CN109949103A - A kind of data processing method, device and electronic equipment - Google Patents
A kind of data processing method, device and electronic equipment Download PDFInfo
- Publication number
- CN109949103A CN109949103A CN201910252933.2A CN201910252933A CN109949103A CN 109949103 A CN109949103 A CN 109949103A CN 201910252933 A CN201910252933 A CN 201910252933A CN 109949103 A CN109949103 A CN 109949103A
- Authority
- CN
- China
- Prior art keywords
- data
- target user
- evaluation
- user
- evaluation data
- 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
Abstract
This application discloses a kind of data processing method, device and electronic equipments, method includes: after customer service system exports the second data for the first data of target user's input, it monitors whether to receive the feedback data that the target user inputs, the feedback data is to describe the target user to the satisfaction of second data;If not receiving the feedback data of target user's input, the evaluation data of the target user are generated, the evaluation data are to simulate the feedback data.As it can be seen that the evaluation data that can simulate its feedback data can be generated for user in the application when user does not have feedback data, to get the evaluation to service in customer service system.
Description
Technical field
This application involves intelligent customer service technical field more particularly to a kind of data processing methods, device and electronic equipment.
Background technique
Intelligent customer service system is more and more widely used in each service field.In intelligent customer service system, need to pass through
It is for statistical analysis to the evaluation of user, to understand the service scenario to user, and then corresponding adjustment is made, such as adjustment pair
Mode or form that user is serviced etc..
The master that user actively inputs the evaluation usually user of intelligent customer service system after service terminates at present
Dynamic feedback information.
Summary of the invention
In view of this, the application provides a kind of data processing method, comprising:
After customer service system exports the second data for the first data of target user's input, monitor whether to receive institute
The feedback data of target user's input is stated, the feedback data is to describe satisfaction of the target user to second data
Degree;
If not receiving the feedback data of target user's input, the evaluation data of the target user, institute are generated
Commentary valence mumber is according to simulate the feedback data.
The above method, it is preferred that generate the evaluation data of the target user, comprising:
Acquire the user action data of the target user;
The user action data are handled, the evaluation data of the target user are obtained.
The above method, it is preferred that the user action data are handled, the review number of the target user is obtained
According to specifically including:
The user action data are handled using disaggregated model, obtain the evaluation data of the target user;
Wherein, the disaggregated model generates in the following manner:
Obtain at least one in the customer service system evaluated service historical action data and its corresponding history evaluation
Data;
Model training is carried out to the historical action data and the history evaluation data using sorting algorithm, to be divided
Class model.
The above method, it is preferred that further include:
Obtain user's portrait information of the target user;
Wherein, after the evaluation data for generating the target user, the method also includes:
Using user portrait information, modify to the evaluation data of the target user, it is revised to obtain
Evaluate data.
The above method, it is preferred that further include:
Obtain the output parameter when customer service system exports second data;
Wherein, after the evaluation data for generating the target user, the method also includes:
It using the output parameter, modifies to the evaluation data of the target user, to obtain revised evaluation
Data.
The above method, it is preferred that further include:
The target user is obtained to the product behavioral data of target product;
Wherein, after the evaluation data in the generation target user, the method also includes:
Using the product behavioral data, modify to the evaluation data of the target user, it is revised to obtain
Evaluate data.
The above method, it is preferred that further include:
The evaluation data of user action data and the target user based on the target user, to the disaggregated model
It optimizes.
The above method, it is preferred that further include:
Evaluation data and the history evaluation data, the entirety for generating the customer service system based on the target user are commented
Valence result.
Present invention also provides a kind of data processing equipments, comprising:
Data monitoring unit, for customer service system for target user input the first data export the second data it
Afterwards, it monitors whether to receive the feedback data that the target user inputs, the feedback data is to describe the target user
To the satisfaction of second data;
Generation unit is evaluated, if the feedback data for not receiving target user's input, generates the target
The evaluation data of user, the evaluation data are to simulate the feedback data.
Present invention also provides a kind of electronic equipment, comprising:
Customer service interactive device, for receiving the first data of target user's input, and for first data output the
Two data;
Processing equipment is evaluated, for monitoring whether to receive the feedback data of target user's input, the feedback coefficient
According to describe the target user to the satisfaction of second data, if not receiving the anti-of target user's input
Data are presented, generate the evaluation data of the target user, the evaluation data are to simulate the feedback data.
It can be seen from the above technical proposal that in a kind of data processing method disclosed in the present application, device and electronic equipment,
After customer service system exports the second data for the first data of target user's input, by monitoring whether that receiving target uses
Family input describes target user to the feedback data of the second data satisfaction, thus if being not received by feedback data, that
The evaluation data for just generating target user, with the feedback data of evaluation data simulated target user.As it can be seen that energy in the application
It is enough that the evaluation data that can simulate its feedback data are generated for user when user does not have feedback data, thus in customer service system
Get the evaluation to service.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart for data processing method that a kind of possible implementation of the embodiment of the present application provides;
Fig. 2 is a kind of part process for data processing method that a kind of possible implementation of the embodiment of the present application provides
Figure;
Fig. 3 is the application exemplary diagram of the embodiment of the present application;
Fig. 4 is a kind of structural representation for data processing equipment that a kind of possible implementation of the embodiment of the present application provides
Figure;
Fig. 5 is the structural schematic diagram for a kind of electronic equipment that a kind of possible implementation of the embodiment of the present application provides;
Fig. 6 is the another application exemplary diagram of the embodiment of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
As shown in Figure 1, being a kind of the embodiment of the present application provided data processing side in one possible implementation
The flow chart of method, this method is suitable for the equipment such as customer care server where intelligent customer service system.
In the present embodiment, this method can specifically include following steps:
Step 101: after customer service system exports the second data for the first data of target user's input, monitoring whether
The feedback data of target user's input is received, if not provided, so executing step 102.
Wherein, when the first data can provide a customer service for customer service system for target user, target user is to visitor
Dress system, which seek advice from etc., operates inputted data, as putd question to the first data of delivery availability or carried out reparation consulting after sale
First data etc..And after the second data refer to that target user inputs the first data to customer service system, customer service system is for the
The reply data of one data, such as reply the second data of delivery availability or the second data of measure of indemnity.
In the present embodiment after customer service system is for first the second data of data recovery, whether target user is inputted instead
Feedback data are monitored, which can describe target user to the satisfaction of the second data, that is to say, that the present embodiment
Whether middle monitoring objective user is directed to the corresponding satisfaction evaluation data of the second data feedback that customer service system is replied, and such as passes through
Control selection inputs very satisfied or unsatisfied satisfaction option and evaluates data, or is liked very much by the customized input of keyboard
Joyous character content evaluation data, etc..
It should be noted that can be after customer service system exports the second data in the present embodiment, starting to execute to monitor is
The no movement for receiving feedback data, and if monitoring movement continue to condition meet after, be still not received by target user
The feedback data of input executes step 102 it is considered that being not connected to the feedback data of target user's input at this time.
For example, certain time length is continued to monitor, if being not connected to mesh after starting to execute monitoring movement in the present embodiment
The feedback data of mark user's input continues to preset duration threshold value, for example, target user was not also to service in more than 12 hours
Input feedback data, at this point, executing step 102;
For another example, it is continued to monitor, after starting to execute monitoring movement if had begun in target user in the present embodiment
Into service (being different from this service) next time still without the feedback data for receiving target user's input, at this point, executing step
Rapid 102.
Step 102: generating the evaluation data of target user.
Wherein, the evaluation data in the present embodiment are monitoring target to analog feedback data, and in the present embodiment
The feedback data of the target user is simulated when user is not to customer service system input feedback data, to generate target user
Evaluation data, be used for subsequent data analysis.
It should be noted that evaluation data generated in the present embodiment can such as be returned at least one evaluation criterion
Complex velocity, reply accuracy or whether help to etc., can specifically be indicated with the mode of verbal description, as reply content is rich
Rich, the reply word contents such as fast speed or content inaccuracy, alternatively, evaluation data can also use grade or star table
Show, can specifically be indicated using numerical value, character representation can also be used, alternatively, indicated using expression packet, etc..For example, numerical value
Higher characterized evaluation satisfaction is higher, character a is higher than the evaluation satisfaction that character c is characterized, smiling face's expression packet compares nutcracker
The evaluation satisfaction that expression packet is characterized is high or 5 stars are higher than the evaluation satisfaction that 1 star is characterized, etc..
By above scheme it is found that in a kind of data processing method provided by the embodiments of the present application, mesh is directed in customer service system
Mark user's input the first data export the second data after, by the description target for monitoring whether to receive target user's input
User is to the feedback data of the second data satisfaction, to just generate target user if being not received by feedback data
Evaluation data, with the feedback data of evaluation data simulated target user.As it can be seen that can be not anti-in user in the present embodiment
The evaluation data that can simulate its feedback data are generated for user when presenting data, to get in customer service system to service
Evaluation.
In one implementation, step 102, specifically can be by with lower section when generating the evaluation data of target user
Formula is realized, as shown in Figure 2:
Step 201: acquiring the user action data of target user.
Wherein, user action data are it is to be understood that consulting behavior operation of the target user conducted in customer service system
Data, including various input action data, such as the input of character input, image or the input operation of expression packet, specifically include input
Content and the input data such as duration, wherein input content can be understood as that target user inputs in customer service system it is various in
Hold, such as one of character, image and expression packet or a variety of, and inputs duration it is to be understood that target user is more than input
Duration used when input content, for example, target user puts question to product function to customer service system, customer service system answers target later
After the function handbook of consumer products, puts question to or reply again after the function handbook that target user replys for customer service system
Duration used.
It should be noted that user action data refer to target user and customer service system current primary in the present embodiment
User action data in service, as shown in Figure 3, target user can log in intelligent customer service system by mobile phone, and carry out
Corresponding movement.Here user action data refer in the current service for the feedback data for being not received by target user's input
Related user action data, being different from the historical action data with feedback data or other does not have the history of feedback data
Action data.
As it can be seen that target user inputs each time, as one-off, primary consulting of the target user in customer service system is taken
In business, it is understood that there may be one or more movement, one or more user action data of these movements composition target user.It needs
Illustrate, user action data can be the relevant data of reference content that target user inputs in customer service system, can also
It can include the feedback data that target user inputs in customer service system, and be not connected to the anti-of target user in the present embodiment
Present data in the case where, in step 201 collected user action data be that target user inputs in customer service system
The relevant data of reference content such as put question to the action data of product function.
Step 202: user action data being handled, the evaluation data of target user are obtained.
Specifically, can be handled in the following manner user action data in the present embodiment:
User action data are handled using disaggregated model, obtain the evaluation data of target user.
Wherein, disaggregated model can use sorting algorithm such as decision tree, logistic regression, naive Bayesian, neural network, branch
Hold vector machine SVM (Support Vector Machine), the realization of K arest neighbors KNN (k-NearestNeighbor) scheduling algorithm is built
Mould.
Specifically, disaggregated model can be accomplished by the following way in building:
Firstly, obtain customer service system at least one evaluated service historical action data and its corresponding history evaluation
Data, and then model training is carried out to historical action data and history evaluation data using sorting algorithm, to obtain classification mould
Type.
Wherein, service has been evaluated it is to be understood that having had in the various completed services that customer service system is recorded
The service of the feedback data of user's input or simulation, and evaluated and be corresponding in service: historical action data and its corresponding
History evaluation data, for example, the evaluation data etc. of interactive action data and user between user and customer service system to customer service.It needs
It is noted that having evaluated in service here may include having the history of target user to evaluate service, it also may include having
The history of other users has evaluated service, correspondingly, include the historical action data of target user in historical action data,
It can wrap the historical action data containing other users, include the history evaluation data of target user in history evaluation data,
Also the history evaluation data containing other users be can wrap.
Correspondingly, in the present embodiment after the historical action data and history evaluation data for acquiring the service of evaluation,
Using these data as the input sample data of sorting algorithm such as SVM or KNN, and model training is carried out, to construct classification
Device, i.e. disaggregated model.Later, the user action data of the target user acquired above are input in disaggregated model
Reason, to obtain the evaluation data of target user.
Based on implementation above, the user action data of target user and the review number of generation are also based in the present embodiment
It is optimized according to disaggregated model, for example, using the user action data of target user and the evaluation data of generation as history number
According in the model construction that is added to disaggregated model or training, thus Optimum Classification model, thus later use disaggregated model again
It is secondary generate review number according to when can be improved evaluation data accuracy.
In addition, being also based on the evaluation data and history evaluation data of target user in the present embodiment, customer service system is generated
The overall evaluation data of system.For example, based on target user and other users to the evaluation data of customer service system in the present embodiment, it is right
The integrity service situation of customer service system is evaluated, as customer service system overall operation is normal, customer service is replied universal relatively slow or returns
Multiple majority inaccuracy etc..
In practical applications, the present embodiment generate target user evaluation data after, can also to evaluation data into
Row amendment, so that evaluation data are more intended to close to the evaluation of target user, for example, being drawn by the user for obtaining target user
Output parameter and target user when as information, customer service system the second data of output is in product behavioral data of target product etc.
It is one or more, modify to the evaluation data of the target user of generation, to obtain revised evaluation data.Specifically
It is as follows:
In one implementation, it can be utilized after the user for obtaining target user draws a portrait information in the present embodiment
User's portrait information modifies to the evaluation data of the target user of generation, to obtain revised evaluation data.
Wherein, the characteristics of user's portrait information can characterize target user, including in personality feature and behavioral characteristic etc.
It is one or more, for example, target user is that mood is irascible or fastidious personality or target user belong to habit and let up on thing
People of object, etc..
The evaluation data of generation are modified based on user's portrait information in the present embodiment, so that evaluation data are more
Stick on the evaluation habit or evaluation behavior of close-target user, so as to improve the accuracy of evaluation data generated.
For example, target user, in this interaction with customer service system, target user shows as width and treats the property of things
Lattice show that target user may be wider sum to the evaluation of customer service system, at this time can be to the evaluation data being currently generated
It is modified, as improved satisfaction numerical value characterized in evaluation data etc..
For another example, target user shows as harsh behaviour style in history interaction, then showing target user to customer service
The evaluation of system may be that comparison is harsh, can be modified to current evaluation data generated at this time, such as reduce evaluation
Satisfaction numerical value etc. characterized in data.
In one implementation, in the present embodiment can obtain customer service system export the second data when output parameter
Later, it using output parameter, modifies to the evaluation data of target user, to obtain revised evaluation data.
Wherein, output parameter can characterize the state that customer service system replys target user, such as mention in target user
The reply speed of customer service system after asking, customer service system reply target user after target user same problem is putd question to again or
Person the state parameters such as puts question to regard to different problems again.
Be modified in the present embodiment based on evaluation data of the output parameter of customer service system to generation, thus avoid because
User personality is formed by the case where malice evaluation or system automatic favorable comment, to improve the accurate of evaluation data generated
Property.
For example, target user is in the interaction with customer service system, the enquirement to target user, customer service system is in 1 second
It replys, shows that customer service system provides service rapidly and efficiently for target user, it at this time can be based on the reply speed to generation
Evaluation data are modified, as improved satisfaction numerical value characterized in evaluation data etc..
For another example, target user is in the interaction with customer service system, and the enquirement to target user, customer service system only once exists
It is replied in 1 second, other times for requiring 5 seconds or more could reply target user, show that customer service system provides for target user
Service there is a situation where slow, can be modified based on evaluation data of the reply speed to generation, be commented as reduced at this time
Satisfaction numerical value etc. of the valence mumber characterized in.
In one implementation, target user can obtained to the product behavioral data of target product in the present embodiment
Later, it using product behavioral data, modifies to the evaluation data of target user, to obtain revised evaluation data.
Wherein, product behavioral data can be understood as after providing service to target user by customer service system, target
Target product provided by target product that user seeks advice from it based on the service that customer service system provides or the customer service system
Satisfaction, as target user buys (satisfaction) to target product again, target user (may to the page browsing of target product
It is satisfied), target user returns goods to the target product to have placed an order or non-payment (dissatisfied) etc..
The evaluation data of generation are repaired based on product behavioral data of the target user to target product in the present embodiment
Just, so that evaluation data are more intended to the evaluation of customer service system close to target user, to improve evaluation generated
The accuracy of data.
For example, target user is terminating and after the interaction of customer service system, is returning to the product page of target product and again
Place an order, show target user's this service corresponding to target product be it is satisfied, at this time can be based on the behavior number that this places an order
It is modified according to the evaluation data of generation, as improved satisfaction numerical value characterized in evaluation data etc..
It is a kind of the embodiment of the present application provided data processing equipment in one possible implementation with reference to Fig. 4
Structural schematic diagram, the device is suitable for the equipment such as customer care server where customer service system.
In the present embodiment, which can specifically include with flowering structure:
Data monitoring unit 401, for being directed to first the second data of input and output of target user's input in customer service system
Later, it monitors whether to receive the feedback data that target user inputs.
Wherein, feedback data is to describe target user to the satisfaction of the second data.
Generation unit 402 is evaluated, if the feedback data for not receiving target user's input, generates target user's
Evaluate data.
Wherein, feedback data of the evaluation data to simulated target user.
In one implementation, evaluation generation unit 402 can generate in the following manner evaluation data:
The user action data of acquisition target user handle the user action data, obtain the mesh later
Mark the evaluation data of user.
Specifically, evaluation generation unit 402 can be obtained ahead of time at least one in the customer service system and evaluate service
Historical action data and its corresponding history evaluation data, and using sorting algorithm to the historical action data and the history
It evaluates data and carries out model training, to obtain disaggregated model.
Correspondingly, evaluation generation unit 402 is handled to the user action data, obtain the target user's
When evaluating data, the user action data can specifically be handled using the disaggregated model that the above training obtains, be obtained
The evaluation data of the target user.
Further, evaluation generation unit 402, can be based on the user of the target user after obtaining evaluation data
The evaluation data of action data and the target user, optimize above disaggregated model, to improve the standard of subsequent evaluation
True property.
By above scheme it is found that in a kind of data processing equipment provided by the embodiments of the present application, mesh is directed in customer service system
Mark user's input the first data export the second data after, by the description target for monitoring whether to receive target user's input
User is to the feedback data of the second data satisfaction, to just generate target user if being not received by feedback data
Evaluation data, with the feedback data of evaluation data simulated target user.As it can be seen that can be not anti-in user in the present embodiment
The evaluation data that can simulate its feedback data are generated for user when presenting data, to get in customer service system to service
Evaluation.
Further, evaluation generation unit 402 can pass through the user of acquisition target user after obtaining evaluation data
Output parameter and target user when the information, customer service system the second data of output of drawing a portrait is to product behavioral data of target product etc.
One of or it is a variety of, modify to the evaluation data of target user, to obtain modified evaluation data.
In addition, generation unit 402 is evaluated after obtaining evaluation data, it can be based on the evaluation data of the target user
And the history evaluation data, generate the overall evaluation result of the customer service system.
It should be noted that the specific implementation of each unit of data processing equipment can be with reference to hereinbefore phase in the present embodiment
Content is answered, and will not be described here in detail.
With reference to Fig. 5, be the embodiment of the present application in one possible implementation provided by a kind of electronic equipment knot
Structure schematic diagram, the electronic equipment can be the equipment such as the customer care server where customer service system.
In the present embodiment, which can specifically include with flowering structure:
Customer service interactive device 501 is exported for receiving the first data of target user's input, and for first data
Second data.
Wherein, customer service interactive device 501 has input equipment and output equipment, and as shown in Figure 5, target user passes through defeated
Enter equipment and input the first data, electronic equipment is that target user exports the second data, realization and target user by output equipment
Between customer service interaction.
Processing equipment 502 is evaluated, for monitoring whether to receive the feedback data of target user's input, the feedback
Data are to describe the target user to the satisfaction of second data, if not receiving target user's input
Feedback data generates the evaluation data of the target user, and the evaluation data are to simulate the feedback data.
Wherein, evaluation processor equipment 502 can be realized by processor, can monitor whether input equipment terminates to mesh
The feedback data of user's input is marked, if not provided, so generating the evaluation data for capableing of analog feedback data.
In addition, can also include memory in electronic equipment, it, can also be to store first to store evaluation data
Data, the second data and various history service datas etc..
In one implementation, evaluation processing equipment 502 can generate in the following manner evaluation data:
The user action data of acquisition target user handle the user action data, obtain the mesh later
Mark the evaluation data of user.
Specifically, evaluation processing equipment 502 can be obtained ahead of time at least one in the customer service system and evaluate service
Historical action data and its corresponding history evaluation data, and using sorting algorithm to the historical action data and the history
It evaluates data and carries out model training, to obtain disaggregated model.
Correspondingly, evaluation processing equipment 502 is handled to the user action data, obtain the target user's
When evaluating data, the user action data can specifically be handled using the disaggregated model that the above training obtains, be obtained
The evaluation data of the target user.
Further, evaluation processing equipment 502, can be based on the user of the target user after obtaining evaluation data
The evaluation data of action data and the target user, optimize above disaggregated model, to improve the standard of subsequent evaluation
True property.
By above scheme it is found that in a kind of electronic equipment provided by the embodiments of the present application, used in customer service system for target
After first data of family input export the second data, by the description target user for monitoring whether to receive target user's input
To the feedback data of the second data satisfaction, to just generate commenting for target user if being not received by feedback data
Valence mumber evidence, with the feedback data of evaluation data simulated target user.As it can be seen that can not there is no feedback coefficient in user in the present embodiment
According to when be generated for user and can simulate the evaluation data of its feedback data, service is commented to be got in customer service system
Valence.
Further, evaluation processing equipment 502 can pass through the user of acquisition target user after obtaining evaluation data
Output parameter and target user when the information, customer service system the second data of output of drawing a portrait is to product behavioral data of target product etc.
One of or it is a variety of, modify to the evaluation data of target user, to obtain modified evaluation data.
In addition, processing equipment 502 is evaluated after obtaining evaluation data, it can be based on the evaluation data of the target user
And the history evaluation data, generate the overall evaluation result of the customer service system.
It should be noted that the specific implementation of each structure of electronic equipment can be with reference to hereinbefore corresponding interior in the present embodiment
Hold, and will not be described here in detail.
Below by taking intelligent customer service system as an example, the technical solution in the present embodiment is illustrated:
Firstly, S is defined as to the primary service (Service) in intelligent customer service system in the present embodiment, by user action
(Action) it is defined as A, user behavior (Behavior) is defined as B, user's evaluation (Evaluation) is defined as E, by
This:
S is the set of B, i.e. S={ B1, B2, B3 ..., Bn }, B are the set of A, i.e. B={ A1, A2, A3 ..., An };
Wherein, A={ content of text text, feedback content feedback, it is intended that content intent ..., evaluation content E },
E={ star star marks tags, illustrates comment... }, i.e. E are a kind of special A.
It is the user that user's evaluation is made by user behavior (B) Lai Weiwei containing user's evaluation (E) in the present embodiment
Behavior (B) speculates a user's evaluation (E).
Wherein, the core element of the present embodiment are as follows:
1, by burying an acquisition user action A and user's evaluation data E to intelligent customer service system.The input each time of user
It is denoted as Ai, the evaluation that user makes is recorded as Ei.
2, using has the user data of user's evaluation E as the input sample data of sorting algorithm (such as SVM, KNN etc.), comes
Structural classification device.
3, deduction is made using user data of the classifier after training to no user evaluation E, inferred results are this time
The user's evaluation E of service.
4, after being inferred to user's evaluation, service data is summarized, obtains data set D={ S1 ..., Sn }, wherein Si=
{ B1 ..., Be }, Be={ A1 ..., Ae }.D, which carries out statistical analysis appropriate, for data sets can be obtained intelligent customer service system
Whole user's evaluation.Such as comment star.
As it can be seen that the data cover amount of user's evaluation is improved, according to known in the present embodiment by the way of supervised learning
The user behavior of user's evaluation infers the evaluations of other users, has reached through user's evaluation and has analyzed the mesh of intelligent customer service
's.
And full-automatic processing is realized in the present embodiment, evaluation of installing machines to replace manual labor eliminates manual read/mark
The time of note, to reduce time cost, human cost.And data cover is strong, by way of intelligently inferring, can incite somebody to action
Total data all stamps user's evaluation label, expands the data set size that analysis uses, and improves the matter of data analysis set
Amount improves the confidence level of analysis result.
Meanwhile the evaluation result inferred in the present embodiment can more reflect true user intention.So as to avoid artificial
The entrained subjective impact of analysis, while the more closing to reality situation of the mode relative to given default value, largely subtract
The distortion risk generated during data record is lacked.
Specifically, may include following processing in the present embodiment in intelligent customer service system as shown in data trend in Fig. 6
Process:
Firstly, obtaining related data by burying a little to intelligent customer service system (behavioral data evaluates data);
Then, for there is the data of user's evaluation as training sample, the disaggregated model for generating corresponding user's evaluation (can make
With decision tree, logistic regression, naive Bayesian, neural network scheduling algorithm);
And for the data of not user's evaluation, the disaggregated model is executed to it, generates the user's evaluation of prediction;
Finally, evaluation (true, to infer) Data Integration of user is got up, it is for statistical analysis.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with other
The difference of embodiment, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is said referring to method part
It is bright.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond scope of the present application.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. a kind of data processing method, comprising:
After customer service system exports the second data for the first data of target user's input, monitor whether to receive the mesh
The feedback data of user's input is marked, the feedback data is to describe the target user to the satisfaction of second data;
If not receiving the feedback data of target user's input, the evaluation data of the target user, institute's commentary are generated
Valence mumber is according to simulate the feedback data.
2. according to the method described in claim 1, generating the evaluation data of the target user, comprising:
Acquire the user action data of the target user;
The user action data are handled, the evaluation data of the target user are obtained.
3. obtaining commenting for the target user according to the method described in claim 2, handling the user action data
Valence mumber evidence, specifically includes:
The user action data are handled using disaggregated model, obtain the evaluation data of the target user;
Wherein, the disaggregated model generates in the following manner:
Obtain historical action data and its corresponding history evaluation data that at least one in the customer service system has evaluated service;
Model training is carried out to the historical action data and the history evaluation data using sorting algorithm, to obtain classification mould
Type.
4. method according to claim 1 or 2, further includes:
Obtain user's portrait information of the target user;
Wherein, after the evaluation data for generating the target user, the method also includes:
Using user portrait information, modify to the evaluation data of the target user, to obtain revised evaluation
Data.
5. method according to claim 1 or 2, further includes:
Obtain the output parameter when customer service system exports second data;
Wherein, after the evaluation data for generating the target user, the method also includes:
It using the output parameter, modifies to the evaluation data of the target user, to obtain revised evaluation data.
6. method according to claim 1 or 2, further includes:
The target user is obtained to the product behavioral data of target product;
Wherein, after the evaluation data in the generation target user, the method also includes:
It using the product behavioral data, modifies to the evaluation data of the target user, to obtain revised evaluation
Data.
7. according to the method described in claim 3, further include:
The evaluation data of user action data and the target user based on the target user carry out the disaggregated model
Optimization.
8. according to the method described in claim 3, further include:
Evaluation data and the history evaluation data based on the target user, generate the overall evaluation knot of the customer service system
Fruit.
9. a kind of data processing equipment, comprising:
Data monitoring unit, for supervising after customer service system exports the second data for the first data of target user's input
The feedback data for whether receiving target user's input is surveyed, the feedback data is to describe the target user to described
The satisfaction of second data;
Generation unit is evaluated, if the feedback data for not receiving target user's input, generates the target user
Evaluation data, the evaluation data are to simulate the feedback data.
10. a kind of electronic equipment, comprising:
Customer service interactive device, for receiving the first data of target user's input, and for the second number of first data output
According to;
Processing equipment is evaluated, for monitoring whether to receive the feedback data of target user's input, the feedback data is used
To describe the target user to the satisfaction of second data, if not receiving the feedback coefficient of target user's input
According to generating the evaluation data of the target user, the evaluation data are to simulate the feedback data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910252933.2A CN109949103B (en) | 2019-03-29 | 2019-03-29 | Data processing method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910252933.2A CN109949103B (en) | 2019-03-29 | 2019-03-29 | Data processing method and device and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109949103A true CN109949103A (en) | 2019-06-28 |
CN109949103B CN109949103B (en) | 2021-10-22 |
Family
ID=67013112
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910252933.2A Active CN109949103B (en) | 2019-03-29 | 2019-03-29 | Data processing method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109949103B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111046162A (en) * | 2019-12-23 | 2020-04-21 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
CN111210232A (en) * | 2019-12-31 | 2020-05-29 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
CN111881268A (en) * | 2020-05-25 | 2020-11-03 | 重庆兆光科技股份有限公司 | Consulting robot evaluation method, system and equipment based on user behavior sequence |
CN112527978A (en) * | 2020-11-10 | 2021-03-19 | 联想(北京)有限公司 | Conversation processing method and equipment |
WO2021073258A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技(深圳)有限公司 | Task follow-up method, apparatus and device based on emotion analysis, and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102081631A (en) * | 2009-11-30 | 2011-06-01 | 国际商业机器公司 | Answer support system and method |
CN103186555A (en) * | 2011-12-28 | 2013-07-03 | 腾讯科技(深圳)有限公司 | Evaluation information generation method and system |
CN105808652A (en) * | 2016-02-26 | 2016-07-27 | 北京京东尚科信息技术有限公司 | Realization method and device of online customer services |
CN107220353A (en) * | 2017-06-01 | 2017-09-29 | 深圳追科技有限公司 | A kind of intelligent customer service robot satisfaction automatic evaluation method and system |
CN108074108A (en) * | 2017-11-02 | 2018-05-25 | 平安科技(深圳)有限公司 | A kind of display methods and its terminal of net recommendation |
CN108133393A (en) * | 2017-12-28 | 2018-06-08 | 新智数字科技有限公司 | Data processing method and system |
CN109241518A (en) * | 2017-07-11 | 2019-01-18 | 北京交通大学 | A kind of detection network navy method based on sentiment analysis |
CN109409759A (en) * | 2018-11-06 | 2019-03-01 | 河南理工大学 | A kind of data digging system and method based on Coal Mine Security Training game |
-
2019
- 2019-03-29 CN CN201910252933.2A patent/CN109949103B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102081631A (en) * | 2009-11-30 | 2011-06-01 | 国际商业机器公司 | Answer support system and method |
CN103186555A (en) * | 2011-12-28 | 2013-07-03 | 腾讯科技(深圳)有限公司 | Evaluation information generation method and system |
CN105808652A (en) * | 2016-02-26 | 2016-07-27 | 北京京东尚科信息技术有限公司 | Realization method and device of online customer services |
CN107220353A (en) * | 2017-06-01 | 2017-09-29 | 深圳追科技有限公司 | A kind of intelligent customer service robot satisfaction automatic evaluation method and system |
CN109241518A (en) * | 2017-07-11 | 2019-01-18 | 北京交通大学 | A kind of detection network navy method based on sentiment analysis |
CN108074108A (en) * | 2017-11-02 | 2018-05-25 | 平安科技(深圳)有限公司 | A kind of display methods and its terminal of net recommendation |
CN108133393A (en) * | 2017-12-28 | 2018-06-08 | 新智数字科技有限公司 | Data processing method and system |
CN109409759A (en) * | 2018-11-06 | 2019-03-01 | 河南理工大学 | A kind of data digging system and method based on Coal Mine Security Training game |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021073258A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技(深圳)有限公司 | Task follow-up method, apparatus and device based on emotion analysis, and storage medium |
CN111046162A (en) * | 2019-12-23 | 2020-04-21 | 联想(北京)有限公司 | Information processing method and device and electronic equipment |
CN111210232A (en) * | 2019-12-31 | 2020-05-29 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
CN111881268A (en) * | 2020-05-25 | 2020-11-03 | 重庆兆光科技股份有限公司 | Consulting robot evaluation method, system and equipment based on user behavior sequence |
CN112527978A (en) * | 2020-11-10 | 2021-03-19 | 联想(北京)有限公司 | Conversation processing method and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109949103B (en) | 2021-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109949103A (en) | A kind of data processing method, device and electronic equipment | |
US11341439B2 (en) | Artificial intelligence and machine learning based product development | |
Bohanec et al. | Decision-making framework with double-loop learning through interpretable black-box machine learning models | |
Zhang | Effective requirements development-A comparison of requirements elicitation techniques | |
Howard | Profiting from innovation: the report of the three-year study from the National Academy of Engineering | |
Duncan et al. | Probabilistic modeling of a sales funnel to prioritize leads | |
Xu et al. | Knowledge value chain: an effective tool to measure knowledge value | |
Oke | Manufacturing quality function deployment: Literature review and future trends | |
Park et al. | A semi-automated filtering technique for software process tailoring using neural network | |
Mabin et al. | Framework for understanding the complementary nature of TOC frames: insights from the product mix dilemma | |
CN112149838A (en) | Method, device, electronic equipment and storage medium for realizing automatic model building | |
US20180285758A1 (en) | Methods for creating and analyzing dynamic trail networks | |
CN108563648A (en) | data display method and device, storage medium and electronic device | |
Meziani et al. | Towards a collaborative business process management methodology | |
Koorneef et al. | Automatic root cause identification using most probable alignments | |
Al-Ahmari | A methodology for selection and evaluation of advanced manufacturing technologies | |
Carmona et al. | Applying case based reasoning for prioritizing areas of business management | |
Bernat et al. | Modelling customer lifetime value in a continuous, non-contractual time setting | |
CN113377640B (en) | Method, medium, device and computing equipment for explaining model under business scene | |
CN113128739A (en) | Prediction method of user touch time, prediction model training method and related device | |
Vasantham et al. | Combination of Scrum Lean-UX-based AI UX Design | |
JP6938259B2 (en) | Generation device, generation method, and generation program | |
CN105912832A (en) | Chronological change prediction system | |
Sumesh et al. | CEA Based Reasoning with the i∗ Framework | |
Nooyen | Predicting the occurrence of complaints within the customer journey based on process mining techniques |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |