CN110210721A - A kind of long-range sign language translation on line customer service distribution method and device - Google Patents
A kind of long-range sign language translation on line customer service distribution method and device Download PDFInfo
- Publication number
- CN110210721A CN110210721A CN201910397814.6A CN201910397814A CN110210721A CN 110210721 A CN110210721 A CN 110210721A CN 201910397814 A CN201910397814 A CN 201910397814A CN 110210721 A CN110210721 A CN 110210721A
- Authority
- CN
- China
- Prior art keywords
- customer service
- sign language
- terminal
- translation
- pond
- 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
- 238000013519 translation Methods 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000004590 computer program Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 229910017435 S2 In Inorganic materials 0.000 claims 1
- 238000005457 optimization Methods 0.000 abstract description 6
- 230000004927 fusion Effects 0.000 abstract description 5
- 230000000694 effects Effects 0.000 abstract description 2
- 230000006854 communication Effects 0.000 description 11
- 238000004891 communication Methods 0.000 description 10
- 230000006872 improvement Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 3
- XCWPUUGSGHNIDZ-UHFFFAOYSA-N Oxypertine Chemical compound C1=2C=C(OC)C(OC)=CC=2NC(C)=C1CCN(CC1)CCN1C1=CC=CC=C1 XCWPUUGSGHNIDZ-UHFFFAOYSA-N 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Machine Translation (AREA)
- Telephonic Communication Services (AREA)
Abstract
The present invention discloses a kind of long-range sign language translation on line customer service distribution method and device, this method step includes: service scenarios type, the solicited message in geographical location that S1. receives the translation request signal that user terminal is sent, and translates needed for including according to the translation request signal acquisition received;S2. the solicited message that will acquire is matched with the customer service information of each sign language customer service terminal respectively, sign language interpreter ability information when customer service information includes geographical location, different service scenarios types, finds out matched target sign language customer service terminal and distributes to user terminal to carry out sign language translation on line.The present invention can take into account regional, scene type and translation ability, and fusion multi information realizes sign language customer service intelligence, optimization distribution, has many advantages, such as that implementation method is simple, allocative efficiency is high and effect is good.
Description
Technical field
The present invention relates to long-range sign language translation on line technical fields more particularly to a kind of long-range sign language interpreter customer service intelligently to divide
Method of completing the square and device.
Background technique
Long-range sign language translation on line is by user by mobile device calling sign language interpreter customer service, by sign language interpreter center or
Mechanism distributes sign language interpreter customer service and carries out real time translation, may be implemented to translate across the real-time hand language of spatial position, compared to intelligence
The interpretative system of sign language interpreter equipment etc., long-range sign language translation on line is more convenient, efficient and cost is more cheap.
It is at present usually directly random by sign language interpreter center or mechanism for the customer service distribution of long-range sign language translation on line
The customer service of current idle is chosen, if the customer service chosen cannot receive current translation duties after selecting rejection, then chooses it again
User terminal and the customer service terminal for receiving task are established audio-video until there is customer service to receive translation duties by the customer service of his free time
Connection.And sign language has region, same to express, sign language play in various regions is different, and the sign language of different geographical differs greatly, in reality
In the communication process of border, spoken dialect is also more serious, and the low efficiency of the above-mentioned customer service method of salary distribution a, subtask possibly even needs
Repeated dispensing process is undergone, translation response delay length is caused, and the customer service distributed probably belongs to different geographical, so that translation
Process is not smooth, there are problems that sign language geographic difference.
To solve the above problems, thering is practitioner to propose the region for dividing customer service according to customer service IP address, according to customer service
IP address selects the customer service of distribution, allows to preferentially distribute the customer service of same area, but whether sign language interpreter is smooth, matter
Amount height additionally depends on many factors such as matching degree, the customer service sign language interpreter level of customer service, as sign language interpreter has a variety of clothes
Business scene type, such as daily communication, medical interrogation, bank finance, kith and kin's communication type, different scenes type need customer service
Personnel must have the translation background knowledge of corresponding professional domain, be likely to provide smooth, high quality translation, and different customer services
The type be good at, being bad at is different, if such as using it is above-mentioned it is single choose customer service according to the scheme of IP address, can not
Corresponding customer service is chosen with different scenes type, it is likely that translation duties are distributed to and are bad at the customer service of the scene type,
Influence translation quality, and if customer service is chosen based on foundation different scenes type, and cannot take into account customer service region problem, it is difficult to
The optimal customer service that Rapid matching meets to region, translation ability etc..
Summary of the invention
The technical problem to be solved in the present invention is that, for technical problem of the existing technology, the present invention provides one
The long-range sign language interpreter customer service smart allocation method and device that kind implementation method is simple, allocative efficiency is high and effect is good, Neng Goujian
Regional, scene type and translation ability are cared for, fusion multi information realizes sign language customer service intelligence, optimization distribution.
In order to solve the above technical problems, technical solution proposed by the present invention are as follows:
A kind of long-range sign language translation on line customer service distribution method, step include:
S1. the translation request signal that user terminal is sent is received, and according to the translation request signal acquisition received
Service scenarios type, the solicited message in geographical location including required translation;
S2. the solicited message that will acquire is matched with the customer service information of each sign language customer service terminal respectively, the visitor
Sign language interpreter ability information when information includes geographical location, different service scenarios types is taken, matched target sign language is found out
Customer service terminal simultaneously distributes to user terminal to carry out sign language translation on line.
As a further improvement of the present invention, in the step S2, first according to service scenarios type in the solicited message
Filter out each sign language customer service terminal met, then the solicited message that will acquire respectively with each sign language customer service terminal for filtering out
The customer service information matched, find out matched target sign language customer service terminal.
As a further improvement of the present invention, the step of step S2 includes:
S21. customer service pond constructs: multiple customer service ponds of the corresponding different service scenarios types of building in advance, each customer service
The letter that the sign language translation ability under corresponding with service scene type meets multiple sign language customer service terminals of necessary requirement is stored in pond
Breath;
S22. customer service pond is screened: filtering out the matched customer service pond according to service scenarios type in the solicited message;
S23. customer service matches: by the solicited message respectively with each sign language customer service terminal in the customer service pond that filters out into
Row matching, finds out matched target sign language customer service terminal distribution to user terminal.
As a further improvement of the present invention, when constructing customer service pond in the step S21, all sign language customer service terminals are pressed
The sign language interpreter ability impinged upon under corresponding with service scene type is ranked up, and chooses specified quantity/or sign language interpreter energy after sequence
The multiple sign language customer service terminals of force value within a preset range form customer service pond.
As a further improvement of the present invention, preferential to choose and service scenarios in the solicited message in the step S22
The identical customer service pond of type, if choose the customer service pond in not available free customer service, according to different service scenarios types it
Between relevance values choose other customer service ponds.
As a further improvement of the present invention, the step of step S23 includes: to calculate separately the solicited message and institute
State geographical location matching value in customer service pond between each sign language customer service terminal, sign language ability matching value, the call history factor and
It is one or more in customer service quality factor, and final matching value is obtained after being weighted, the size according to the final matching value
Determine matched target sign language customer service terminal, the geographical location matching value is according to geographical location and sign language in the solicited message
Matching degree between the geographical location of customer service terminal obtains, and the sign language ability matching value is according to the sign language of sign language customer service terminal
Standardization level and/or sign language hierarchical level obtain, and the call history factor is exhaled twice according to sign language customer service terminal is adjacent
The interval time cried obtains, and the customer service quality factor obtains the scoring of sign language customer service terminal according to user.
It as a further improvement of the present invention, further include every time updating step after the completion of service, the update step includes:
The scoring for obtaining user and sign language customer service terminal is scored using the user got and is updated described in corresponding sign language customer service terminal
Customer service information and each customer service pond, and described in updating when being weighted each weighted factor weight;Each weighted factor
Weight updated value obtained with specific reference to the final matching value determination that the user of acquisition scores and is calculated.
As a further improvement of the present invention, if the customer service pond currently selected for destination service field in the solicited message
Customer service pond is closed on corresponding to the relevant service scenarios type of scape type, uses sign language customer service terminal after the completion of each service
Scoring updates relevance values between destination service scene type and the related service scene type of selection.
As a further improvement of the present invention, the step of step S2 are as follows: calculate separately the solicited message and each hand
In geographical location matching value, sign language ability matching value, the call history factor and customer service quality factor between language customer service terminal
It is one or more, and final matching value is obtained after being weighted, the size according to the final matching value determines matched target
Sign language customer service terminal, the geographical location matching value is according to the geography in geographical location in the solicited message and sign language customer service terminal
Matching degree between position obtains, and the sign language ability matching value is whole according to the sign language standardization level of user and sign language customer service
The sign language hierarchical level at end obtains, and the call history factor is according to sign language customer service terminal adjacent called interval time twice
It obtains, the customer service quality factor obtains the scoring of sign language customer service terminal according to user.
A kind of long-range sign language translation on line customer service distributor, the computer-readable storage including being stored with computer program
Medium, which is characterized in that such as above-mentioned method is realized when the computer program executes.
Compared with the prior art, the advantages of the present invention are as follows:
1, the long-range sign language translation on line customer service distribution method of the present invention and device, by receiving turning over for user terminal transmission
When translating request signal, the solicited messages such as service scenarios type, geographical location in acquisition request signal, by the solicited message and each visitor
A variety of customer service information such as sign language interpreter ability when taking geographical location, the difference service scenarios type of terminal carry out comprehensive matchings,
Regional, scene type and translation ability, the multi informations such as fusion region, service scenarios type and translation ability can be taken into account
It realizes sign language customer service intelligence, optimization distribution, can effectively improve reasonable, the validity of customer service distribution, improve translation quality, together
When ensure allocative efficiency, reduce translation on line operating lag.
2, the long-range sign language translation on line customer service distribution method of the present invention and device, further by first according to service scenarios class
Type is initially screened, and quickly orients the range of lookup, then according to the information such as geographical location, translation ability in solicited message into
The accurate matching of row, can effectively shorten match time, while ensure matched performance, allow to quickly find and meet the requirements
Optimal sign language customer service terminal, improve service quality.
3, the long-range sign language translation on line customer service distribution method of the present invention and device, further by according to different service scenarios
Type constructs corresponding customer service pond, and the satisfactory sign language customer service terminal of translation ability under different service scenarios types is drawn in advance
Divide into customer service pond, after the service scenarios Sort positioning to corresponding customer service pond according to request, is further screened from customer service pond
The sign language customer service terminal met out can be further improved the efficiency of customer service distribution, can quickly distribute as far as possible and meet need
The best sign language customer service terminal asked is to user terminal.
4, the long-range sign language translation on line customer service distribution method of the present invention and device, further when determining matched range,
It is preferential to choose the identical customer service pond of service scenarios type, if without idle customer service, then according to the correlation between service scenarios type
Property choose other customer service ponds, can to avoid common type call frequently or calling peak period caused by specific type sign language customer service not
Sufficient problem, while can distribute translation background correlation big customer service as far as possible.
5, the long-range sign language translation on line customer service distribution method of the present invention and device, further by fully considering geographical position
It sets, the factors such as sign language is horizontal, history call and quality, fusion geographical location, sign language be horizontal, history call and quality etc. are more
Information can obtain satisfactory best sign language customer service terminal distribution to user terminal, realize optimal sign language customer service distribution.
6, the long-range sign language translation on line customer service distribution method of the present invention and device use use after the completion of further service every time
The scoring of family and sign language customer service terminal updates user, the customer service information of sign language customer service terminal, each customer service pond, and respectively adds when weighting
The weight of weight factor can carry out continuous renewal optimization to distribution by each service state, and if the customer service pond that selects for
Related service scene type closes on customer service pond, updates correlation between scene type using the scoring of sign language customer service terminal, can
With the correlation between the feedback updated different scenes type by real time service, to continue to optimize allocation performance.
Detailed description of the invention
Fig. 1 is the implementation process schematic diagram of the long-range sign language translation on line customer service distribution method of the present embodiment.
Fig. 2 is the implementation process schematic diagram of concrete application embodiment medium-long range sign language translation on line customer service distribution of the present invention.
Specific embodiment
Below in conjunction with Figure of description and specific preferred embodiment, the invention will be further described, but not therefore and
It limits the scope of the invention.
Embodiment 1:
As shown in Figure 1, the long-range sign language translation on line customer service distribution method of the present embodiment, step include:
S1. the translation request signal that user terminal is sent is received, and includes according to the translation request signal acquisition received
Service scenarios type, the solicited message in geographical location of required translation;
S2. the solicited message that will acquire is matched with the customer service information of each sign language customer service terminal respectively, customer service packet
Sign language interpreter ability information when geographical location, different service scenarios types is included, matched target sign language customer service terminal is found out
And user terminal is distributed to carry out sign language translation on line.
When translation request signal of the present embodiment by receiving user terminal transmission, service scenarios in acquisition request signal
The solicited messages such as type, geographical location, when by the solicited message from the geographical location of each customer service terminal, different service scenarios types
A variety of customer service information such as sign language interpreter ability carry out comprehensive matchings, regional, scene type and translation ability can be taken into account,
The multi informations such as fusion region, service scenarios type and translation ability realize sign language customer service intelligence, optimization distribution, can be effective
Reasonable, the validity of customer service distribution are improved, improves translation quality, while ensuring allocative efficiency, reduces translation on line operating lag.
In the present embodiment step S2, it is whole that each sign language customer service met is first filtered out according to service scenarios type in solicited message
End, then the solicited message that will acquire matched with the customer service information of each sign language customer service terminal filtered out respectively, finds out
The target sign language customer service terminal matched.The present embodiment is quickly oriented and is looked by first initially being screened according to service scenarios type
The range looked for, then accurately matched according to information such as geographical location, translation abilities in solicited message, can effectively shorten matching
Time, while ensuring matched performance, allow to quickly find satisfactory optimal sign language customer service terminal, improves service
Quality.
In the present embodiment, the step of step S2, includes:
S21. customer service pond constructs: multiple customer service ponds of the corresponding different service scenarios types of building in advance, in each customer service pond
It is stored with the information that the sign language translation ability under corresponding with service scene type meets multiple sign language customer service terminals of necessary requirement;
S22. customer service pond is screened: filtering out matched customer service pond according to service scenarios type in solicited message;
S23. customer service matches: solicited message is matched with each sign language customer service terminal in the customer service pond filtered out respectively,
Matched target sign language customer service terminal distribution is found out to user terminal.
The present embodiment is by constructing corresponding customer service pond according to different service scenarios types, under different service scenarios types
The satisfactory sign language customer service terminal of translation ability is divided in advance in customer service pond, receives the request signal of user terminal transmission
Afterwards, after according to the service scenarios Sort positioning to corresponding customer service pond of request, further screening meets from customer service pond hand
Language customer service terminal can be further improved the efficiency of customer service distribution, can quickly distribute the sign language of meet demand as far as possible
Customer service terminal is to user terminal.
When constructing customer service pond in the present embodiment, in step S21, by all sign language customer service terminals according in corresponding with service scene
Sign language interpreter ability under type is ranked up, and specified quantity or sign language interpreter ability value be within a preset range after choosing sequence
Multiple sign language customer service terminals form customer service pond, when such as constructing the customer service pond of corresponding daily communication scene type, first by all hands
Language customer service terminal is ranked up according to the sign language interpreter ability value of daily communication scene, and n sign language customer services are whole before taking after sequence
End forms customer service pond, or sign language translation ability value size after sequence is taken to be greater than the sign language customer service terminal shape for presetting ability value threshold value
At customer service pond, above-mentioned n, ability value threshold value can specifically be set according to actual needs.
It is preferential to choose customer service pond identical with service scenarios type in solicited message in step S22 in the present embodiment, if
Not available free customer service in the customer service pond of selection chooses other customer service ponds according to the correlation between different service scenarios types.No
With between service scenarios type there is certain correlation to be good at one type type if the correlation between two types is larger
Customer service generally also can satisfy another type of translation demand, such as the correlation between daily communication type and kith and kin's communication type
It is larger, in the case where the corresponding customer service pond not available free customer service of daily communication type, chooses kith and kin and link up the corresponding visitor of type
It takes pond generally also and can satisfy translation demand.The call number of certain service scenarios type may be more compared to other types, or
Peak period is called at some, may exist does not have idle sign language customer service terminal in customer service pond in some service scenarios type,
The present embodiment is preferential to choose the identical customer service pond of service scenarios type when determining matched range, if without idle customer service, then
Other customer service ponds are chosen according to the correlation between service scenarios type, can call frequently or call peak to avoid common type
Specific type sign language customer service deficiency problem caused by phase, while can distribute translation background correlation big customer service as far as possible.
The present embodiment specifically first calculates the relevance values R between different service scenarios types, after receiving user's request,
Without idle customer service in the corresponding customer service pond of service scenarios type of user's request, then choose and the service scenarios type correlation
Be worth it is maximum close on customer service pond, if closing on the still not available free customer service in customer service pond, then successively choose it according to relevance values size
He closes on customer service pond, until finding the customer service pond of available free customer service.
In the present embodiment, the step of step S23 includes: to calculate separately each sign language customer service terminal in solicited message and customer service pond
Between geographical location matching value, sign language ability matching value, the call history factor and customer service quality factor etc., and be weighted
After obtain final matching value, the size according to final matching value determines matched target sign language customer service terminal, geographical location matching
Value is obtained according to the matching degree between geographical location in solicited message and the geographical location of sign language customer service terminal, to characterize geography
Location matches degree, sign language ability matching value is according to the sign language standardization level of user and the sign language level etc. of sign language customer service terminal
Grade obtains, to characterize sign language ability matching degree, the call history factor adjacent called twice according to sign language customer service terminal
It is obtained every the time, to characterize history call influence degree, customer service quality factor scoring to sign language customer service terminal according to user
It arrives, to characterize customer service quality, can fully consider the factors such as geographical location, sign language level, history call and quality, match
Satisfactory best sign language customer service terminal distribution is to user terminal.
In concrete application embodiment, each matching value, factor values are calculated first:
Geographical location matching value VlocIt can specifically determine as the following formula:
I.e. when the two geographical location is consistent, then forIt otherwise is 0,For the corresponding weight of geographical location matching value, value
It can specifically set according to actual needs.
Sign language ability matching degree and sign language standardization level, sign language are horizontal related, the present embodiment by user sign language mark
Quasi-ization degree, customer service sign language hierarchical level jointly determine sign language ability matching value, sign language ability matching value specifically can be by formula:
Β * sign language standardization level * sign language hierarchical level value is calculated, and wherein Β is the corresponding weight of sign language ability matching value.
If same customer service called again after completing a translation duties, shows that the matching degree of the customer service is higher,
Call time interval is shorter twice, and corresponding matching degree is higher, and the present embodiment fully considers history call to the shadow of matching degree
It rings, while considering the influence of history call length of time, determined using based on the attenuation function of the interval time called twice
The call history factor.In concrete application embodiment, if last calling is the customer service, recorded by formula (2):
Wherein, t is the adjacent interval time called twice, if last do not call the customer service, being recorded as 0, ε is to exhale
Cry historical factors corresponding weight.
Customer service quality factor can specifically score by formula Q=γ * user and be calculated, and wherein γ is corresponding for customer service quality factor
Weight.
Then (3) calculate final matching value Score as the following formula:
The geographical location Score=matching value+sign language ability matching value+call history factor+customer service quality factor (3)
The customer service in customer service pool list is ranked up according to the size of final matching value Score, distribution sort is near preceding
Customer service to user terminal, complete distribution.
Above-mentioned weighted factor can specifically determine according to actual needs, can also increase other weighted factors, to merge more
More information realization distribution.
It further include every time updating step after the completion of service, updating step includes: to obtain user and sign language in the present embodiment
The scoring of customer service terminal is scored using the user got and updates the customer service information of corresponding sign language customer service terminal and update again
Each customer service pond, and the weight of each weighted factor when being weighted is updated, it can be by each active service quality condition to distribution
Carry out continuous renewal optimization.Specifically the service scoring FB1 that user services this can be obtained after the completion of each service, and
The sign language standardization level scoring FB3 of smooth degree scoring FB2, user that sign language customer service translates this, use service to score
FB1 updates user's scoring of corresponding sign language customer service terminal, customer service quality information, and it is whole to update corresponding sign language customer service using scoring FB2
The sign language interpreter ability value under current service scene type is held, the sign language standardization level of user is updated using scoring FB3, is made
The information that can constantly update user and sign language customer service terminal is obtained, above-mentioned update specifically can be used to be taken plus after newest score value
The modes such as mean value are realized.
The present embodiment updates in step, and the final matching value for scoring and being calculated according to the user of acquisition, which determines, respectively to be added
The weight updated value of weight factor can adaptively be adjusted the weight of every weighted information by each active service quality state,
To continue to optimize matching performance.
In concrete application embodiment, the update of weight parameter is carried out according to the following formula:
B=Β+α (Socre2–FB1*FB2) (5)
ε=ε+α (Socre2–FB22) (6)
Wherein α is updating factor.
If customer service pond is closed on for related service scene type in the customer service pond that the present embodiment further selects, servicing every time
It after the completion further include that correlation between scene type is updated to the smooth degree scoring FB2 that this is translated using sign language customer service terminal
It the step of value, can be by the correlation between the feedback updated different scenes type of real time service.
The more new formula of relevance values R specifically:
R=R+0.001 (FB2-n) (7)
Wherein n is default median, can be set according to actual needs, specifically can use 4.
As shown in Fig. 2, the present invention realizes long-range sign language translation on line visitor using the above method in concrete application embodiment
Take the detailed process of distribution are as follows:
Step 1: customer service pond divides
The field that each sign language interpreter is good at is different, is that different sign language interpreters is provided with not according to service scenarios type
Same scene sign language interpreter ability value Vscene, as shown in table 1.
Table 1: sign language interpreters different scenes ability Data-Statistics.
Customer service pond is formed according to the sign language customer service that the ability value of each sign language interpreter chooses specified quantity, and according to ability
Value size is arranged, and forms customer service pool list, as shown in table 2.
Table 2: customer service pool list.
The case where there may be not available free sign language customer services when the consideration corresponding customer service pond of calling, calculates different service scenarios
Relevance parameter between type, the relevance values table being calculated is as shown in table 3, for idle sign language no in customer service pond
When customer service, the sign language interpreters met successively can be found from the customer service pond closed on.
Table 3: the relevance values of different service scenarios types.
Scene: | Daily communication | Medical interrogation | Bank finance | Government police | Kith and kin link up | Other content |
Daily communication | 1 | 0.76 | 0.84 | 0.85 | 0.92 | 0.86 |
Medical interrogation | 0.76 | 1 | 0.52 | 0.62 | 0.53 | 0.42 |
Bank finance | 0.84 | 0.52 | 1 | 0.71 | 0.56 | 0.57 |
Government police | 0.85 | 0.62 | 0.71 | 1 | 0.67 | 0.67 |
Kith and kin link up | 0.92 | 0.53 | 0.56 | 0.67 | 1 | 0.83 |
Other content | 0.86 | 0.42 | 0.57 | 0.67 | 0.83 | 1 |
Step 2: the service scenarios type of user obtains
User terminal to server initiate call request after, server to user terminal send a scene list, such as with
Above-mentioned corresponding kith and kin's communication, bank finance, medical interrogation, government department etc., user click pair according to oneself translation content
The service scenarios type answered obtains the service scenarios type that user clicks.
Step 3: the customer service distribution based on customer service pond
Step 3.1: after getting the service scenarios type of user's selection, first inquiry and service scenarios selected by user
The corresponding customer service pond of type judges whether available free customer service executes step 3.2, otherwise hold if available free customer service in the customer service pond
Row step 3.3;
Step 3.2: extracting idle customer service list from current customer service pond and carry out customer service distribution;
Step 3.3: according to relevance values between different service scenarios types, relevance values are higher closes on customer service for selection
Pond is closed in customer service pond from this and extracts idle customer service list progress customer service distribution.
When above-mentioned carry out customer service distribution, detailed step are as follows:
The customer service information of each sign language customer service terminal, a certain in the solicited message of acquisition user and the customer service pond currently selected
The solicited message table of user is as shown in table 4, and the customer service information table of a certain sign language customer service terminal is as shown in table 5.
Table 4: the solicited message table of user.
Table 5: the customer service information table of sign language customer service terminal.
Calculate separately each matching value between user and sign language customer service terminal:
Geographical location matching value:
If the two geographical location is consistent, forIt otherwise is 0;
Sign language ability matching value=Β * sign language standardization level * sign language is horizontal;
If last time calling is the customer service call historical factorsIf the last time does not call the customer service,
It is recorded as 0;
Customer service quality factor Q=γ * user scoring;
The final geographical location the matching value Score=matching value+sign language ability matching value+call history factor+customer service quality
The factor
The customer service in customer service pool list is ranked up according to Score, and distribution sort is near preceding customer service.
Step 4: updating
Step 4.1: parameter updates
After the completion of service, obtain scoring FB1 that user services this and sign language customer service terminal this is translated it is suitable
The scoring FB2 of sharp degree, to the scoring FB3 of user's sign language standardization level, by user's scoring of corresponding sign language customer service terminal plus
It is averaged and is updated after upper newest value of feedback FB1, user's sign language standardization level parameter is added into newest value of feedback FB3
After be averaged and be updated, will translation ability value of the corresponding sign language customer service terminal under current service scene type plus newest
It feeds back to after FB2 to be averaged and be updated.
Step 4.2: weight updates
Updated respectively using FB1 geographical location matching value, sign language ability matching value, the call history factor, customer service quality because
The weight parameter of son:
B=Β+α (Socre2–FB1*FB2)
ε=ε+α (Socre2–FB22)
Step 4.3: relevance parameter updates
If current customer service pond is that selection closes on customer service pond, relevance parameter is updated again:
R=R+0.001 (FB2-4)
Step 4.4: customer service pond updates
According to ability value of the sign language customer service terminal of update under service scenarios type, the customer service in customer service pond is updated again
List.
The long-range sign language translation on line customer service distributor of the present embodiment, including being stored with the computer-readable of computer program
Storage medium, computer program realize the above method when executing.
Embodiment 2:
The step of difference from Example 1 is, the present embodiment step S2 are as follows: calculate separately solicited message and each sign language
One in geographical location matching value, sign language ability matching value, the call history factor and customer service quality factor between customer service terminal
Kind is a variety of, and final matching value is obtained after being weighted, and the size according to final matching value determines matched target sign language visitor
Terminal is taken, geographical location matching value is according to the matching between geographical location in solicited message and the geographical location of sign language customer service terminal
Degree obtains, and sign language ability matching value is obtained according to the sign language standardization level of user, the sign language hierarchical level of sign language customer service terminal
Arrive, the call history factor is obtained according to sign language customer service terminal adjacent called interval time twice, customer service quality factor according to
User obtains the scoring of sign language customer service terminal.I.e. the present embodiment uses geographical location, sign language level, history call and quality
Etc. the weighted values of factors directly calculate the matching value of each customer service terminal, realization principle and the basic phase of embodiment 1 of weighting
Together, realization is more simple, can merge the multi informations such as geographical location, sign language level, history call and quality and distribute to realize,
It can be adapted for requiring in more loose scene real-time, allocative efficiency.
Above-mentioned only presently preferred embodiments of the present invention, is not intended to limit the present invention in any form.Although of the invention
It has been disclosed in a preferred embodiment above, however, it is not intended to limit the invention.Therefore, all without departing from technical solution of the present invention
Content, technical spirit any simple modifications, equivalents, and modifications made to the above embodiment, should all fall according to the present invention
In the range of technical solution of the present invention protection.
Claims (10)
1. a kind of long-range sign language translation on line customer service distribution method, which is characterized in that step includes:
S1. the translation request signal that user terminal is sent is received, and includes according to the translation request signal acquisition received
Service scenarios type, the solicited message in geographical location of required translation;
S2. the solicited message that will acquire is matched with the customer service information of each sign language customer service terminal respectively, the customer service letter
Sign language interpreter ability information when breath includes geographical location, different service scenarios types, finds out matched target sign language customer service
Terminal simultaneously distributes to user terminal to carry out sign language translation on line.
2. long-range sign language translation on line customer service distribution method according to claim 1, which is characterized in that the step S2
In, each sign language customer service terminal met is first filtered out according to service scenarios type in the solicited message, then will acquire described
Solicited message is matched with the customer service information of each sign language customer service terminal filtered out respectively, finds out matched target hand
Language customer service terminal.
3. long-range sign language translation on line customer service distribution method according to claim 2, which is characterized in that the step S2's
Step includes:
S21. customer service pond constructs: multiple customer service ponds of the corresponding different service scenarios types of building in advance, in each customer service pond
It is stored with the information that the sign language translation ability under corresponding with service scene type meets multiple sign language customer service terminals of necessary requirement;
S22. customer service pond is screened: filtering out the matched customer service pond according to service scenarios type in the solicited message;
S23. customer service matches: the solicited message being carried out with sign language customer service terminal each in the customer service pond that filters out respectively
Match, finds out matched target sign language customer service terminal distribution to user terminal.
4. long-range sign language translation on line customer service distribution method according to claim 3, which is characterized in that the step S21
When middle building customer service pond, all sign language customer service terminals are arranged according to the sign language interpreter ability under corresponding with service scene type
Sequence, the multiple sign language customer service terminals of specified quantity/or sign language interpreter ability value within a preset range form customer service after choosing sequence
Pond.
5. long-range sign language translation on line customer service distribution method according to claim 3 or 4, which is characterized in that the step
It is preferential to choose the customer service pond identical with service scenarios type in the solicited message in S22, if the customer service pond chosen
Inside not available free customer service chooses other customer service ponds according to the relevance values between different service scenarios types.
6. long-range sign language translation on line customer service distribution method according to claim 3 or 4, which is characterized in that the step
The step of S23 includes: the geographical location calculated separately in the solicited message and the customer service pond between each sign language customer service terminal
It is one or more in matching value, sign language ability matching value, the call history factor and customer service quality factor, and obtained after being weighted
To final matching value, the size according to the final matching value determines matched target sign language customer service terminal, the geographical location
Matching value is obtained according to the matching degree between geographical location in the solicited message and the geographical location of sign language customer service terminal, institute
Sign language ability matching value is stated to be obtained according to the sign language standardization level and/or sign language hierarchical level of sign language customer service terminal, it is described to exhale
Make historical factors be obtained according to sign language customer service terminal adjacent called interval time twice, the customer service quality factor according to
Family obtains the scoring of sign language customer service terminal.
7. long-range sign language translation on line customer service distribution method according to claim 6, which is characterized in that service is completed every time
After further include updating step, the step that updates includes: to obtain the scoring of user and sign language customer service terminal, uses what is got
User, which scores, updates the customer service information for corresponding to sign language customer service terminal and each customer service pond, and is weighted described in update
When each weighted factor weight;The weight updated value of each weighted factor is scored and is calculated with specific reference to the user of acquisition
To the final matching value determination obtain.
8. long-range sign language translation on line customer service distribution method according to claim 7, which is characterized in that the update step
Further include: if the customer service pond currently selected is service scenarios type relevant to destination service scene type in the solicited message
Corresponding closes on customer service pond, updates destination service scene type using the scoring of sign language customer service terminal after the completion of each service
The relevance values between the related service scene type of selection.
9. long-range sign language translation on line customer service distribution method according to claim 1, which is characterized in that the step S2's
Step are as follows: calculate separately the geographical location matching value between the solicited message and each sign language customer service terminal, sign language ability matches
It is one or more in value, the call history factor and customer service quality factor, and final matching value is obtained after being weighted, according to institute
The size for stating final matching value determines matched target sign language customer service terminal, and the geographical location matching value is believed according to the request
Matching degree in breath between geographical location and the geographical location of sign language customer service terminal obtains, the sign language ability matching value according to
The sign language standardization level of user and the sign language hierarchical level of sign language customer service terminal obtain, and the call history factor is according to sign language
Customer service terminal adjacent called interval time twice obtains, and the customer service quality factor is according to user to sign language customer service terminal
Scoring obtains.
10. a kind of long-range sign language translation on line customer service distributor, the computer-readable storage including being stored with computer program
Medium, which is characterized in that the method as described in any one of claim 1~9 is realized when the computer program executes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910397814.6A CN110210721B (en) | 2019-05-14 | 2019-05-14 | Remote sign language online translation customer service distribution method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910397814.6A CN110210721B (en) | 2019-05-14 | 2019-05-14 | Remote sign language online translation customer service distribution method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110210721A true CN110210721A (en) | 2019-09-06 |
CN110210721B CN110210721B (en) | 2023-11-21 |
Family
ID=67787193
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910397814.6A Active CN110210721B (en) | 2019-05-14 | 2019-05-14 | Remote sign language online translation customer service distribution method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110210721B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112581008A (en) * | 2020-12-25 | 2021-03-30 | 上海高顿教育科技有限公司 | Enterprise-level question-answering automatic distribution system based on shift-scheduling weighting |
CN112668464A (en) * | 2020-12-25 | 2021-04-16 | 株洲手之声信息科技有限公司 | Chinese sign language translation model construction method and device fusing scene matching |
CN112866494A (en) * | 2020-12-16 | 2021-05-28 | 华人运通(上海)云计算科技有限公司 | Intelligent customer service cooperation system and method for vehicle |
CN113158694A (en) * | 2021-04-23 | 2021-07-23 | 深圳鸿祥源科技有限公司 | Online translation terminal system based on 5G network and method thereof |
WO2021179703A1 (en) * | 2020-10-20 | 2021-09-16 | 平安科技(深圳)有限公司 | Sign language interpretation method and apparatus, computer device, and storage medium |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004282134A (en) * | 2003-03-12 | 2004-10-07 | Seiko Epson Corp | Sign language translation system and sign language translation program |
CN101742004A (en) * | 2009-12-23 | 2010-06-16 | 中兴通讯股份有限公司 | Call transfer method and device in multi-call central system |
US20100332281A1 (en) * | 2009-06-26 | 2010-12-30 | Microsoft Corporation | Task allocation mechanisms and markets for acquiring and harnessing sets of human and computational resources for sensing, effecting, and problem solving |
CN102609409A (en) * | 2012-02-13 | 2012-07-25 | 浙江飞常哆科技有限公司 | Online translation method, device, system and server |
CN104539814A (en) * | 2014-11-26 | 2015-04-22 | 小米科技有限责任公司 | Customer service call forwarding method and device |
CN105592237A (en) * | 2014-10-24 | 2016-05-18 | ***通信集团公司 | Method and apparatus for session switching, and intelligent customer service robot |
CN107231374A (en) * | 2017-07-08 | 2017-10-03 | 长沙手之声信息科技有限公司 | Deaf person's remote chat method based on online sign language interpreter |
CN107562738A (en) * | 2017-09-08 | 2018-01-09 | 合肥安华信息科技有限公司 | A kind of sign language interpretation method based on user's request |
KR20180013820A (en) * | 2017-08-07 | 2018-02-07 | (주)씨제이텔레닉스 | System and method for analyzing customer type using speech analysis |
CN108846590A (en) * | 2018-07-10 | 2018-11-20 | 广州市联普翻译有限公司 | A kind of translation on line order matching process, system, storage medium and terminal |
-
2019
- 2019-05-14 CN CN201910397814.6A patent/CN110210721B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004282134A (en) * | 2003-03-12 | 2004-10-07 | Seiko Epson Corp | Sign language translation system and sign language translation program |
US20100332281A1 (en) * | 2009-06-26 | 2010-12-30 | Microsoft Corporation | Task allocation mechanisms and markets for acquiring and harnessing sets of human and computational resources for sensing, effecting, and problem solving |
CN101742004A (en) * | 2009-12-23 | 2010-06-16 | 中兴通讯股份有限公司 | Call transfer method and device in multi-call central system |
CN102609409A (en) * | 2012-02-13 | 2012-07-25 | 浙江飞常哆科技有限公司 | Online translation method, device, system and server |
CN105592237A (en) * | 2014-10-24 | 2016-05-18 | ***通信集团公司 | Method and apparatus for session switching, and intelligent customer service robot |
CN104539814A (en) * | 2014-11-26 | 2015-04-22 | 小米科技有限责任公司 | Customer service call forwarding method and device |
CN107231374A (en) * | 2017-07-08 | 2017-10-03 | 长沙手之声信息科技有限公司 | Deaf person's remote chat method based on online sign language interpreter |
KR20180013820A (en) * | 2017-08-07 | 2018-02-07 | (주)씨제이텔레닉스 | System and method for analyzing customer type using speech analysis |
CN107562738A (en) * | 2017-09-08 | 2018-01-09 | 合肥安华信息科技有限公司 | A kind of sign language interpretation method based on user's request |
CN108846590A (en) * | 2018-07-10 | 2018-11-20 | 广州市联普翻译有限公司 | A kind of translation on line order matching process, system, storage medium and terminal |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021179703A1 (en) * | 2020-10-20 | 2021-09-16 | 平安科技(深圳)有限公司 | Sign language interpretation method and apparatus, computer device, and storage medium |
CN112866494A (en) * | 2020-12-16 | 2021-05-28 | 华人运通(上海)云计算科技有限公司 | Intelligent customer service cooperation system and method for vehicle |
CN112866494B (en) * | 2020-12-16 | 2023-01-24 | 华人运通(上海)云计算科技有限公司 | Intelligent customer service cooperation system and method for vehicle |
CN112581008A (en) * | 2020-12-25 | 2021-03-30 | 上海高顿教育科技有限公司 | Enterprise-level question-answering automatic distribution system based on shift-scheduling weighting |
CN112668464A (en) * | 2020-12-25 | 2021-04-16 | 株洲手之声信息科技有限公司 | Chinese sign language translation model construction method and device fusing scene matching |
CN112668464B (en) * | 2020-12-25 | 2022-12-02 | 株洲手之声信息科技有限公司 | Chinese sign language translation model construction method and device fusing scene matching |
CN113158694A (en) * | 2021-04-23 | 2021-07-23 | 深圳鸿祥源科技有限公司 | Online translation terminal system based on 5G network and method thereof |
CN113158694B (en) * | 2021-04-23 | 2023-01-13 | 深圳鸿祥源科技有限公司 | Online translation terminal system based on 5G network and method thereof |
Also Published As
Publication number | Publication date |
---|---|
CN110210721B (en) | 2023-11-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110210721A (en) | A kind of long-range sign language translation on line customer service distribution method and device | |
CN108269128B (en) | Advertisement putting method, device, equipment and storage medium | |
CN107426621B (en) | A kind of method and system showing any active ues image in mobile terminal direct broadcasting room | |
CN106204120A (en) | Advertisement broadcast method and system, advertisement playing device and method of work thereof | |
CN107528761A (en) | A kind of method and system of wechat public number automatic information reply | |
CN105898424A (en) | Video recommendation method and system and server | |
CN109389501A (en) | A kind of calculating equipment, computing system | |
Ogunlade et al. | Knowledge levels of extension agents and their perceived impact of climate change on extension service provision in Ghana | |
CN106535083A (en) | Method and device for providing service for remote UE (User Equipment) | |
Whyte | On studying China at a distance | |
CN107205051A (en) | The method and processing system, intelligent refrigerator, cloud server of intelligence message | |
CN103309963A (en) | Intelligent customer service processing method and system | |
CN109788448A (en) | Service number subregion tandem method and system | |
CN110213450A (en) | Based on polycentric sign language online customer service allocation management method, apparatus and system | |
CN104812067B (en) | Smallcell base station callings optimization method and system | |
Keleher et al. | Connecting isolated communities: Quantitative evidence on the adoption of community cellular networks in the Philippines | |
Kisiel et al. | Functioning of the Polish LAG in the Context of the Leader Initiative | |
CN105120345B (en) | A kind of method for ordering television channels and system based on range statistics and switching times | |
EP1266515B1 (en) | Method for operating a telecommunication system and telecommunication arrangement | |
CN109996331A (en) | Paging method and device, policy distribution method are with regard to device, network element and storage medium | |
CN110246386A (en) | A kind of intelligent ideological education display control program and its method based on cloud platform | |
CN105393616A (en) | Exchange, wireless network control device, wireless network device, and wireless communication network system | |
Mol | Joint environmental policymaking in Europe: between deregulation and political modernization | |
Aenlle et al. | Best Practices for Communicating about Outdoor Residential Water Conservation: AEC703/WC366, 9/2020 | |
Zimbalist | Reasons Why Hosting the Olympics is a Loser’s Game’ |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20231023 Address after: Zhuzhou Yandi University, 229 Shaoshan Road, Tianyuan District, Zhuzhou City, Hunan Province, 412007 Applicant after: Zhuzhou shouzhisheng Information Technology Co.,Ltd. Address before: Yandi University, 229 Shaoshan Road, Tianyuan District, Zhuzhou City, Hunan Province, 412007 Applicant before: CHANGSHA SHOUZHISHENG INFORMATION TECHNOLOGY CO.,LTD. |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |