresponding
Answer；When described server inquires the instruction of described removal request in described corpus
In answer corresponding to described client questions time, by answer corresponding for described client questions
The first corresponding weighted value reduces.
When server is after client returns recommendation answer, and contact staff can be by visitor
Family end returns removal request instruction to server, thinks and does not conforms to deleting contact staff
Suitable answer.But it is understood that, certain contact staff thinks that recommendation answer is also
Time improper, do not represent all contact staff and all think improper, therefore work as service
When device receives the removal request instruction that described client returns, server is at described language material
Storehouse inquires the answer that the described client questions in the instruction of described removal request is corresponding
Time, the first corresponding for answer corresponding for described client questions weighted value is reduced.
Step S406, described service end receives the amendment instruction of described client, described
Amendment instruction include the answer corresponding to client questions of described client request amendment with
And amended answer；Repair described in inquiring in described corpus when described server
When changing answer corresponding to the described client questions in instruction, described in described corpus
The answer that client questions is corresponding replaces with described amended answer, and by described language material
First weighted value corresponding to answer that client questions described in storehouse is corresponding increases.
When server is after client returns recommendation answer, and contact staff can be by visitor
Family end returns amendment instruction to server, so that recommendation answer is modified by server.
It is understood that the credibility of answer after contact staff manual amendment is higher,
Therefore after server receives the amendment instruction that client returns, by institute in described corpus
State answer corresponding to client questions and replace with described amended answer, and by institute's predicate
The first weighted value that described in material storehouse, answer corresponding to client questions is corresponding increases.
Step S407, described service end receives the interpolation instruction of described client, described
Add answer and the request interpolation of described client that instruction includes that client questions is corresponding
Satellite information；Refer to when described server inquires described interpolation in described corpus
The answer that described client questions in order is corresponding, adds described satellite information to correspondence
Described language material blocks of knowledge, and by corresponding for corresponding for described client questions answer
One weighted value increases.
Server can receive the interpolation instruction that contact staff is sent, institute by client
State and add instruction and include that answer that client questions is corresponding and the request of described client add
The satellite information added, wherein said satellite information can include adding key word, add
Classification information etc., it is to be understood that add when certain answer has more user
Satellite information time, its credibility will be higher, therefore receive described interpolation when server
During instruction, the first corresponding for answer corresponding for described client questions weighted value is increased.
Step S408, described service end receives the newly-built instruction of described client, described
Newly-built instruction includes that described client asks the answer that newly-built client questions is corresponding；
Described server creates client questions described in described newly-built instruction in described corpus
Corresponding answer, and by the first corresponding for answer corresponding for described client questions weighted value
It is entered as the 3rd predetermined threshold value.
Server can receive the newly-built instruction that contact staff is sent by client, with
Server is made to add the answer that client questions is corresponding, it is to be understood that contact staff
The answer that the answer actively added is collected the most in a large number relative to server, its credibility
Higher.After the most described service end receives the newly-built instruction of described client, can be right
The first corresponding for answer corresponding for described client questions weighted value is entered as higher power
Weight values.Wherein, the 3rd predetermined threshold value can be arranged the most flexibly.
Step S409, described service end receives the second input data of described client,
Judge that described second input data are identical with described recommendation answer, when described second defeated
Entering data and described recommendation answer when differing, described server will described second input
Data are added in described corpus.
When contact staff receives, by client, the recommendation answer that server sends, visitor
The personnel that take do not select described recommendation answer by client, but pass through client
Have sent the second input data, when server receives described second input data, will
Described second input data are added in corpus as new language material blocks of knowledge, logical
Cross server collection language material blocks of knowledge the most actively so that corpus is increasingly
Abundant, further such that the recommendation answer that server returns to client according to corpus
More accurate.
The data processing method that the embodiment of the present invention provides, is possible not only to solve upper one real
Execute the soluble problem of example, also by the mode of user feedback to server structure in advance
Language material blocks of knowledge in the corpus built is optimized so that server is according to language material
The recommendation answer that storehouse returns to client is more accurate.
Refer to Fig. 5, be that the data shown in Fig. 2 that the embodiment of the present invention provides process
The high-level schematic functional block diagram of device.Described data processing equipment 210 includes word-dividing mode
211, the first processing module 212, computing module 213, order module 214.
Described word-dividing mode 211, receives the input data of client for server,
Described input data are carried out participle, it is thus achieved that the first word set after participle；
Described first processing module 212, for described server by quantity of information less than the
The word of one predetermined threshold value removes from described first word set, it is thus achieved that the second word set；
Described computing module 213, for described server according to described second word set,
Calculate respectively and be pre-stored in the corpus previously generated institute in each language material blocks of knowledge
State the second weighted value that answer is corresponding, each language material knowledge in wherein said corpus
Unit includes answer that client questions is corresponding and the first weight corresponding to described answer
Value；
Described order module 214, for described server according to institute in described corpus
Described first weighted value and described that answer described in predicate material blocks of knowledge is corresponding
Described answer is ranked up by two weighted values, and according to ranking results to described client
Return and recommend answer.
The most each module can be by software code realization, now, and above-mentioned each module
Can be stored in the memorizer of server.The most each module is equally by hardware such as
IC chip realizes.
Refer at a kind of data that Fig. 6, Fig. 6 are another embodiment of the present invention offers
The high-level schematic functional block diagram of reason device, described data processing equipment 310 includes participle mould
Block the 211, first processing module 212, computing module 213, order module 214, the
Two processing modules 315, the 3rd processing module 316, fourth processing module the 317, the 5th
Processing module 318 and the 6th processing module 319.
Described second processing module 315, for described order module 214 at described clothes
Business device returns to described client to described client to be recommended after answer, described service
End receives the removal request instruction of described client, and the instruction of described removal request includes institute
State the answer that the client questions of client request deletion is corresponding；When described server is in institute
State the described client questions inquired in corpus in the instruction of described removal request corresponding
During answer, the first corresponding for answer corresponding for described client questions weighted value is reduced.
Described 3rd processing module 316, for described order module 214 at described clothes
Business device returns to described client to described client to be recommended after answer, described service
End receives the amendment instruction of described client, and described amendment instruction includes described client
Hold answer and amended answer that the client questions asking amendment is corresponding；When described
Server inquires the described client questions in described amendment instruction in described corpus
During corresponding answer, answer corresponding for client questions described in described corpus is replaced
For described amended answer, and by corresponding for client questions described in described corpus
The first weighted value increase that answer is corresponding.
Described fourth processing module 317, for described order module 214 at described clothes
Business device returns to described client to described client to be recommended after answer, described service
End receives the interpolation instruction of described client, and described interpolation instruction includes client questions
The satellite information that corresponding answer and the request of described client are added；When described service
It is corresponding that device inquires the described described client questions added in instruction in described corpus
Answer, described satellite information is added to correspondence described language material blocks of knowledge, and
The first corresponding for answer corresponding for described client questions weighted value is increased.
Described 5th processing module 318, is in institute for described order module 214
State server to described client after described client returns recommendation answer, described
Service end receives the newly-built instruction of described client, and described newly-built instruction includes described
Client asks the answer that newly-built client questions is corresponding；Described server is at institute's predicate
Material creates, in storehouse, the answer that described in described newly-built instruction, client questions is corresponding, and by institute
State the first weighted value corresponding to answer corresponding to client questions to be entered as the 3rd and preset threshold
Value.
Described 6th processing module 319, for being for described order module 214
Return to described client to described client at described server and recommend after answer,
Described service end receives the second input data of described client, it is judged that described second defeated
It is identical with described recommendation answer for entering data, when described second input data push away with described
Recommending answer when differing, described second input data are added to described by described server
In corpus.
The language material that server is built in advance by described device also by the mode of user feedback
Language material blocks of knowledge in storehouse is optimized so that server according to corpus to client
The recommendation answer that end returns is more accurate.
The most each module can be by software code realization, now, and above-mentioned each module
Can be stored in the memorizer of server.The most each module is equally by hardware such as
IC chip realizes.
It should be noted that each embodiment in this specification all uses the side gone forward one by one
Formula describes, and what each embodiment stressed is the difference with other embodiments,
Between each embodiment, identical similar part sees mutually.
The data processing equipment that the embodiment of the present invention is provided, it realizes principle and generation
Technique effect identical with preceding method embodiment, for briefly describe, Apparatus and system
The not mentioned part of embodiment part, refers to corresponding contents in preceding method embodiment.
In several embodiments provided herein, it should be understood that disclosed
Apparatus and method, it is also possible to realize by another way.Device described above
Embodiment is only that schematically such as, flow chart and block diagram in accompanying drawing show
The device of multiple embodiments according to the present invention, method and computer program product can
Architectural framework, function and the operation that can realize.In this, in flow chart or block diagram
Each square frame can represent a module, program segment or a part for code, described
A part for module, program segment or code comprises one or more for realizing regulation
The executable instruction of logic function.It should also be noted that at some as the realization replaced
In mode, the function that marked in square frame can also be marked in accompanying drawing to be different from
Order occurs.Such as, two continuous print square frames can essentially perform substantially in parallel,
They can also perform sometimes in the opposite order, and this is depending on involved function.
It is also noted that each square frame in block diagram and/or flow chart and block diagram and/
Or the combination of the square frame in flow chart, can be with performing the function of regulation or the special of action
Hardware based system realize, or can refer to computer with specialized hardware
The combination of order realizes.
It addition, each functional module in each embodiment of the present invention can be integrated in one
Rise and form an independent part, it is also possible to be modules individualism, it is also possible to
Two or more modules are integrated to form an independent part.
If described function realizes and as independent product using the form of software function module
When product are sold or use, can be stored in a computer read/write memory medium.
Based on such understanding, technical scheme is the most in other words to prior art
The part of the part contributed or this technical scheme can be with the form of software product
Embodying, this computer software product is stored in a storage medium, if including
Dry instruction with so that computer equipment (can be personal computer, server,
Or the network equipment etc.) perform the whole or portion of method described in each embodiment of the present invention
Step by step.And aforesaid storage medium includes: USB flash disk, portable hard drive, read-only storage
Device (ROM, Read-Only Memory), random access memory (RAM, Random
Access Memory), various Jie that can store program code such as magnetic disc or CD
Matter.It should be noted that in this article, the relation of such as first and second or the like
Term is used merely to separate an entity or operation with another entity or operating space
Come, and exist any this between not necessarily requiring or imply these entities or operating
Actual relation or order.And, term " includes ", " comprising " or it is any
Other variants are intended to comprising of nonexcludability, so that include a series of key element
Process, method, article or equipment not only include those key elements, but also include
Other key elements being not expressly set out, or also include for this process, method,
Article or the intrinsic key element of equipment.In the case of there is no more restriction, by language
The key element that sentence " including ... " limits, it is not excluded that in the mistake including described key element
Journey, method, article or equipment there is also other identical element.
The foregoing is only the preferred embodiments of the present invention, be not limited to this
Invention, for a person skilled in the art, the present invention can have various change and
Change.All within the spirit and principles in the present invention, any amendment of being made, equivalent
Replacement, improvement etc., should be included within the scope of the present invention.It should also be noted that
Similar label and letter in following accompanying drawing represent similar terms, therefore, once certain
One is defined in an accompanying drawing, then need not carry out it in accompanying drawing subsequently
Definition and explanation further.
The above, the only detailed description of the invention of the present invention, but the protection of the present invention
Scope is not limited thereto, and any those familiar with the art is in the present invention
In the technical scope disclosed, change can be readily occurred in or replace, all should contain at this
Within bright protection domain.Therefore, protection scope of the present invention should described be wanted with right
The protection domain asked is as the criterion.
Claims (12)
1. a data processing method, is applied to customer service field, it is characterised in that institute
The method of stating includes:
Server receives the input data of client, and described input data are carried out participle,
Obtain the first word set after participle；
Quantity of information is less than the word of the first predetermined threshold value from described first word by described server
Concentration is removed, it is thus achieved that the second word set；
Described server, according to described second word set, calculates respectively and is pre-stored in pre-Mr.
The second weight that in the corpus become, described in each language material blocks of knowledge, answer is corresponding
Value, each language material blocks of knowledge in wherein said corpus includes that client questions is corresponding
Answer and the first weighted value corresponding to described answer；
Described server is answered according to language material blocks of knowledge described in described corpus
Described answer is carried out by described first weighted value and described second weighted value that case is corresponding
Sequence, and return recommendation answer according to ranking results to described client.
Method the most according to claim 1, it is characterised in that in described service
Device is to described client after described client returns recommendation answer, and described method is also
Including:
Described service end receives the removal request instruction of described client, and described deletion please
Instruction is asked to include answer corresponding to client questions that the request of described client is deleted；
In described server inquires the instruction of described removal request in described corpus
Answer corresponding to described client questions time, by answer pair corresponding for described client questions
The first weighted value answered reduces.
Method the most according to claim 1, it is characterised in that in described service
Device is to described client after described client returns recommendation answer, and described method is also
Including:
Described service end receives the amendment instruction of described client, in described amendment instruction
Including the answer corresponding to client questions of described client request amendment and amended
Answer；
Institute in described server inquires described amendment instruction in described corpus
When stating answer corresponding to client questions, client questions described in described corpus is corresponding
Answer replace with described amended answer, and by client described in described corpus
The first weighted value puing question to corresponding answer corresponding increases.
Method the most according to claim 1, it is characterised in that in described service
Device is to described client after described client returns recommendation answer, and described method is also
Including:
Described service end receives the interpolation instruction of described client, in described interpolation instruction
The attached letter that the answer corresponding including client questions and the request of described client are added
Breath；
When described server inquires the described institute added in instruction in described corpus
State the answer that client questions is corresponding, described satellite information is added to institute's predicate of correspondence
Material blocks of knowledge, and by the first corresponding for answer corresponding for described client questions weighted value
Increase.
Method the most according to claim 1, it is characterised in that in described service
Device is to described client after described client returns recommendation answer, and described method is also
Including:
Described service end receives the newly-built instruction of described client, in described newly-built instruction
Including the answer that the client questions that the request of described client is newly-built is corresponding；
Described server creates client described in described newly-built instruction in described corpus
Put question to corresponding answer, and by the first corresponding for answer corresponding for described client questions power
Weight values is entered as the 3rd predetermined threshold value.
Method the most according to claim 1, it is characterised in that in described service
Device is to described client after described client returns recommendation answer, and described method is also
Including:
Described service end receives the second input data of described client, it is judged that described the
Two input data are identical with described recommendation answer, when described second inputs data and institute
Stating recommendation answer when differing, described second input data are added to by described server
In described corpus.
7. a data processing equipment, is applied to customer service field, it is characterised in that institute
State device to include:
Word-dividing mode, receives the input data of client, to described defeated for server
Enter data and carry out participle, it is thus achieved that the first word set after participle；
First processing module, presets threshold by quantity of information less than first for described server
The word of value removes from described first word set, it is thus achieved that the second word set；
Computing module, for described server according to described second word set, calculates respectively
It is pre-stored in answer pair described in each language material blocks of knowledge in the corpus previously generated
The second weighted value answered, each language material blocks of knowledge in wherein said corpus includes
Answer that client questions is corresponding and the first weighted value corresponding to described answer；
Order module, knows according to language material described in described corpus for described server
Know described first weighted value and described second weighted value that described in unit, answer is corresponding
Described answer is ranked up, and returns recommendation according to ranking results to described client
Answer.
Device the most according to claim 7, it is characterised in that described device is also
Including the second processing module,
Described second processing module, for described order module at described server to institute
Stating client to return to described client and recommend after answer, described service end receives institute
Stating the removal request instruction of client, the instruction of described removal request includes described client
Ask the answer that the client questions of deletion is corresponding；When described server is at described corpus
In when inquiring answer corresponding to described client questions in the instruction of described removal request,
The first corresponding for answer corresponding for described client questions weighted value is reduced.
Device the most according to claim 7, it is characterised in that described device is also
Including the 3rd processing module,
Described 3rd processing module, for described order module at described server to institute
Stating client to return to described client and recommend after answer, described service end receives institute
Stating the amendment instruction of client, described amendment instruction includes that the request of described client is repaiied
Answer that the client questions that changes is corresponding and amended answer；When described server exists
Described corpus inquires corresponding the answering of described client questions in described amendment instruction
During case, answer corresponding for client questions described in described corpus is replaced with described in repair
Answer after changing, and answer corresponding for client questions described in described corpus is corresponding
First weighted value increase.
Device the most according to claim 7, it is characterised in that described device
Also include fourth processing module,
Described fourth processing module, for described order module at described server to institute
Stating client to return to described client and recommend after answer, described service end receives institute
Stating the interpolation instruction of client, described interpolation instruction includes corresponding the answering of client questions
The satellite information that case and the request of described client are added；When described server is described
Corpus inquires the answer that the described described client questions added in instruction is corresponding,
Described satellite information is added to the described language material blocks of knowledge of correspondence, and by described visitor
The first weighted value that family puts question to corresponding answer corresponding increases.
11. devices according to claim 7, it is characterised in that described device
Also include the 5th processing module,
Described 5th processing module, is at described server for described order module
To described client after described client returns recommendation answer, described service termination
Receiving the newly-built instruction of described client, described newly-built instruction includes that described client please
The answer that the client questions built of looking for novelty is corresponding；Described server is created in described corpus
Build the answer that described in described newly-built instruction, client questions is corresponding, and described client is carried
Ask that first weighted value corresponding to answer of correspondence is entered as the 3rd predetermined threshold value.
12. devices according to claim 7, it is characterised in that described device
Also include the 6th processing module,
Described 6th processing module, is at described server for described order module
To described client after described client returns recommendation answer, described service termination
Receive the second input data of described client, it is judged that described second input data are and institute
State recommendation answer identical, when described second input data differ with described recommendation answer
Time, described second input data are added in described corpus by described server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610200992.1A CN105893552B (en) | 2016-03-31 | 2016-03-31 | Data processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610200992.1A CN105893552B (en) | 2016-03-31 | 2016-03-31 | Data processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105893552A true CN105893552A (en) | 2016-08-24 |
| CN105893552B CN105893552B (en) | 2020-05-05 |
Family
ID=57013193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610200992.1A Active CN105893552B (en) | 2016-03-31 | 2016-03-31 | Data processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105893552B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110880077A (en) * | 2019-11-25 | 2020-03-13 | 重庆五生信息技术有限公司 | Enterprise intelligent consultation cloud platform |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050086045A1 (en) * | 2003-10-17 | 2005-04-21 | National Institute Of Information And Communications Technology | Question answering system and question answering processing method |
| US20120078891A1 (en) * | 2010-09-28 | 2012-03-29 | International Business Machines Corporation | Providing answers to questions using multiple models to score candidate answers |
| CN103493045A (en) * | 2011-01-18 | 2014-01-01 | 谷歌公司 | Automated answers to online questions |
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| US20150235130A1 (en) * | 2014-02-19 | 2015-08-20 | International Business Machines Corporation | NLP Duration and Duration Range Comparison Methodology Using Similarity Weighting |
| US20160078341A1 (en) * | 2014-09-16 | 2016-03-17 | International Business Machines Corporation | Building a Domain Knowledge and Term Identity Using Crowd Sourcing |
| CN106844400A (en) * | 2015-12-07 | 2017-06-13 | 南京中兴新软件有限责任公司 | Intelligent response method and device |
-
2016
- 2016-03-31 CN CN201610200992.1A patent/CN105893552B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050086045A1 (en) * | 2003-10-17 | 2005-04-21 | National Institute Of Information And Communications Technology | Question answering system and question answering processing method |
| US20120078891A1 (en) * | 2010-09-28 | 2012-03-29 | International Business Machines Corporation | Providing answers to questions using multiple models to score candidate answers |
| CN103493045A (en) * | 2011-01-18 | 2014-01-01 | 谷歌公司 | Automated answers to online questions |
| CN104598445A (en) * | 2013-11-01 | 2015-05-06 | 腾讯科技（深圳）有限公司 | Automatic question-answering system and method |
| US20150235130A1 (en) * | 2014-02-19 | 2015-08-20 | International Business Machines Corporation | NLP Duration and Duration Range Comparison Methodology Using Similarity Weighting |
| CN103902652A (en) * | 2014-02-27 | 2014-07-02 | 深圳市智搜信息技术有限公司 | Automatic question-answering system |
| US20160078341A1 (en) * | 2014-09-16 | 2016-03-17 | International Business Machines Corporation | Building a Domain Knowledge and Term Identity Using Crowd Sourcing |
| CN106844400A (en) * | 2015-12-07 | 2017-06-13 | 南京中兴新软件有限责任公司 | Intelligent response method and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110880077A (en) * | 2019-11-25 | 2020-03-13 | 重庆五生信息技术有限公司 | Enterprise intelligent consultation cloud platform |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105893552B (en) | 2020-05-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10452702B2 (en) | Data clustering | |
| CN107491547A (en) | Searching method and device based on artificial intelligence | |
| US11514498B2 (en) | System and method for intelligent guided shopping | |
| US20200175381A1 (en) | Candidate visualization techniques for use with genetic algorithms | |
| CN109766949A (en) | Convolutional neural networks light weight method, device and electronic equipment | |
| CN110413867B (en) | Method and system for content recommendation | |
| EP3671526A1 (en) | Dependency graph based natural language processing | |
| CN105894183A (en) | Project evaluation method and apparatus | |
| CN107807935B (en) | Using recommended method and device | |
| CN108898413A (en) | A kind of traveller's management method and device | |
| CN110069629A (en) | House transaction task processing method, equipment, storage medium and device | |
| CN109947983A (en) | Video recommendation method, system, terminal and computer readable storage medium | |
| Kuhrmann et al. | A mapping study on the feasibility of method engineering | |
| CN108182200B (en) | Keyword expansion method and device based on semantic similarity | |
| CN110378543A (en) | Leaving office Risk Forecast Method, device, computer equipment and storage medium | |
| CN105812473A (en) | Data processing method and device | |
| Ye et al. | Big data processing framework for manufacturing | |
| Keshavarz et al. | Efficient upper and lower bounding methods for flowshop sequence-dependent group scheduling problems | |
| CN109492149A (en) | Crawler task processing method and device | |
| CN117787290A (en) | Drawing prompting method and device based on knowledge graph | |
| CN105893552A (en) | Method and device for processing data | |
| CN110851343A (en) | Test method and device based on decision tree | |
| CN108763260A (en) | Test question searching method and system and terminal equipment | |
| Brdjanin et al. | Towards the automated business model-driven conceptual database design | |
| CN117370650A (en) | Cloud computing data recommendation method based on service combination hypergraph convolutional network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Chang Inventor after: Qu Feng Inventor after: Ying Hailiang Inventor after: Zhao Xuemin Inventor after: Jiang Ling Inventor before: Chen Chang Inventor before: Qu Feng Inventor before: Ying Hailiang Inventor before: Zhao Xuemin Inventor before: Jiang Ling |
|
| CB03 | Change of inventor or designer information | ||
| CB02 | Change of applicant information |
Address after: 610000 Huayang Street, Tianfu New District, Chengdu City, Sichuan Province, No. 1, No. 2, No. 19 Building, Unit 2, 1903 Applicant after: Chengdu Xiaoduo Technology Co., Ltd. Address before: 610000 846, southern section of Tianfu Avenue, Huayang street, Tianfu New District, Chengdu, Sichuan Applicant before: CHENGDU XIAODUO TECH CO., LTD. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |