CN109767261A - Products Show method, apparatus, computer equipment and storage medium - Google Patents
Products Show method, apparatus, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109767261A CN109767261A CN201811546833.2A CN201811546833A CN109767261A CN 109767261 A CN109767261 A CN 109767261A CN 201811546833 A CN201811546833 A CN 201811546833A CN 109767261 A CN109767261 A CN 109767261A
- Authority
- CN
- China
- Prior art keywords
- micro
- expression
- information
- obtains
- client
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000014509 gene expression Effects 0.000 claims abstract description 195
- 230000001815 facial effect Effects 0.000 claims abstract description 74
- 238000012549 training Methods 0.000 claims description 136
- 238000006243 chemical reaction Methods 0.000 claims description 41
- 238000003066 decision tree Methods 0.000 claims description 29
- 230000002996 emotional effect Effects 0.000 claims description 28
- 238000004590 computer program Methods 0.000 claims description 25
- 238000003062 neural network model Methods 0.000 claims description 18
- 238000005070 sampling Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 3
- 239000000047 product Substances 0.000 claims 38
- 230000005540 biological transmission Effects 0.000 claims 1
- 239000006227 byproduct Substances 0.000 claims 1
- 238000013473 artificial intelligence Methods 0.000 abstract 1
- 238000010801 machine learning Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 6
- 239000000284 extract Substances 0.000 description 6
- 238000004891 communication Methods 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 238000007637 random forest analysis Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000036651 mood Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010016275 Fear Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 210000004709 eyebrow Anatomy 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
This application involves the machine learning in artificial intelligence, a kind of Products Show method, apparatus, computer equipment and storage medium are provided.The described method includes: carrying out recognition of face by obtaining client's facial image according to client's facial image, obtaining recognition result, obtain customer information according to recognition result；Micro- expressive features are extracted according to client's facial image, micro- expressive features are input in the micro- Expression Recognition model trained, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- expression of client；Corresponding customer anger state is determined according to the micro- expression of client, client characteristics information is obtained according to customer anger state and customer information, in the Products Show model that client characteristics information input has been trained, obtain output of products vector, corresponding recommended products information is obtained according to output of products vector, recommended products information is sent into terminal.It can be improved the specific aim and accuracy of recommendation using this method.
Description
Technical field
This application involves Internet technical field, more particularly to a kind of Products Show method, apparatus, computer equipment and
Storage medium.
Background technique
With the development of current insurance industry, the type of insurance products is more and more, and the selectable content of client is increasingly
It is more, cause client that difficulty is selected to increase.At present.It is all after client's proposition demand, insurance business personnel pass through face-to-face with client
Communication exchange is carried out, determines the product that client needs, the recommendation that client relies primarily on business personnel selects.Insurance business
Personnel are when recommending, the problem of product often recommended and customer demand are not inconsistent, specific aim and accuracy all compared with
It is low, it is difficult to meet customer need.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of specific aim that can be improved Products Show and accurate
Products Show method, apparatus, computer equipment and the storage medium of property.
A kind of Products Show method, which comprises
Client's facial image is obtained, recognition of face is carried out according to client's facial image, obtains recognition result, is tied according to identification
Fruit obtains customer information；
Micro- expressive features are extracted according to client's facial image, micro- expressive features are input to the micro- Expression Recognition mould trained
In type, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- expression of client；
Corresponding customer anger state is determined according to the micro- expression of client, and visitor is obtained according to customer anger state and customer information
Family characteristic information in the Products Show model for having trained client characteristics information input, obtains output of products vector, according to product
Output vector obtains corresponding recommended products information, and recommended products information is sent terminal.
Corresponding customer anger state is determined according to the micro- expression of client in one of the embodiments, according to customer anger
State and customer information obtain client characteristics information, in the Products Show model that client characteristics information input has been trained, obtain
Output of products vector obtains corresponding recommended products information according to output of products vector, by recommended products information send terminal it
Afterwards, further includes:
Current time is obtained, by current time, customer anger state, customer information and the write-in of recommended products information association are pre-
If promotion and conversion template obtains target promotion and conversion, target promotion and conversion is saved.
Corresponding recommended products information is obtained according to output of products vector in one of the embodiments, by recommended products
Information sends terminal, comprising:
Corresponding recommended products information is obtained according to output of products vector, obtains the current recommended products letter that terminal is submitted
Recommended products information is sent terminal when recommended products information and current recommended products information are inconsistent by breath.
The generation step for the Products Show model trained in one of the embodiments, comprising:
Obtain history promotion and conversion, according to history promotion and conversion obtain historic customer information, historic customer emotional state and
Corresponding history recommended products information；
Historic customer characteristic information is extracted according to historic customer information and historic customer emotional state, recommends to produce according to history
Product information obtains historical product output vector；
Using historic customer characteristic information as the input of neural network model, using historical product output vector as nerve net
The label of network model is trained, when reaching preset condition, the Products Show model trained.
The generation step for the micro- Expression Recognition model trained in one of the embodiments, comprising:
Historic customer facial image and corresponding micro- expression are obtained, according to historic customer facial image and corresponding micro- expression
Obtain original training set；
It puts back to sampling at random from original training set, obtains target training set；
Corresponding micro- expressive features collection is obtained according to target training set, randomly selects the micro- expression in part from micro- expressive features collection
Feature obtains the micro- expressive features collection of target；
It obtains dividing micro- expressive features using gini index algorithm from target expression feature set, it is special using micro- expression is divided
Sign divides target training set, sub- training set is obtained, using sub- training set as target training set；
It returns and corresponding micro- expressive features collection is obtained according to target training set, it is micro- to randomly select part from micro- expressive features collection
Expressive features, the step of obtaining target micro- expressive features collection execution obtain the micro- expression decision of target when reaching preset condition
Tree；
The step of return puts back to sampling at random from original training set, obtains target training set execution, it is default when reaching
When the micro- expression decision tree of the target of number, micro- Expression Recognition model for having been trained.
A kind of Products Show device, described device include:
Face recognition module carries out recognition of face according to client's facial image, obtains for obtaining client's facial image
To recognition result, customer information is obtained according to the recognition result；
Micro- Expression Recognition module, it is for extracting micro- expressive features according to client's facial image, micro- expression is special
Sign is input in the micro- Expression Recognition model trained, and obtains micro- expression output feature, is obtained according to micro- expression output feature
To the micro- expression of client；
Recommended products obtains module, for determining corresponding customer anger state according to the micro- expression of the client, according to institute
It states customer anger state and the customer information obtains client characteristics information, the production that the client characteristics information input has been trained
In product recommended models, output of products vector is obtained, corresponding recommended products information is obtained according to the output of products vector, by institute
It states recommended products information and sends terminal.
Described device in one of the embodiments, further includes:
Report acquisition module obtains historic customer letter according to the history promotion and conversion for obtaining history promotion and conversion
Breath, historic customer emotional state and corresponding history recommended products information；
Extraction module, it is special for extracting historic customer according to the historic customer information and the historic customer emotional state
Reference breath, obtains historical product output vector according to history recommended products information；
Network training module, for using the historic customer characteristic information as the input of neural network model, by history
Output of products vector is trained as the label of the neural network model, when reaching preset condition, obtains described instructed
Experienced Products Show model.
The generation step of the micro- Expression Recognition model trained in one of the embodiments, comprising:
Initial sample obtains module, for obtaining historic customer facial image and corresponding micro- expression, according to the history
Client's facial image and corresponding micro- expression obtain original training set；
Training set obtains module, for putting back to sampling at random from the original training set, obtains target training set；
Feature set obtains module, for obtaining corresponding micro- expressive features collection according to the target training set, from described micro-
Expressive features collection randomly selects the micro- expressive features in part, obtains the micro- expressive features collection of target；
Sub- training set obtains module, micro- for obtaining dividing using gini index algorithm from the target expression feature set
Expressive features divide the target training set using the micro- expressive features of division, obtain sub- training set, by son training
Collection is used as target training set；
Decision tree obtains module, corresponding micro- expressive features collection is obtained according to the target training set for returning, from institute
The step of stating micro- expressive features collection and randomly select the micro- expressive features in part, obtaining target micro- expressive features collection execution, it is pre- when reaching
If when condition, obtaining the micro- expression decision tree of target；
Model obtains module, puts back to sampling at random from the original training set for returning, obtains target training set
The step of execute, when the micro- expression decision tree of the target for reaching preset number, obtain the micro- Expression Recognition model trained.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
Device performs the steps of when executing the computer program
Client's facial image is obtained, recognition of face is carried out according to client's facial image, obtains recognition result, is tied according to identification
Fruit obtains customer information；
Micro- expressive features are extracted according to client's facial image, micro- expressive features are input to the micro- Expression Recognition mould trained
In type, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- expression of client；
Corresponding customer anger state is determined according to the micro- expression of client, and visitor is obtained according to customer anger state and customer information
Family characteristic information in the Products Show model for having trained client characteristics information input, obtains output of products vector, according to product
Output vector obtains corresponding recommended products information, and recommended products information is sent terminal.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
It is performed the steps of when row
Client's facial image is obtained, recognition of face is carried out according to client's facial image, obtains recognition result, is tied according to identification
Fruit obtains customer information；
Micro- expressive features are extracted according to client's facial image, micro- expressive features are input to the micro- Expression Recognition mould trained
In type, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- expression of client；
Corresponding customer anger state is determined according to the micro- expression of client, and visitor is obtained according to customer anger state and customer information
Family characteristic information in the Products Show model for having trained client characteristics information input, obtains output of products vector, according to product
Output vector obtains corresponding recommended products information, and recommended products information is sent terminal.
The said goods recommended method, device, computer equipment and storage medium, by obtaining client's facial image, according to
Client's facial image carries out recognition of face, obtains recognition result, obtains customer information according to recognition result；According to client's face figure
As extracting micro- expressive features, micro- expressive features is input in the micro- Expression Recognition model trained, it is special to obtain micro- expression output
Sign exports feature according to micro- expression and obtains the micro- expression of client；Corresponding customer anger state is determined according to the micro- expression of client, according to
Customer anger state and customer information obtain client characteristics information, the Products Show model that client characteristics information input has been trained
In, output of products vector is obtained, corresponding recommended products information is obtained according to output of products vector, recommended products information is sent
Terminal.By obtaining the facial image of client in real time, customer information is obtained, the mood shape of client is obtained by micro- Expression Recognition
Then state obtains real-time recommended products information using Products Show model according to customer information and the emotional state of client, will
Recommended products information sends terminal, and terminal is shown, business personnel believes according to the recommended products that real-time customer data obtains
Breath is recommended, and can be improved the specific aim and accuracy of recommendation, meets customer need.
Detailed description of the invention
Fig. 1 is the application scenario diagram of Products Show method in one embodiment；
Fig. 2 is the flow diagram of Products Show method in one embodiment；
Fig. 3 is the flow diagram of training Products Show model in one embodiment；
Fig. 4 is the flow diagram of the micro- Expression Recognition model of training in one embodiment；
Fig. 5 is the structural block diagram of Products Show device in one embodiment；
Fig. 6 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Products Show method provided by the present application, can be applied in application environment as shown in Figure 1.Wherein, video is adopted
Acquisition means 102 are communicated by network with server 104, and server 104 is communicated with terminal 106 by network.Service
Device 104 obtains client's facial image that image collecting device 102 acquires, and carries out recognition of face according to client's facial image, obtains
Recognition result obtains customer information according to recognition result；Server 104 extracts micro- expressive features according to client's facial image, will
Micro- expressive features are input in the micro- Expression Recognition model trained, and are obtained micro- expression output feature, are exported according to micro- expression special
Obtain the micro- expression of client；Server 104 determines corresponding customer anger state according to the micro- expression of client, according to customer anger shape
State and customer information obtain client characteristics information, in the Products Show model that client characteristics information input has been trained, are produced
Product output vector obtains corresponding recommended products information according to output of products vector, and recommended products information is sent terminal 106.
Wherein, terminal 106 can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and portable
Wearable device, video acquisition device 10 may include camera, and server 104 can be either multiple with independent server
The server cluster of server composition is realized.
In one embodiment, as shown in Fig. 2, providing a kind of Products Show method, it is applied in Fig. 1 in this way
It is illustrated for server, comprising the following steps:
S202 obtains client's facial image, carries out recognition of face according to client's facial image, obtains recognition result, according to
Recognition result obtains customer information.
Wherein, client's facial image refers to the client's facial image obtained using image collecting device.Image collecting device
For acquiring facial image, which can be camera, video camera or scanning means etc..Customer information is for retouching
The information for stating client, essential information and client including client have bought the information of product.
Specifically, server can intercept video frame at interval of preset duration from the video client of real-time collecting in advance,
Video image is obtained, Face datection is carried out according to the video image and obtains client's facial image.It is carried out according to client's facial image
Recognition of face obtains recognition result, obtains customer information according to recognition result.Recognition of face is based on facial feature information of people
Carry out a kind of biological identification technology of identification.Image or video flowing containing face are acquired with video camera or camera, and
Automatic detection and tracking face in the picture, and then a series of the relevant technologies of face recognition are carried out to the face detected, lead to
Often it is also referred to as Identification of Images, face recognition.Recognition of face includes: Image Acquisition, Face datection, facial image pretreatment, face
Image characteristics extraction and facial image you match and identification.
S204 extracts micro- expressive features according to client's facial image, micro- expressive features is input to the micro- expression trained
In identification model, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- expression of client.
Wherein, micro- expressive features are used to reflect the feature of the face of customer image, comprising: upper face feature, lower face are special
Sign, mouth feature and other features etc..Micro- Expression Recognition model refers to according to historic customer facial image and client's face figure
As the micro- expression of corresponding history is trained to obtain using random forests algorithm, for carrying out the identification of micro- expression.Micro- expression
Output is characterized in the feature of micro- expression for describing to obtain.Micro- expression has micro- expression of 7 big types, including happiness type
Micro- expression, the micro- expression of frightened type, the micro- expression of sad type, fears that the micro- expression of type, pleasantly surprised type are micro- at the micro- expression of angry type
Expression and the micro- expression of natural type.
Specifically, server extracts micro- expressive features according to client's facial image, and micro- expressive features are input to and have been trained
Micro- Expression Recognition model in, obtain micro- expression output feature, server is according to the micro- expression output spy set in training
It seeks peace and presets the corresponding relationship of micro- expression and obtain the corresponding micro- expression of client of micro- expression output feature.Carrying out micro- Expression Recognition
When, each decision tree can all export the corresponding output feature of a micro- expression, and the micro- expression for counting all decision trees outputs is defeated
The accounting of feature out, the high output feature of accounting are that model exports feature.Such as 100 decision trees, wherein 98 output features
Be the calm corresponding output feature of expression, 2 output is the corresponding output feature of other expressions, then Expression Recognition model is known
Other result is the calm corresponding output feature of expression.
S206 determines corresponding customer anger state according to the micro- expression of client, according to customer anger state and customer information
Client characteristics information is obtained, in the Products Show model that client characteristics information input has been trained, obtains output of products vector, root
Corresponding recommended products information is obtained according to output of products vector, recommended products information is sent into terminal.
Wherein, customer anger state refers to the emotional state of existing customer, each micro- expression corresponding one sets in advance
The emotional state set.Client characteristics information refers to for describing the specific characteristic information of client, including client's current emotional shape
State characteristic information, client's essential information characteristic information and client have bought Product Feature Information etc..Products Show model is root
It is trained, is used for using neural network algorithm according to historic customer characteristic information and corresponding history recommended products information
Prediction to recommended products.The terminal is the corresponding terminal of business personnel, can be smart phone.
Specifically, server determines corresponding customer anger state according to the micro- expression of client, according to customer anger state and
Customer information obtains client characteristics information, and in the Products Show model that client characteristics information input has been trained, it is defeated to obtain product
Outgoing vector obtains corresponding recommended products information according to output of products vector, recommended products information is sent terminal, terminal is incited somebody to action
To recommended products information shown so that business personnel carries out Products Show to client according to the recommended products of display.Than
Such as, when the product currently recommended when client is dissatisfied, at this point, the emotional state of client can be it is half-believing, half-doubting, finally according to visitor
The new recommended products information that the emotional state at family and the essential information of client obtain, by new recommended products information terminal into
Row display, screen reminding business personnel carry out new recommendation.In one embodiment, the recommended products information obtained can have more
It is a, when in recommended products information including the product information currently recommended, then illustrate that the recommended products of business personnel meets client
Demand, when not including the product information currently recommended in recommended products information, then business personnel reselects recommended products information
In product recommended, meet customer need.
In the said goods recommended method, by obtaining client's facial image, recognition of face is carried out according to client's facial image,
Recognition result is obtained, customer information is obtained according to recognition result.Micro- expressive features are extracted according to client's facial image, by micro- expression
Feature is input in the micro- Expression Recognition model trained, and is obtained micro- expression output feature, is exported feature according to micro- expression and obtain
The micro- expression of client.Corresponding customer anger state is determined according to the micro- expression of client, is obtained according to customer anger state and customer information
To client characteristics information, in the Products Show model that client characteristics information input has been trained, output of products vector is obtained, according to
Output of products vector obtains corresponding recommended products information, recommended products information is sent terminal, business personnel is according to real-time
The recommended products information that customer data obtains is recommended, and be can be improved the specific aim and accuracy of recommendation, is met customer need.
In one embodiment, after step S206, i.e., corresponding customer anger shape is being determined according to the micro- expression of client
State obtains client characteristics information according to customer anger state and customer information, the product that client characteristics information input has been trained
In recommended models, output of products vector is obtained, corresponding recommended products information is obtained according to output of products vector, by recommended products
After information sends terminal, further comprise the steps of:
Current time is obtained, by current time, customer anger state, customer information and the write-in of recommended products information association are pre-
If promotion and conversion template obtains target promotion and conversion, target promotion and conversion is saved.
Wherein, presetting promotion and conversion template is pre-set for recording the report of recommended products relevant information.
Specifically, server obtain current time, by customer anger state, customer information and recommended products at this time giggle
You are associated in the promotion and conversion template that write-in is pre-set, and obtain target promotion and conversion, target promotion and conversion is saved in number
It subsequent check and uses according to facilitating in library.
In one embodiment, corresponding recommended products information is obtained according to output of products vector, by recommended products information
Send terminal, comprising:
Corresponding recommended products information is obtained according to output of products vector, obtains the current recommended products letter that terminal is submitted
Recommended products information is sent terminal when recommended products information and current recommended products information are inconsistent by breath.
Specifically, server gets the current recommended products information of terminal submission, which is business people
Before member is that client links up, the server submitted by terminal.After communication starts, server is according to output of products
Vector obtains corresponding recommended products information, the recommended products that server judges current recommended products information and is calculated at this time
Whether information is consistent, for example can be judged according to product number, can be carried out judging according to name of product etc..When pushing away
When recommending product information and inconsistent current recommended products information, the recommended products information being calculated is sent into terminal.Work as recommendation
When product information is consistent with current recommended products information, the suitable prompt of recommended products can be sent to terminal.It being capable of business people
Member is more easily handled according to obtained information, and the efficiency and accuracy of business personnel's recommended products are improved.
In one embodiment, recommended products information can also be converted to voice messaging, server is by the language after conversion
It is played out in the voice device for the business personnel that message breath is sent, so that business personnel obtains recommended products information.
In one embodiment, as shown in figure 3, the generation step for the Products Show model trained, comprising steps of
S302 obtains history promotion and conversion, obtains historic customer information, historic customer mood shape according to history promotion and conversion
State and corresponding history recommended products information.
Specifically, server obtains history promotion and conversion, obtains historic customer information, history visitor according to history promotion and conversion
Family emotional state and corresponding history recommended products information.
S304 extracts historic customer characteristic information according to historic customer information and historic customer emotional state, according to history
Recommended products information obtains historical product output vector.
Specifically, server extracts historic customer characteristic information according to historic customer information and historic customer emotional state,
Historical product output vector is obtained according to history recommended products information.
S306, using historic customer characteristic information as the input of neural network model, using historical product output vector as
The label of neural network model is trained, when reaching preset condition, the Products Show model trained.
Wherein, preset condition is the condition for being used to training of judgement and completing pre-set, including the number of iterations reaches most
The value of big number and evaluation function reaches preset threshold.
Specifically, server is using historic customer characteristic information as the input of neural network model, by historic customer feature
Input of the information as neural network model, when reaching preset condition, the Products Show model trained.Firstly, into
Row initialization, i.e., carry out random assignment to the weight of neural network and biasing, and select first input sample at random.Setting changes
For frequency n, to front transfer and the output of each hidden layer and output layer is found out.Output layer is found out (to mark with expected output
Label) deviation E, reverse propagated error finds out the error of each hidden layer.Adjust the weight and biasing of neuron.Judge sample
Whether this, which trains, terminates, if it is not, the number of iterations adds 1.If so, inputting next sample.Wherein, using mean square deviation is minimized, come
Error size is measured, formula:
In the above-described embodiments, by obtain history promotion and conversion, according to history promotion and conversion obtain historic customer information,
Historic customer emotional state and corresponding history recommended products information, mention according to historic customer information and historic customer emotional state
Historic customer characteristic information is taken, historical product output vector is obtained according to history recommended products information, historic customer feature is believed
The input as neural network model is ceased, is trained historical product output vector as the label of neural network model, when
When reaching preset condition, the Products Show model trained can train Products Show model in advance.Carrying out product
It can directly be used when recommendation, raising obtains the efficiency of recommended products.
In one embodiment, as shown in figure 4, the generation step for the micro- Expression Recognition model trained, comprising:
S402 obtains historic customer facial image and corresponding micro- expression, according to historic customer facial image and corresponding
Micro- expression obtains original training set.
Specifically, server obtains historic customer facial image and corresponding micro- expression.Wherein, historic customer facial image
It can be and obtained according in the communication video of history service personnel and client by recognition of face.According to historic customer facial image
The micro- expressive features of historic customer are obtained, the micro- expression of history is obtained according to corresponding micro- expression and exports feature, according to obtained history
The micro- expressive features of client and the micro- expression output feature of history obtain original training set.
S404 puts back to sampling at random from original training set, obtains target training set.
S406 obtains corresponding micro- expressive features collection according to target training set, randomly selects part from micro- expressive features collection
Micro- expressive features obtain the micro- expressive features collection of target.
Wherein, target training set refers to the set that the sample randomly selected from original training set obtains.Micro- expressive features
Collection refers to the set being made of micro- expressive features, is extracted by the video pictures in history video.Server is according to target
History facial image in training set carries out the face picture that Face datection obtains, obtains the micro- expression of history according to face picture
Feature obtains the micro- expressive features collection of history according to the micro- expressive features of obtained history, then concentrates institute from the micro- expressive features of history
And it chooses any number of feature and obtains the micro- expressive features collection of target.
S408 obtains dividing micro- expressive features using gini index algorithm from target expression feature set, micro- using dividing
Expressive features divide target training set, obtain sub- training set, using sub- training set as target training set.
Wherein, dividing expressive features is the feature for dividing the sample in target training set.For example, dividing expressive features
It raises up for eyebrow exterior angle, then the sample that the upper lip in target training set above mentions is divided into a sub- training set, target is instructed
The upper lip for practicing concentration above mentions the sample not mentioned above and is divided into another sub- training set.
Specifically, server is special using the optimal division expression of gini index algorithm picks according to the micro- expressive features collection of target
Sign, divides target training set using obtained division expressive features, obtains sub- training set.Using sub- training set as target
Training set, when the target training set is first node, which is root node, then a left side can be obtained after being divided
Training set and right training set all regard left training set and right training set as target training set.Gini index is indicated in sample set
In a sample chosen at random by the probability of misclassification.It chooses optimal division expressive features and refers to that selection gini index is the smallest
Feature is divided, is handled the feature two-value of the target training set according to the division feature.
S410 is returned and is obtained corresponding micro- expressive features collection according to target training set, randomly selects from micro- expressive features collection
The micro- expressive features in part, the step of obtaining target micro- expressive features collection execution obtain the micro- expression of target when reaching preset condition
Decision tree.
Wherein, preset condition refers to that micro- expression of sample in target training set is all same micro- expression.
Specifically, server judges whether to have reached preset condition, when not up to preset condition, i.e., in target training set
Micro- expression label of sample is not identical, and server return step S406 is executed at this time, until micro- table in target training set
When feelings label is all identical, i.e. when micro- expression label in each sub- training set is all identical, objective decision tree is just obtained.For example,
Obtained micro- expression label in a sub- training set is all uneasiness, then illustrates that the sub- training set label is identical, then son training
Collection has just reached preset condition, at this time, it is also necessary to judge whether other sub- training sets have reached preset condition, when all sons
Training set has all reached preset condition, has just obtained an objective decision tree at this time.
S412, return the step of putting back to sampling at random from original training set, obtaining target training set execution, when reaching
When the micro- expression decision tree of the target of preset number, micro- Expression Recognition model for having been trained.
Specifically, when obtaining objective decision tree, whether the quantity that server judges objective decision tree has reached default
Number just obtained micro- expression Random Forest model when the quantity of objective decision tree has reached preset number.Micro- expression
Random Forest model is just used as micro- Expression Recognition model.When the quantity of objective decision tree is not up to preset number, server is returned
It returns step S404 to be executed, until the quantity of objective decision tree has reached preset number.
In above-described embodiment, historic customer facial image and corresponding micro- expression are obtained, according to historic customer facial image
Original training set is obtained with corresponding micro- expression.It puts back to sampling at random from original training set, obtains target training set.According to
Target training set obtains corresponding micro- expressive features collection, randomly selects the micro- expressive features in part from micro- expressive features collection, obtains mesh
Mark micro- expressive features collection.It obtains dividing micro- expressive features using gini index algorithm from target expression feature set, uses division
Micro- expressive features divide target training set, obtain sub- training set, using sub- training set as target training set.Return to basis
Target training set obtains corresponding micro- expressive features collection, randomly selects the micro- expressive features in part from micro- expressive features collection, obtains mesh
The step of marking micro- expressive features collection execution obtains the micro- expression decision tree of target when reaching preset condition.It returns from initial sample
The step of putting back to sampling at random, obtain target training set execution is concentrated, when the micro- expression decision tree of the target for reaching preset number
When, micro- Expression Recognition model for having been trained.It can improve micro- expression directly using the micro- Expression Recognition model trained and know
Other efficiency.
In a specific application scenarios, insurance business person first passes through service terminal and submits the guarantor to be recommended to server
Dangerous product A information.When insurance business person when linking up face-to-face with client, passes through video camera and record video client information, service
Device receives the video client information of recording, extracts video frame according to 5 seconds intervals, obtains client's picture, carries out recognition of face,
Recognition result is obtained, the customer information of server preservation is obtained according to recognition result.Then micro- expression knowledge is carried out to client's picture
Not, the corresponding micro- expression of client's facial image is obtained, customer anger state is obtained according to micro- expression.At this point, server is according to visitor
Family emotional state and customer information calculate the insurance product information of recommendation using Products Show model, when the insurance products of recommendation
When including insurance products A in information, the suitable prompt of product is sent to the corresponding terminal of business personnel, when the insurance products of recommendation
When not including insurance products A in information, the insurance product information recommended is sent to the corresponding terminal of business personnel, guarantor can be submitted
The specific aim and accuracy of dangerous Products Show, meet customer need.And the service success rate of insurance business personnel is improved, meet industry
The demand of business personnel.
It should be understood that although each step in the flow chart of Fig. 2-4 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-4
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 5, providing a kind of Products Show device 500, comprising: face recognition module
502, micro- Expression Recognition module 504 and recommended products obtain module 506, in which:
Face recognition module 502 carries out recognition of face according to client's facial image, obtains for obtaining client's facial image
To recognition result, customer information is obtained according to recognition result；
Micro- Expression Recognition module 504 inputs micro- expressive features for extracting micro- expressive features according to client's facial image
Into the micro- Expression Recognition model trained, micro- expression output feature is obtained, feature is exported according to micro- expression and obtains the micro- table of client
Feelings；
Recommended products obtains module 506, for determining corresponding customer anger state according to the micro- expression of client, according to client
Emotional state and customer information obtain client characteristics information, in the Products Show model that client characteristics information input has been trained,
Output of products vector is obtained, corresponding recommended products information is obtained according to output of products vector, recommended products information is sent eventually
End.
In the said goods recommendation apparatus, client's facial image is identified to obtain visitor by face recognition module 502
Family information is identified micro- expression of client by client's facial image in micro- Expression Recognition module 504, obtains customer anger shape
State finally obtains in module 506 in recommended products, obtains corresponding recommended products by customer anger state and customer information and believes
Breath, can be improved the specific aim and accuracy of recommendation, meets customer need.
In one embodiment, Products Show device 500, further includes:
Report obtains module, for obtaining current time, by current time, customer anger state, customer information and recommendation
Default promotion and conversion template is written in product information association, obtains target promotion and conversion, target promotion and conversion is saved.
In one embodiment, recommended products obtains module 506, comprising:
Recommended products judgment module obtains terminal for obtaining corresponding recommended products information according to output of products vector
The current recommended products information submitted believes recommended products when recommended products information and current recommended products information are inconsistent
Breath sends terminal.
In one embodiment, Products Show device 500, comprising:
Historical report obtains module, for obtaining history promotion and conversion, obtains historic customer letter according to history promotion and conversion
Breath, historic customer emotional state and corresponding history recommended products information；
Feature obtains module, for extracting historic customer feature letter according to historic customer information and historic customer emotional state
Breath, obtains historical product output vector according to history recommended products information；
Model training module, for using historic customer characteristic information as the input of neural network model, by historical product
Output vector is trained as the label of neural network model, when reaching preset condition, the Products Show trained
Model.
In one embodiment, Products Show device 500, comprising:
Initial sample obtains module, for obtaining historic customer facial image and corresponding micro- expression, according to historic customer
Facial image and corresponding micro- expression obtain original training set；
Training set obtains module, for putting back to sampling at random from original training set, obtains target training set；
Feature set obtains module, for obtaining corresponding micro- expressive features collection according to target training set, from micro- expressive features
Collection randomly selects the micro- expressive features in part, obtains the micro- expressive features collection of target；
Sub- training set obtains module, divides micro- expression for obtaining from target expression feature set using gini index algorithm
Feature divides target training set using micro- expressive features are divided, obtains sub- training set, instructs sub- training set as target
Practice collection；
Decision tree obtains module, corresponding micro- expressive features collection is obtained according to target training set for returning, from micro- expression
The step of feature set randomly selects the micro- expressive features in part, obtains target micro- expressive features collection execution, when reaching preset condition,
Obtain the micro- expression decision tree of target；
Model obtains module, puts back to sampling at random from original training set for returning, obtains the step of target training set
It is rapid to execute, when the micro- expression decision tree of the target for reaching preset number, micro- Expression Recognition model for having been trained.
Specific about Products Show device limits the restriction that may refer to above for Products Show method, herein not
It repeats again.Modules in the said goods recommendation apparatus can be realized fully or partially through software, hardware and combinations thereof.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing recommended products related data.The network interface of the computer equipment is used for and external terminal
It is communicated by network connection.To realize a kind of Products Show method when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 6, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with
Computer program, which performs the steps of when executing computer program obtains client's facial image, according to client's face
Image carries out recognition of face, obtains recognition result, obtains customer information according to recognition result；It is extracted according to client's facial image micro-
Micro- expressive features are input in the micro- Expression Recognition model trained by expressive features, micro- expression output feature are obtained, according to micro-
Expression output feature obtains the micro- expression of client；Corresponding customer anger state is determined according to the micro- expression of client, according to customer anger
State and customer information obtain client characteristics information, in the Products Show model that client characteristics information input has been trained, obtain
Output of products vector obtains corresponding recommended products information according to output of products vector, and recommended products information is sent terminal.
In one embodiment, acquisition current time is also performed the steps of when processor executes computer program, will be worked as
Default promotion and conversion template is written in preceding time, customer anger state, customer information and recommended products information association, obtains target and pushes away
Report is recommended, target promotion and conversion is saved.
In one embodiment, it also performs the steps of when processor executes computer program according to output of products vector
Obtain corresponding recommended products information, obtain the current recommended products information that terminal is submitted, when recommended products information with ought be pushed forward
Recommend product information it is inconsistent when, by recommended products information send terminal.
In one embodiment, it is also performed the steps of when processor executes computer program and obtains history promotion and conversion,
Historic customer information, historic customer emotional state and corresponding history recommended products information are obtained according to history promotion and conversion；Root
Historic customer characteristic information is extracted according to historic customer information and historic customer emotional state, is obtained according to history recommended products information
Historical product output vector；Using historic customer characteristic information as the input of neural network model, by historical product output vector
Label as neural network model is trained, when reaching preset condition, the Products Show model trained.
In one embodiment, it is also performed the steps of when processor executes computer program and obtains historic customer face
Image and corresponding micro- expression obtain original training set according to historic customer facial image and corresponding micro- expression；From initial sample
This concentration puts back to sampling at random, obtains target training set；Corresponding micro- expressive features collection is obtained according to target training set, from micro-
Expressive features collection randomly selects the micro- expressive features in part, obtains the micro- expressive features collection of target；It is used from target expression feature set
Gini index algorithm obtains dividing micro- expressive features, is divided using micro- expressive features are divided to target training set, obtains son
Training set, using sub- training set as target training set；It returns and corresponding micro- expressive features collection is obtained according to target training set, from micro-
The step of expressive features collection randomly selects the micro- expressive features in part, obtains target micro- expressive features collection execution, when reaching default item
When part, the micro- expression decision tree of target is obtained；Sampling is put back in return at random from original training set, obtains the step of target training set
It is rapid to execute, when the micro- expression decision tree of the target for reaching preset number, micro- Expression Recognition model for having been trained.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor obtains client's facial image, carries out face knowledge according to client's facial image
Not, recognition result is obtained, customer information is obtained according to recognition result；Micro- expressive features are extracted according to client's facial image, it will be micro-
Expressive features are input in the micro- Expression Recognition model trained, and are obtained micro- expression output feature, are exported feature according to micro- expression
Obtain the micro- expression of client；Corresponding customer anger state is determined according to the micro- expression of client, is believed according to customer anger state and client
Breath obtains client characteristics information, in the Products Show model that client characteristics information input has been trained, obtains output of products vector,
Corresponding recommended products information is obtained according to output of products vector, recommended products information is sent into terminal.
In one embodiment, acquisition current time is also performed the steps of when computer program is executed by processor, it will
Default promotion and conversion template is written in current time, customer anger state, customer information and recommended products information association, obtains target
Promotion and conversion saves target promotion and conversion.
In one embodiment, also performed the steps of when computer program is executed by processor according to output of products to
Corresponding recommended products information is measured, the current recommended products information that terminal is submitted is obtained, when recommended products information and currently
When recommended products information is inconsistent, recommended products information is sent into terminal.
In one embodiment, it is also performed the steps of when computer program is executed by processor and obtains history recommendation report
It accuses, historic customer information, historic customer emotional state and corresponding history recommended products information is obtained according to history promotion and conversion；
Historic customer characteristic information is extracted according to historic customer information and historic customer emotional state, is obtained according to history recommended products information
To historical product output vector；Using historic customer characteristic information as the input of neural network model, by historical product export to
It measures and is trained as the label of neural network model, when reaching preset condition, the Products Show model trained.
In one embodiment, it is also performed the steps of when computer program is executed by processor and obtains historic customer people
Face image and corresponding micro- expression obtain original training set according to historic customer facial image and corresponding micro- expression；From initial
Sampling is put back in sample set at random, obtains target training set；Corresponding micro- expressive features collection is obtained according to target training set, from
Micro- expressive features collection randomly selects the micro- expressive features in part, obtains the micro- expressive features collection of target；Make from target expression feature set
It is obtained dividing micro- expressive features with gini index algorithm, target training set is divided using micro- expressive features are divided, is obtained
Sub- training set, using sub- training set as target training set；It returns and corresponding micro- expressive features collection is obtained according to target training set, from
The step of micro- expressive features collection randomly selects the micro- expressive features in part, obtains target micro- expressive features collection execution, it is default when reaching
When condition, the micro- expression decision tree of target is obtained；Sampling is put back in return at random from original training set, obtains target training set
Step executes, when the micro- expression decision tree of the target for reaching preset number, micro- Expression Recognition model for having been trained.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of Products Show method, which comprises
Client's facial image is obtained, recognition of face is carried out according to client's facial image, recognition result is obtained, according to the knowledge
Other result obtains customer information；
Micro- expressive features are extracted according to client's facial image, micro- expressive features are input to the micro- expression trained and are known
In other model, micro- expression output feature is obtained, the micro- expression of client is obtained according to micro- expression output feature；
Corresponding customer anger state is determined according to the micro- expression of the client, is believed according to the customer anger state and the client
Breath obtains client characteristics information, in the Products Show model that the client characteristics information input has been trained, obtains output of products
Vector obtains corresponding recommended products information according to the output of products vector, and the recommended products information is sent terminal.
2. the method according to claim 1, wherein described determine corresponding client according to the micro- expression of the client
Emotional state obtains client characteristics information according to the customer anger state and the customer information, the client characteristics is believed
In the Products Show model that breath input has been trained, output of products vector is obtained, is obtained according to the output of products vector corresponding
Recommended products information, after recommended products information transmission terminal, further includes:
Current time is obtained, by the current time, the customer anger state, the customer information and recommended products letter
Default promotion and conversion template is written in breath association, obtains target promotion and conversion, and the target promotion and conversion is saved.
3. being produced the method according to claim 1, wherein obtaining corresponding recommendation according to the output of products vector
The recommended products information is sent terminal by product information, comprising:
Corresponding recommended products information is obtained according to the output of products vector, obtains the current recommended products letter that terminal is submitted
The recommended products information is sent institute when the recommended products information and the inconsistent current recommended products information by breath
State terminal.
4. the method according to claim 1, wherein the generation step of the Products Show model trained,
Include:
Obtain history promotion and conversion, according to the history promotion and conversion obtain historic customer information, historic customer emotional state and
Corresponding history recommended products information；
Historic customer characteristic information is extracted according to the historic customer information and the historic customer emotional state, is pushed away according to history
It recommends product information and obtains historical product output vector；
Using the historic customer characteristic information as the input of neural network model, using historical product output vector as the mind
Label through network model is trained, and when reaching preset condition, obtains the Products Show model trained.
5. the method according to claim 1, wherein the generation of the micro- Expression Recognition model trained walks
Suddenly, comprising:
Historic customer facial image and corresponding micro- expression are obtained, according to the historic customer facial image and corresponding micro- expression
Obtain original training set；
It puts back to sampling at random from the original training set, obtains target training set；
Corresponding micro- expressive features collection is obtained according to the target training set, it is micro- to randomly select part from micro- expressive features collection
Expressive features obtain the micro- expressive features collection of target；
It obtains dividing micro- expressive features using gini index algorithm from the target expression feature set, divides micro- table using described
Feelings feature divides the target training set, obtains sub- training set, using sub- training set as target training set；
It returns and corresponding micro- expressive features collection is obtained according to the target training set, randomly select portion from micro- expressive features collection
Point micro- expressive features, the step of obtaining target micro- expressive features collection execution obtain the micro- expression of target and determine when reaching preset condition
Plan tree；
The step of return puts back to sampling at random from the original training set, obtains target training set execution, it is default when reaching
When the micro- expression decision tree of the target of number, the micro- Expression Recognition model trained is obtained.
6. a kind of Products Show device, which is characterized in that described device includes:
Face recognition module carries out recognition of face according to client's facial image, is known for obtaining client's facial image
Not as a result, obtaining customer information according to the recognition result；
Micro- Expression Recognition module, it is for extracting micro- expressive features according to client's facial image, micro- expressive features are defeated
Enter into the micro- Expression Recognition model trained, obtain micro- expression output feature, visitor is obtained according to micro- expression output feature
The micro- expression in family；
Recommended products obtains module, for determining corresponding customer anger state according to the micro- expression of the client, according to the visitor
Family emotional state and the customer information obtain client characteristics information, and the product that the client characteristics information input has been trained is pushed away
It recommends in model, obtains output of products vector, corresponding recommended products information is obtained according to the output of products vector, is pushed away described
It recommends product information and sends terminal.
7. device according to claim 6, which is characterized in that described device, further includes:
Report acquisition module obtains historic customer information according to the history promotion and conversion, goes through for obtaining history promotion and conversion
History customer anger state and corresponding history recommended products information；
Extraction module, for extracting historic customer feature letter according to the historic customer information and the historic customer emotional state
Breath, obtains historical product output vector according to history recommended products information；
Network training module, for using the historic customer characteristic information as the input of neural network model, by historical product
Output vector is trained as the label of the neural network model, when reaching preset condition, obtains described trained
Products Show model.
8. device according to claim 6, which is characterized in that the generation of the micro- Expression Recognition model trained walks
Suddenly, comprising:
Initial sample obtains module, for obtaining historic customer facial image and corresponding micro- expression, according to the historic customer
Facial image and corresponding micro- expression obtain original training set；
Training set obtains module, for putting back to sampling at random from the original training set, obtains target training set；
Feature set obtains module, for obtaining corresponding micro- expressive features collection according to the target training set, from micro- expression
Feature set randomly selects the micro- expressive features in part, obtains the micro- expressive features collection of target；
Sub- training set obtains module, divides micro- expression for obtaining from the target expression feature set using gini index algorithm
Feature divides the target training set using the micro- expressive features of division, obtains sub- training set, sub- training set is made
For target training set；
Decision tree obtains module, according to the target training set obtains corresponding micro- expressive features collection for returning, from described micro-
The step of expressive features collection randomly selects the micro- expressive features in part, obtains target micro- expressive features collection execution, when reaching default item
When part, the micro- expression decision tree of target is obtained；
Model obtains module, puts back to sampling at random from the original training set for returning, obtains the step of target training set
It is rapid to execute, when the micro- expression decision tree of the target for reaching preset number, obtain the micro- Expression Recognition model trained.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 5 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 5 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811546833.2A CN109767261A (en) | 2018-12-18 | 2018-12-18 | Products Show method, apparatus, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811546833.2A CN109767261A (en) | 2018-12-18 | 2018-12-18 | Products Show method, apparatus, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109767261A true CN109767261A (en) | 2019-05-17 |
Family
ID=66451304
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811546833.2A Pending CN109767261A (en) | 2018-12-18 | 2018-12-18 | Products Show method, apparatus, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109767261A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298683A (en) * | 2019-05-22 | 2019-10-01 | 深圳壹账通智能科技有限公司 | Information popularization method, apparatus, equipment and medium based on micro- expression |
| CN110321477A (en) * | 2019-05-24 | 2019-10-11 | 平安科技（深圳）有限公司 | Information recommendation method, device, terminal and storage medium |
| CN110334669A (en) * | 2019-07-10 | 2019-10-15 | 深圳市华腾物联科技有限公司 | A kind of method and apparatus of morphological feature identification |
| CN110415048A (en) * | 2019-08-06 | 2019-11-05 | 即悟(上海)智能科技有限公司 | A kind of advertisement recommended method, equipment, system, computer equipment and storage medium |
| CN110415041A (en) * | 2019-07-31 | 2019-11-05 | 中国工商银行股份有限公司 | Recommended method, recommendation apparatus, equipment and storage medium |
| CN110458644A (en) * | 2019-07-05 | 2019-11-15 | 深圳壹账通智能科技有限公司 | A kind of information processing method and relevant device |
| CN110532971A (en) * | 2019-09-02 | 2019-12-03 | 京东方科技集团股份有限公司 | Image procossing and device, training method and computer readable storage medium |
| CN110648203A (en) * | 2019-09-19 | 2020-01-03 | 中国银行股份有限公司 | Bank product information pushing method and device |
| CN111179031A (en) * | 2019-12-23 | 2020-05-19 | 第四范式（北京）技术有限公司 | Training method, device and system for commodity recommendation model |
| CN111222044A (en) * | 2019-12-31 | 2020-06-02 | 深圳Tcl数字技术有限公司 | Information recommendation method and device based on emotion perception and storage medium |
| CN113010784A (en) * | 2021-03-17 | 2021-06-22 | 北京十一贝科技有限公司 | Method, apparatus, electronic device, and medium for generating prediction information |
| CN113393301A (en) * | 2021-06-30 | 2021-09-14 | 中国工商银行股份有限公司 | Product pushing method and device |
| CN113515702A (en) * | 2021-07-07 | 2021-10-19 | 北京百度网讯科技有限公司 | Content recommendation method, model training method, device, equipment and storage medium |
| CN113628008A (en) * | 2021-08-11 | 2021-11-09 | 中国工商银行股份有限公司 | Intelligent marketing method, device, equipment and medium |
| CN115375001A (en) * | 2022-07-11 | 2022-11-22 | 重庆旅游云信息科技有限公司 | Tourist emotion assessment method and device for scenic spot |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104202718A (en) * | 2014-08-05 | 2014-12-10 | 百度在线网络技术（北京）有限公司 | Method and device for providing information for user |
| CN104462468A (en) * | 2014-12-17 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Information supply method and device |
| CN107203897A (en) * | 2017-04-24 | 2017-09-26 | 广东数相智能科技有限公司 | A kind of evaluation method of Products Show degree, apparatus and system |
| CN107424019A (en) * | 2017-08-15 | 2017-12-01 | 京东方科技集团股份有限公司 | The art work based on Emotion identification recommends method, apparatus, medium and electronic equipment |
| CN108154398A (en) * | 2017-12-27 | 2018-06-12 | 广东欧珀移动通信有限公司 | Method for information display, device, terminal and storage medium |
| CN108564414A (en) * | 2018-04-23 | 2018-09-21 | 帷幄匠心科技（杭州）有限公司 | Method of Commodity Recommendation based on behavior under line and system |
-
2018
- 2018-12-18 CN CN201811546833.2A patent/CN109767261A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104202718A (en) * | 2014-08-05 | 2014-12-10 | 百度在线网络技术（北京）有限公司 | Method and device for providing information for user |
| CN104462468A (en) * | 2014-12-17 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Information supply method and device |
| WO2016095621A1 (en) * | 2014-12-17 | 2016-06-23 | 百度在线网络技术（北京）有限公司 | Information providing method, apparatus, and computer device |
| CN107203897A (en) * | 2017-04-24 | 2017-09-26 | 广东数相智能科技有限公司 | A kind of evaluation method of Products Show degree, apparatus and system |
| CN107424019A (en) * | 2017-08-15 | 2017-12-01 | 京东方科技集团股份有限公司 | The art work based on Emotion identification recommends method, apparatus, medium and electronic equipment |
| CN108154398A (en) * | 2017-12-27 | 2018-06-12 | 广东欧珀移动通信有限公司 | Method for information display, device, terminal and storage medium |
| CN108564414A (en) * | 2018-04-23 | 2018-09-21 | 帷幄匠心科技（杭州）有限公司 | Method of Commodity Recommendation based on behavior under line and system |
Non-Patent Citations (2)
| Title |
|---|
| 朱龙珠;宫立华;刘鲲鹏;杨菁;赵强;: "基于随机森林算法的投诉预警模型优化方法", 电力信息与通信技术, no. 08, 31 August 2018 (2018-08-31) * |
| 钟生海;邱玉辉;: "基于情感语义的个性化推荐研究", 西南师范大学学报(自然科学版), no. 04, 31 August 2011 (2011-08-31) * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298683A (en) * | 2019-05-22 | 2019-10-01 | 深圳壹账通智能科技有限公司 | Information popularization method, apparatus, equipment and medium based on micro- expression |
| CN110321477A (en) * | 2019-05-24 | 2019-10-11 | 平安科技（深圳）有限公司 | Information recommendation method, device, terminal and storage medium |
| WO2020238023A1 (en) * | 2019-05-24 | 2020-12-03 | 平安科技（深圳）有限公司 | Information recommendation method and apparatus, and terminal and storage medium |
| CN110458644A (en) * | 2019-07-05 | 2019-11-15 | 深圳壹账通智能科技有限公司 | A kind of information processing method and relevant device |
| CN110334669A (en) * | 2019-07-10 | 2019-10-15 | 深圳市华腾物联科技有限公司 | A kind of method and apparatus of morphological feature identification |
| CN110334669B (en) * | 2019-07-10 | 2021-06-08 | 深圳市华腾物联科技有限公司 | Morphological feature recognition method and equipment |
| CN110415041A (en) * | 2019-07-31 | 2019-11-05 | 中国工商银行股份有限公司 | Recommended method, recommendation apparatus, equipment and storage medium |
| CN110415048A (en) * | 2019-08-06 | 2019-11-05 | 即悟(上海)智能科技有限公司 | A kind of advertisement recommended method, equipment, system, computer equipment and storage medium |
| CN110532971B (en) * | 2019-09-02 | 2023-04-28 | 京东方科技集团股份有限公司 | Image processing apparatus, training method, and computer-readable storage medium |
| CN110532971A (en) * | 2019-09-02 | 2019-12-03 | 京东方科技集团股份有限公司 | Image procossing and device, training method and computer readable storage medium |
| US11961327B2 (en) | 2019-09-02 | 2024-04-16 | Boe Technology Group Co., Ltd. | Image processing method and device, classifier training method, and readable storage medium |
| CN110648203A (en) * | 2019-09-19 | 2020-01-03 | 中国银行股份有限公司 | Bank product information pushing method and device |
| CN111179031A (en) * | 2019-12-23 | 2020-05-19 | 第四范式（北京）技术有限公司 | Training method, device and system for commodity recommendation model |
| CN111179031B (en) * | 2019-12-23 | 2023-09-26 | 第四范式（北京）技术有限公司 | Training method, device and system for commodity recommendation model |
| CN111222044A (en) * | 2019-12-31 | 2020-06-02 | 深圳Tcl数字技术有限公司 | Information recommendation method and device based on emotion perception and storage medium |
| CN113010784A (en) * | 2021-03-17 | 2021-06-22 | 北京十一贝科技有限公司 | Method, apparatus, electronic device, and medium for generating prediction information |
| CN113010784B (en) * | 2021-03-17 | 2024-02-06 | 北京十一贝科技有限公司 | Method, apparatus, electronic device and medium for generating prediction information |
| CN113393301A (en) * | 2021-06-30 | 2021-09-14 | 中国工商银行股份有限公司 | Product pushing method and device |
| CN113515702A (en) * | 2021-07-07 | 2021-10-19 | 北京百度网讯科技有限公司 | Content recommendation method, model training method, device, equipment and storage medium |
| CN113628008A (en) * | 2021-08-11 | 2021-11-09 | 中国工商银行股份有限公司 | Intelligent marketing method, device, equipment and medium |
| CN115375001A (en) * | 2022-07-11 | 2022-11-22 | 重庆旅游云信息科技有限公司 | Tourist emotion assessment method and device for scenic spot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109767261A (en) | Products Show method, apparatus, computer equipment and storage medium | |
| Jain et al. | Hybrid deep neural networks for face emotion recognition | |
| US10303982B2 (en) | Systems and methods for machine learning enhanced by human measurements | |
| CN108304882B (en) | Image classification method and device, server, user terminal and storage medium | |
| CN109766766A (en) | Employee work condition monitoring method, device, computer equipment and storage medium | |
| CN109508638A (en) | Face Emotion identification method, apparatus, computer equipment and storage medium | |
| US11244202B2 (en) | System and method for generation of unseen composite data objects | |
| CN111028305A (en) | Expression generation method, device, equipment and storage medium | |
| Abd El Meguid et al. | Fully automated recognition of spontaneous facial expressions in videos using random forest classifiers | |
| CN109409198A (en) | AU detection model training method, AU detection method, device, equipment and medium | |
| US11908240B2 (en) | Micro-expression recognition method based on multi-scale spatiotemporal feature neural network | |
| CN109711874A (en) | User's portrait generation method, device, computer equipment and storage medium | |
| CN109886110A (en) | Micro- expression methods of marking, device, computer equipment and storage medium | |
| CN111523421B (en) | Multi-person behavior detection method and system based on deep learning fusion of various interaction information | |
| CN110458008A (en) | Method for processing video frequency, device, computer equipment and storage medium | |
| Liu et al. | A 3 GAN: an attribute-aware attentive generative adversarial network for face aging | |
| CN110418204B (en) | Video recommendation method, device, equipment and storage medium based on micro expression | |
| CN109816200A (en) | Task method for pushing, device, computer equipment and storage medium | |
| CN110705490B (en) | Visual emotion recognition method | |
| CN110148007A (en) | Merchandise news detection method, device, computer equipment and storage medium | |
| US20220237917A1 (en) | Video comparison method and apparatus, computer device, and storage medium | |
| CN113435335B (en) | Microscopic expression recognition method and device, electronic equipment and storage medium | |
| Ousmane et al. | Automatic recognition system of emotions expressed through the face using machine learning: Application to police interrogation simulation | |
| JP2018005638A (en) | Image recognition model learning device, image recognition unit, method and program | |
| CN113780145A (en) | Sperm morphology detection method, sperm morphology detection device, computer equipment and storage medium |
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 |