CN110321477A - Information recommendation method, device, terminal and storage medium - Google Patents
Information recommendation method, device, terminal and storage medium Download PDFInfo
- Publication number
- CN110321477A CN110321477A CN201910441455.XA CN201910441455A CN110321477A CN 110321477 A CN110321477 A CN 110321477A CN 201910441455 A CN201910441455 A CN 201910441455A CN 110321477 A CN110321477 A CN 110321477A
- Authority
- CN
- China
- Prior art keywords
- information
- emotion
- type
- user
- micro
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to intelligent recommendation technical field, a kind of information recommendation method, device, terminal and storage medium are provided.The information recommendation method includes: when specified content information is shown on the interface for detecting customer service terminal device, the facial image of the specified content information user of browsing before controlling the camera acquisition display screen of customer service terminal device, and face characteristic information is extracted from facial image；Face characteristic information is inputted into micro- Expression Recognition model and carries out micro- Expression analysis, obtains the type of emotion of user；Recommendation information associated with specified content information is inquired from database according to type of emotion, recommendation information is placed on the designated position at the interface of customer service terminal device and is shown.The present invention can analyze in real time whether user is interested in the information content currently browsed according to the type of emotion of user, to recommend the information content being closer to the current attention rate of user to it, realize and precisely recommend.
Description
Technical field
The present invention relates to intelligent recommendation technical fields more particularly to a kind of information recommendation method, device, terminal and storage to be situated between
Matter.
Background technique
In recent years, With the fast development of internet, information is just in explosive growth, how to be screened from a large amount of information
The interested content of user becomes the research emphasis of internet area out, and therefore, information recommendation technology is also achieved in recent years
Bigger progress can recommend interested information to user, to meet user demand.
When existing recommendation information, often by the historical viewings data of analysis user, according to historical viewings data screening
The interested information of user out, because the interest that historical viewings data can only characterize user's entirety is biased to, it is difficult to accurately understand user
Therefore the information content of current interest when pushing content information, is easy the attention rate current with user and differs increasing, in real time
The precision of push is poor.
Summary of the invention
The present invention provides a kind of information recommendation method, device, terminal and storage medium, when solving current recommendation information,
It is difficult to accurately understand the information content of user's current interest, the content information of the push attention rate current with user, which differs, to be added
Greatly, the poor problem of the precision pushed in real time.
To solve the above problems, the present invention adopts the following technical scheme:
The present invention provides a kind of information recommendation method, includes the following steps:
Content information is shown on the interface of customer service terminal device, when detecting that be currently at display state is in specified
When holding information, start the camera of customer service terminal device；
It controls the camera and acquires the browsing specified content information before the display screen of the customer service terminal device and use
The facial image at family, and face characteristic information is extracted from the facial image；
Face characteristic information is inputted into micro- Expression Recognition model and carries out micro- Expression analysis, obtains the mood class of the user
Type；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Recommendation information associated with the specified content information is inquired from database according to the type of emotion,
The recommendation information is placed on the designated position at the interface of the customer service terminal device and is shown；Wherein, described
The corresponding at least one recommendation information of each type of emotion.
In one embodiment, described that face characteristic information is inputted to the step that micro- Expression Recognition model carries out micro- Expression analysis
Suddenly, comprising:
Go out the corresponding first eigenvector of face characteristic information using micro- Expression Recognition model extraction；
First eigenvector second feature vector corresponding with the type of emotion in micro- expression data library is carried out apart from degree
Amount obtains the closely located second feature vector with the first eigenvector as target signature from micro- expression data library
Vector；
The moos index of each type of emotion of user is calculated according to the target feature vector, and it is corresponding to obtain maximum moos index
Type of emotion.
In one embodiment, the step of the moos index that each type of emotion of user is calculated according to the target feature vector
Suddenly, comprising:
Closely located target feature vector is added in the mood label of respective dimensions, the mark of each type of emotion is obtained
Label value；
The specific gravity for calculating the total label value of label value Zhan of each type of emotion obtains the corresponding mood of each type of emotion of the user
Index；Wherein, total label value is the sum of the label value of each type of emotion.
In one embodiment, before the corresponding type of emotion of the maximum moos index of the acquisition, further includes:
Judge whether maximum moos index is greater than preset value；
If so, obtaining the corresponding type of emotion of maximum moos index；
Otherwise the camera for starting customer service terminal device resurveys the facial image of the user.
In one embodiment, it is described inquired from database according to the type of emotion it is related to the specified content information
The step of recommendation information of connection, comprising:
When the type of emotion is glad, happy or satisfied active mood, from database inquiry with it is currently assigned
The similar recommendation information of content information；
Otherwise the recommendation information opposite with currently assigned content information is inquired from database.
It is in one embodiment, described to input face characteristic information before micro- Expression Recognition model carries out micro- Expression analysis,
Further include:
It obtains facial image sample set and each facial image sample corresponds to fixed sample type of emotion；
Convolutional neural networks model is trained using the facial image sample set and sample type of emotion, until receiving
When holding back, micro- Expression Recognition model is obtained.
In one embodiment, described to utilize the facial image sample set and sample type of emotion to convolutional neural networks mould
After type is trained, further includes:
Based on preset loss function, the loss of convolutional neural networks model is calculated；
When loss is higher than certain value, the weight ginseng connected between each node in the convolutional neural networks model is adjusted
Number, to convolutional neural networks model re -training, until obtaining optimal weight parameter.
A kind of information recommending apparatus provided by the invention, comprising:
Starting module is currently at display when detecting for showing content information on the interface of customer service terminal device
State is when specifying content information, to start the camera of customer service terminal device；
Control module, the browsing finger before the display screen of the customer service terminal device is acquired for controlling the camera
Determine the facial image of content information user, and extracts face characteristic information from the facial image；
Analysis module carries out micro- Expression analysis for face characteristic information to be inputted micro- Expression Recognition model, described in acquisition
The type of emotion of user；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Display module, it is associated with the specified content information for being inquired from database according to the type of emotion
The recommendation information is placed on the designated position at the interface of the customer service terminal device and shows by recommendation information
Show；Wherein, the corresponding at least one recommendation information of each type of emotion.
The present invention provides a kind of terminal, including memory and processor, is stored with computer-readable finger in the memory
It enables, when the computer-readable instruction is executed by the processor, so that the processor executes as above described in any item letters
The step of ceasing recommended method.
The present invention provides a kind of storage medium, is stored thereon with computer program, and the computer program is held by processor
When row, as above described in any item information recommendation methods are realized.
Compared with the existing technology, technical solution of the present invention at least has following advantage:
Information recommendation method provided by the invention works as inspection by showing content information on the interface of customer service terminal device
Measure be currently at display state be specified content information when, just start the camera of customer service terminal device, and described in control
Camera acquires the facial image of the browsing specified content information user before the display screen of the customer service terminal device, and from
Face characteristic information is extracted in the facial image；Then face characteristic information is inputted into micro- Expression Recognition model and carries out micro- table
Mutual affection analysis, obtains the type of emotion of the user；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks of training
Model；Recommendation associated with the specified content information is finally inquired from database according to the type of emotion to believe
Breath, the recommendation information is placed on the designated position at the interface of the customer service terminal device and is shown, thus root
It is whether interested in the information content currently browsed that user is analyzed in real time according to the type of emotion of user, to work as to its recommendation with user
The information content that preceding attention rate is closer to is realized and is precisely recommended.
Detailed description of the invention
Fig. 1 is the implementation environment figure of the information recommendation method provided in one embodiment of the invention；
Fig. 2 is a kind of embodiment flow diagram of information recommendation method of the present invention；
Fig. 3 is another embodiment flow diagram of information recommendation method of the present invention, is mainly illustrated according to the mood class
Type inquires the specific steps of recommendation information associated with the specified content information from database；
Fig. 4 is another embodiment flow diagram of information recommendation method of the present invention, is mainly illustrated face characteristic information
Input the specific steps that micro- Expression Recognition model carries out micro- Expression analysis；
Fig. 5 is a kind of embodiment module frame chart of information recommending apparatus of the present invention；
Fig. 6 is the internal structure block diagram of terminal in one embodiment of the invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described.
In some processes of the description in description and claims of this specification and above-mentioned attached drawing, contain according to
Multiple operations that particular order occurs, but it should be clearly understood that these operations can not be what appears in this article suitable according to its
Sequence is executed or is executed parallel, and the serial number of operation such as S11, S12 etc. be only used for distinguishing each different operation, serial number
It itself does not represent and any executes sequence.In addition, these processes may include more or fewer operations, and these operations can
To execute or execute parallel in order.It should be noted that the description such as " first " herein, " second ", is for distinguishing not
Same message, equipment, module etc., does not represent sequencing, does not also limit " first " and " second " and be different type.
It will appreciated by the skilled person that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or wirelessly coupling.It is used herein to arrange
Diction "and/or" includes one or more associated wholes for listing item or any cell and all combinations.
It will appreciated by the skilled person that unless otherwise defined, all terms used herein (including technology art
Language and scientific term), there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, it should be understood that have in the context of the prior art
The consistent meaning of meaning, and unless idealization or meaning too formal otherwise will not be used by specific definitions as here
To explain.
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description in which the same or similar labels are throughly indicated same or similar element or has same or like function
Element.Obviously, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, those skilled in the art's every other implementation obtained without creative efforts
Example, shall fall within the protection scope of the present invention.
Fig. 1 is the implementation environment figure of the information recommendation method provided in one embodiment, as shown in Figure 1, in the implementation ring
In border, including server 110, terminal 120, terminal 120 can be connect by network with server.Wherein, above-mentioned network can wrap
Include internet, 2G/3G/4G, wifi etc..
It should be noted that server 110 can be independent physical server or terminal, it is also possible to multiple physics clothes
The server cluster for device composition of being engaged in can be to provide the basic cloud computing service such as Cloud Server, cloud database, cloud storage and CDN
Cloud Server.
Terminal 120 can be smart phone, customer service terminal device, tablet computer, laptop, desktop computer, intelligence
Energy speaker, smartwatch etc., however, it is not limited to this.
Referring to Fig. 2, the present invention provides a kind of information recommendation methods, when solving current recommendation information, it is difficult to accurate
Understand the information content of user's current interest, the attention rate that the content information of push is current with user differs increasing, pushes away in real time
The poor problem of the precision sent.In a kind of wherein embodiment, the information recommendation method may include following steps:
S21, content information is shown on the interface of customer service terminal device, be currently at referring to for display state when detecting
When determining content information, start the camera of customer service terminal device；
In the present embodiment, can show plurality of kinds of contents information on the interface of customer service terminal device, as product information, advertisement,
The information such as audio-video, news.When user's browsing content information, it can arbitrarily switch the content information of displaying on interface, work as detection
When the content information currently browsed to user is specified content information, such as certain given ad, then start customer service terminal device
Camera.Wherein, the customer service terminal device can be the terminal devices such as banking terminal machine, Self-service Tickets, self-help inquiry apparatus.
In one embodiment, respective identification can be set for every content information, and by the mark of specified content information to arrange
The form of table is saved, when detecting the identification record of the content information shown on current customer service terminal device in lists
When, then it is assumed that the content information of current presentation is specified content information.
S22, the control camera acquire the browsing specified content letter before the display screen of the customer service terminal device
The facial image of user is ceased, and extracts face characteristic information from the facial image；
In the present embodiment, camera acquisition can control to browse the user of specified content information just on customer service terminal device
One or more facial image, then facial image is positioned, and all face characteristics are extracted from facial image
Information.Wherein, the face characteristic information includes the significant condition of the human face characteristic points such as eyes, nose, ear.
S23, face characteristic information is inputted to the micro- Expression analysis of micro- Expression Recognition model progress, obtains the mood of the user
Type；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Micro- expression is that the mankind attempt to constrain or when hiding real feelings, leakage it is very of short duration, be unable to autonomous control
Facial expression is the effective clue of lie identification.Therefore, it can obtain user's by micro- expression of analysis user's facial image
Type of emotion.It is closed as shown that user's corners of the mouth parts a little, eyes are micro- in facial image, when making the expression of smile, then it represents that user's
Type of emotion is glad, happy；When user's brows are tightly knitted, then it represents that the type of emotion of user is angry, detest.
In the present embodiment, all face characteristic informations extracted in facial image can be inputted micro- Expression Recognition mould
Type obtains type of emotion when user browses currently assigned content information using micro- Expression Recognition model analysis.Wherein, the feelings
Thread type includes the moods such as happy, surprised, detest, anger.Micro- Expression Recognition model is the base obtained after repetition training
In the deep learning model of convolutional neural networks, the type of emotion of user can be used to identify.
S24, recommendation associated with the specified content information is inquired from database according to the type of emotion
The recommendation information is placed on the designated position at the interface of the customer service terminal device and shows by information；Wherein,
The corresponding at least one recommendation information of each type of emotion.
In the present embodiment, it after the type of emotion for obtaining active user, can be inquired from database according to the type of emotion
When recommendation information associated with specified content information such as the type of emotion of active user is glad, then it represents that user couple
Specified content information is interested, then it is subsequent to its recommendation information when, can recommend similar with specified content information
Content information, and on the designated position at the interface that the recommendation information is placed on customer service terminal device, such as by recommendation
The brief introduction of information is shown in the top at interface, after user browses the brief introduction to recommendation information generate interest when, pass through by
The recommendation information is pulled down to the display that whole content information is carried out among interface.
Information recommendation method provided by the invention works as inspection by showing content information on the interface of customer service terminal device
Measure be currently at display state be specified content information when, just start the camera of customer service terminal device, and described in control
Camera acquires the facial image of the browsing specified content information user before the display screen of the customer service terminal device, and from
Face characteristic information is extracted in the facial image；Then face characteristic information is inputted into micro- Expression Recognition model and carries out micro- table
Mutual affection analysis, obtains the type of emotion of the user；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks of training
Model；Recommendation associated with the specified content information is finally inquired from database according to the type of emotion to believe
Breath, the recommendation information is placed on the designated position at the interface of the customer service terminal device and is shown, thus root
It is whether interested in the information content currently browsed that user is analyzed in real time according to the type of emotion of user, to work as to its recommendation with user
The information content that preceding attention rate is closer to is realized and is precisely recommended.
In one embodiment, described to be inquired from database according to the type of emotion as shown in figure 3, in step s 24
It the step of recommendation information associated with the specified content information, may particularly include:
S241, when the type of emotion is glad, happy or satisfied active mood, from database inquiry with it is current
The specified similar recommendation information of content information；
S242, the recommendation information opposite with currently assigned content information is otherwise inquired from database.
In the present embodiment, when getting the type of emotion of active user, if the type of emotion of active user be it is glad,
When happiness etc. indicates positive mood, then it represents that user is more satisfied to the specified content information currently browsed, pushes away to it
When recommending content information, then recommendation information similar with currently assigned content information is inquired from database.Conversely, when current
When the type of emotion of user is that anger, sadness etc. indicate passive mood, then it represents that user believes the specified content currently browsed
Breath is lost interest in, and when to its recommendation information, is then inquired from database and is pushed away with what currently assigned content information differed greatly
Recommend content information.
In order to better understand the technical program, it is illustrated by taking financial product information recommendation as an example below:
When user uses banking terminal machine, the financial product shown on random browser interface on banking terminal machine is produced
A certain specified financing type of financial product is shown when detecting accordingly to the favorable rating of financial product in life on terminating machine
When, then facial image when user browses the financial product can be acquired by camera, and face is extracted from facial image
Then characteristic information analyzes face characteristic information by micro- Expression Recognition model, obtains the type of emotion of user, when working as
When the type of emotion of preceding user is glad, then other financing type of financial similar with specified financial product can be recommended to produce to it
Product.When the type of emotion of active user is to detest, then it can recommend debt-credit class product to it, thus according to the real-time feelings of user
The accurate recommendation of thread type realization content information.
In one embodiment, described that face characteristic information is inputted into micro- Expression Recognition mould as shown in figure 4, in step S23
Type carries out the step of micro- Expression analysis, may particularly include:
S231, go out the corresponding first eigenvector of face characteristic information using micro- Expression Recognition model extraction；
In the present embodiment, when extracting face characteristic information from facial image, facial image center can first be extracted
The face characteristic information in domain, to be positioned to facial image, then according to the face characteristic information of facial image central area
The face characteristic information in other regions of facial image is successively extracted, it is finally all to what is extracted using micro- Expression Recognition model
Face characteristic information carries out geometrical characteristic vector construction, so that it is mapped to corresponding first eigenvector, it is special to improve first
Levy the extraction rate of vector.
S232, by corresponding with the type of emotion in the micro- expression data library second feature vector of first eigenvector carry out away from
From measurement, the closely located second feature vector with the first eigenvector is obtained from micro- expression data library as target
Feature vector；
The present embodiment also needs to obtain the corresponding second feature vector of all type of emotion, saves it in micro- expression data library
In, it, can be by the mood in all first eigenvectors of facial image and micro- expression data library when carrying out distance metric calculating
The corresponding second feature vector of type carries out distance metric, to obtain the fisrt feature with facial image from micro- expression data library
Several second feature vectors similar in vector distance, obtain target feature vector, so that the user emotion can be characterized by filtering out
The feature vector of type.Wherein, micro- expression data inventory contains the corresponding second feature vector of each type of emotion.
S233, the moos index that each type of emotion of user is calculated according to the target feature vector obtain maximum mood and refer to
The corresponding type of emotion of number.
In the present embodiment, can the type of emotion according to belonging to each target feature vector, calculate each mood class of the user
The corresponding moos index of type, when the target feature vector of affiliated type of emotion is more, then the corresponding mood of the type of emotion refers to
Number is higher, and obtains the corresponding type of emotion of maximum moos index.For example, after distance metric, obtained target signature to
Amount is Y1, Y2, Y3, Y4, Y5, Y6, Y7, in these target feature vectors, belong to glad target feature vector have Y1, Y2,
Y3 and Y4, then glad corresponding moos index can be expressed as 4, and belonging to surprised target feature vector is Y5 and Y6, then surprised
Corresponding moos index can be expressed as 2, and the target feature vector for belonging to detest is Y7, then detesting corresponding moos index can be with
It is expressed as 1, to obtain the corresponding moos index of all type of emotion of the user, and the maximum moos index of the user can be obtained
It is 4, the corresponding type of emotion of the moos index is happiness, i.e., when the specified content information that the user currently browses is financing eka-gold
When melting product, then it represents that the user is interested in such financial product, then can inquire similar with such financial product
Type of financial Products Show of managing money matters is to user.
In one embodiment, described that each type of emotion of user is calculated according to the target feature vector in step S233
Moos index the step of, may particularly include:
Closely located target feature vector is added in the mood label of respective dimensions, the mark of each type of emotion is obtained
Label value；
The specific gravity for calculating the total label value of label value Zhan of each type of emotion obtains the corresponding mood of each type of emotion of the user
Index；Wherein, total label value is the sum of the label value of each type of emotion.
In the present embodiment, can analyze to obtain several dimensions type of emotion and its corresponding mood label, it is such as happy,
Similar or closely located target feature vector is added to corresponding dimension by the mood label of the dimensions such as sad, worried, irritated
In mood label, the label value of each type of emotion is obtained.For example, obtained target feature vector has 7 after distance metric
A, respectively Y1, Y2, Y3, Y4, Y5, Y6, Y7, total label value are 7.In these target feature vectors, belong to glad mesh
Mark feature vector has 4, i.e. Y1, Y2, Y3 and Y4, and corresponding label value is 4, then glad corresponding moos index can also indicate
ForI.e. 57%；Belonging to surprised target feature vector has 2, i.e. Y5 and Y6, and corresponding label value is 2, then surprised corresponding
Moos index can also be expressed asI.e. 28.6%；The target feature vector for belonging to detest has 1, i.e. Y7, corresponding label value
It is 1, then detesting corresponding moos index can also be expressed asI.e. 14.4%.Wherein, maximum moos index is 57%.
In one embodiment, in step S233, before the corresponding type of emotion of the maximum moos index of the acquisition, may be used also
Include:
Judge whether maximum moos index is greater than preset value；
If so, obtaining the corresponding type of emotion of maximum moos index；
Otherwise the camera for starting customer service terminal device resurveys the facial image of the user.
When carrying out mood analysis to user, it is possible that the mood of user is biased to unconspicuous situation, thus can not
The type of emotion for accurately obtaining the user, when leading to recommendation information, recommend precision reduce the case where.Therefore, in this implementation
In example, when maximum moos index is calculated, it can further judge whether maximum moos index is greater than preset value, when maximum feelings
When thread index is greater than preset value, then it represents that the user is biased to apparent mood, and then it is corresponding to obtain maximum moos index
Type of emotion.When maximum moos index is lower than preset value, then it represents that the user does not have apparent mood to be biased to, it is therefore desirable to logical
The camera for crossing starting customer service terminal device resurveys the facial image of the user, until obtaining more apparent mood class
Type.
In one embodiment, described that face characteristic information is inputted into the micro- table of micro- Expression Recognition model progress in step S23
Before mutual affection analysis, it may also include that
It obtains facial image sample set and each facial image sample corresponds to fixed sample type of emotion；
Convolutional neural networks model is trained using the facial image sample set and sample type of emotion, until receiving
When holding back, micro- Expression Recognition model is obtained.
In the present embodiment, fixed sample mood is corresponded to using facial image sample set and each facial image sample
Convolutional neural networks model is trained, micro- Expression Recognition model is obtained.When training, the quantity of facial image sample set is got over
More, then trained effect is better, and the precision of analysis of obtained micro- Expression Recognition model is also higher.
In one embodiment, described to utilize the facial image sample set and sample type of emotion to convolutional neural networks mould
After type is trained, it may also include that
Based on preset loss function, the loss of convolutional neural networks model is calculated；
When loss is higher than certain value, the weight ginseng connected between each node in the convolutional neural networks model is adjusted
Number, to convolutional neural networks model re -training, until obtaining optimal weight parameter.
In the present embodiment, each base of the convolutional neural networks model includes several nodes, base with
Node between base is in a kind of state connected entirely, and the connection between node usually has a weight parameter.Right
Before convolutional neural networks model is trained, the weight parameter between node is the parameter value being arbitrarily arranged.To convolution mind
When being trained through network model, the face template image pattern collection of magnanimity can be inputted in convolutional neural networks model, so
The loss for calculating convolutional neural networks model afterwards, judges whether its loss is greater than preset value, if so, adjusting the convolutional Neural
Then the weight parameter connected between each node in network model calculates its loss to model re -training, until loss is less than
Or it is equal to preset value, the optimal weight parameter connected between each node in convolutional neural networks model is obtained, to be trained
Qualified micro- Expression Recognition model.
Referring to FIG. 5, the embodiment of the present invention also provides a kind of information recommending apparatus, in a kind of the present embodiment, including open
Dynamic model block 31, control module 32, analysis module 33 and display module 34.Wherein,
Starting module 31, for showing content information on the interface of customer service terminal device, when detect be currently at it is aobvious
When show state is specified content information, start the camera of customer service terminal device；
In the present embodiment, can show plurality of kinds of contents information on the interface of customer service terminal device, as product information, advertisement,
The information such as audio-video, news.When user's browsing content information, it can arbitrarily switch the content information of displaying on interface, work as detection
When the content information currently browsed to user is specified content information, such as certain given ad, then start customer service terminal device
Camera.Wherein, the customer service terminal device can be the terminal devices such as banking terminal machine, Self-service Tickets, self-help inquiry apparatus.
In one embodiment, respective identification can be set for every content information, and by the mark of specified content information to arrange
The form of table is saved, when detecting the identification record of the content information shown on current customer service terminal device in lists
When, then it is assumed that the content information of current presentation is specified content information.
Control module 32, described in the browsing before the display screen of the customer service terminal device is acquired for controlling the camera
The facial image of specified content information user, and face characteristic information is extracted from the facial image；
In the present embodiment, camera acquisition can control to browse the user of specified content information just on customer service terminal device
One or more facial image, then facial image is positioned, and all face characteristics are extracted from facial image
Information.Wherein, the face characteristic information includes the significant condition of the human face characteristic points such as eyes, nose, ear.
Analysis module 33 carries out micro- Expression analysis for face characteristic information to be inputted micro- Expression Recognition model, obtains institute
State the type of emotion of user；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Micro- expression is that the mankind attempt to constrain or when hiding real feelings, leakage it is very of short duration, be unable to autonomous control
Facial expression is the effective clue of lie identification.Therefore, it can obtain user's by micro- expression of analysis user's facial image
Type of emotion.It is closed as shown that user's corners of the mouth parts a little, eyes are micro- in facial image, when making the expression of smile, then it represents that user's
Type of emotion is glad, happy；When user's brows are tightly knitted, then it represents that the type of emotion of user is angry, detest.
In the present embodiment, all face characteristic informations extracted in facial image can be inputted micro- Expression Recognition mould
Type obtains type of emotion when user browses currently assigned content information using micro- Expression Recognition model analysis.Wherein, the feelings
Thread type includes the moods such as happy, surprised, detest, anger.Micro- Expression Recognition model is the base obtained after repetition training
In the deep learning model of convolutional neural networks, the type of emotion of user can be used to identify.
Display module 34, it is associated with the specified content information for being inquired from database according to the type of emotion
Recommendation information, the recommendation information is placed on the designated position at the interface of the customer service terminal device and is carried out
Display；Wherein, the corresponding at least one recommendation information of each type of emotion.
In the present embodiment, it after the type of emotion for obtaining active user, can be inquired from database according to the type of emotion
When recommendation information associated with specified content information such as the type of emotion of active user is glad, then it represents that user couple
Specified content information is interested, then it is subsequent to its recommendation information when, can recommend similar with specified content information
Content information, and on the designated position at the interface that the recommendation information is placed on customer service terminal device, such as by recommendation
The brief introduction of information is shown in the top at interface, after user browses the brief introduction to recommendation information generate interest when, pass through by
The recommendation information is pulled down to the display that whole content information is carried out among interface.
Information recommending apparatus provided by the invention works as inspection by showing content information on the interface of customer service terminal device
Measure be currently at display state be specified content information when, just start the camera of customer service terminal device, and described in control
Camera acquires the facial image of the browsing specified content information user before the display screen of the customer service terminal device, and from
Face characteristic information is extracted in the facial image；Then face characteristic information is inputted into micro- Expression Recognition model and carries out micro- table
Mutual affection analysis, obtains the type of emotion of the user；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks of training
Model；Recommendation associated with the specified content information is finally inquired from database according to the type of emotion to believe
Breath, the recommendation information is placed on the designated position at the interface of the customer service terminal device and is shown, thus root
It is whether interested in the information content currently browsed that user is analyzed in real time according to the type of emotion of user, to work as to its recommendation with user
The information content that preceding attention rate is closer to is realized and is precisely recommended.
In one embodiment, the analysis module 33 is also configured to
Go out the corresponding first eigenvector of face characteristic information using micro- Expression Recognition model extraction；
First eigenvector second feature vector corresponding with the type of emotion in micro- expression data library is carried out apart from degree
Amount obtains the closely located second feature vector with the first eigenvector as target signature from micro- expression data library
Vector；Wherein, micro- expression data inventory contains the corresponding second feature vector of each type of emotion；
The moos index of each type of emotion of user is calculated according to the target feature vector, and it is corresponding to obtain maximum moos index
Type of emotion.
In one embodiment, the analysis module 33 is also configured to
Closely located target feature vector is added in the mood label of respective dimensions, the mark of each type of emotion is obtained
Label value；
The specific gravity for calculating the total label value of label value Zhan of each type of emotion obtains the corresponding mood of each type of emotion of the user
Index；Wherein, total label value is the sum of the label value of each type of emotion.
In one embodiment, the analysis module 33 is also configured to:
Judge whether maximum moos index is greater than preset value；
If so, obtaining the corresponding type of emotion of maximum moos index；
Otherwise the camera for starting customer service terminal device resurveys the facial image of the user.
In one embodiment, the display module 34 is also configured to
When the type of emotion is glad, happy or satisfied active mood, from database inquiry with it is currently assigned
The similar recommendation information of content information；
Otherwise the recommendation information opposite with currently assigned content information is inquired from database.
In one embodiment, the information recommending apparatus further include:
Module is obtained, corresponds to fixed sample mood class for obtaining facial image sample set and each facial image sample
Type；
Training module, for using the facial image sample set and sample type of emotion to convolutional neural networks model into
Row training, until obtaining micro- Expression Recognition model when convergence.
In one embodiment, the training module is also configured to
Based on preset loss function, the loss of convolutional neural networks model is calculated；
When loss is higher than certain value, the weight ginseng connected between each node in the convolutional neural networks model is adjusted
Number, to convolutional neural networks model re -training, until obtaining optimal weight parameter.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method
Embodiment in be described in detail, no detailed explanation will be given here.
A kind of terminal provided by the invention, including memory and processor are stored in the memory computer-readable
Instruction, when the computer-readable instruction is executed by the processor, so that processor execution is as above described in any item
The step of information recommendation method.
In one embodiment, the terminal is a kind of computer equipment, as shown in Figure 6.Computer described in the present embodiment
Equipment can be the equipment such as server, personal computer and the network equipment.The computer equipment includes processor 402, deposits
The devices such as reservoir 403, input unit 404 and display unit 405.It will be understood by those skilled in the art that the equipment shown in Fig. 4
Structure devices do not constitute the restriction to all devices, may include components more more or fewer than diagram, or combine certain
Component.Memory 403 can be used for storing computer program 401 and each functional module, and the operation of processor 402 is stored in memory
403 computer program 401, thereby executing the various function application and data processing of equipment.Memory can be interior storage
Device or external memory, or including both built-in storage and external memory.Built-in storage may include read-only memory (ROM),
Programming ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory or
Random access memory.External memory may include hard disk, floppy disk, ZIP disk, USB flash disk, tape etc..Memory packet disclosed in this invention
Include but be not limited to the memory of these types.Memory disclosed in this invention is only used as example rather than as restriction.
Input unit 404 is used to receive the input of signal, and receives the keyword of user's input.Input unit 404 can
Including touch panel and other input equipments.Touch panel collects the touch operation of user on it or nearby and (for example uses
Family uses the operations of any suitable object or attachment on touch panel or near touch panel such as finger, stylus), and root
According to the corresponding attachment device of preset driven by program；Other input equipments can include but is not limited to physical keyboard, function
One of key (such as broadcasting control button, switch key etc.), trace ball, mouse, operating stick etc. are a variety of.Display unit
405 can be used for showing the information of user's input or be supplied to the information of user and the various menus of computer equipment.Display is single
The forms such as liquid crystal display, Organic Light Emitting Diode can be used in member 405.Processor 402 is the control centre of computer equipment, benefit
With the various pieces of various interfaces and the entire computer of connection, by running or executing the software being stored in memory 402
Program and/or module, and the data being stored in memory are called, perform various functions and handle data.
As one embodiment, the computer equipment includes: one or more processors 402, memory 403, and one
Or multiple computer programs 401, wherein one or more of computer programs 401 are stored in memory 403 and are matched
It is set to and is executed by one or more of processors 402, one or more of computer programs 401 are configured to carry out above
Information recommendation method described in embodiment.
In one embodiment, the invention also provides a kind of storage medium for being stored with computer-readable instruction, the meters
When calculation machine readable instruction is executed by one or more processors, so that one or more processors execute above- mentioned information recommendation side
Method.It is deposited for example, the storage medium can be ROM, random access memory (RAM), CD-ROM, tape, floppy disk and light data
Store up equipment etc..
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, which can be stored in a storage medium, the program
When being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, storage medium above-mentioned can for magnetic disk, CD, only
Read non-volatile memory mediums or random access memory (Random such as storage memory (Read-Only Memory, ROM)
Access Memory, RAM) etc..
Based on the above embodiments it is found that the maximum beneficial effect of the present invention is:
Information recommendation method, device, terminal and storage medium provided by the invention, by the interface of customer service terminal device
Upper displaying content information, when detect be currently at display state be specified content information when, just starting customer service terminal device
Camera, and control the camera and acquire the browsing specified content information before the display screen of the customer service terminal device
The facial image of user, and face characteristic information is extracted from the facial image；Then face characteristic information is inputted micro-
Expression Recognition model carries out micro- Expression analysis, obtains the type of emotion of the user；Wherein, micro- Expression Recognition model is instruction
Practice qualified convolutional neural networks model；It is finally inquired from database according to the type of emotion and the specified content information
The recommendation information is placed on the designated position at the interface of the customer service terminal device by associated recommendation information
On shown, so that whether analyze user in real time according to the type of emotion of user interested in the information content currently browsed,
Recommend the information content being closer to the current attention rate of user to it with subsequent, realizes and precisely recommend.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (10)
1. a kind of information recommendation method characterized by comprising
Content information is shown on the interface of customer service terminal device, when detecting that be currently at display state is specified content letter
When breath, start the camera of customer service terminal device；
It controls the camera and acquires the browsing specified content information user's before the display screen of the customer service terminal device
Facial image, and face characteristic information is extracted from the facial image；
Face characteristic information is inputted into micro- Expression Recognition model and carries out micro- Expression analysis, obtains the type of emotion of the user；Its
In, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Recommendation information associated with the specified content information is inquired from database according to the type of emotion, by institute
It states and is shown on the designated position at the interface that recommendation information is placed on the customer service terminal device；Wherein, described each
The corresponding at least one recommendation information of type of emotion.
2. information recommendation method according to claim 1, which is characterized in that described that face characteristic information is inputted micro- expression
Identification model carries out the step of micro- Expression analysis, comprising:
Go out the corresponding first eigenvector of face characteristic information using micro- Expression Recognition model extraction；
First eigenvector second feature vector corresponding with the type of emotion in micro- expression data library is subjected to distance metric, from
The closely located second feature vector with the first eigenvector is obtained in micro- expression data library as target feature vector；
The moos index of each type of emotion of user is calculated according to the target feature vector, obtains the corresponding feelings of maximum moos index
Thread type.
3. information recommendation method according to claim 2, which is characterized in that described to be calculated according to the target feature vector
The step of moos index of each type of emotion of user, comprising:
Closely located target feature vector is added in the mood label of respective dimensions, the label of each type of emotion is obtained
Value；
The specific gravity for calculating the total label value of label value Zhan of each type of emotion obtains the corresponding mood of each type of emotion of the user and refers to
Number；Wherein, total label value is the sum of the label value of each type of emotion.
4. information recommendation method according to claim 2, which is characterized in that the corresponding feelings of the maximum moos index of the acquisition
Before thread type, further includes:
Judge whether maximum moos index is greater than preset value；
If so, obtaining the corresponding type of emotion of maximum moos index；
Otherwise the camera for starting customer service terminal device resurveys the facial image of the user.
5. information recommendation method according to claim 1, which is characterized in that it is described according to the type of emotion from database
The step of middle inquiry associated with specified content information recommendation information, comprising:
When the type of emotion is glad, happy or satisfied active mood, inquiry and currently assigned content from database
The similar recommendation information of information；
Otherwise the recommendation information opposite with currently assigned content information is inquired from database.
6. information recommendation method according to claim 1, which is characterized in that described that face characteristic information is inputted micro- expression
Identification model carries out before micro- Expression analysis, further includes:
It obtains facial image sample set and each facial image sample corresponds to fixed sample type of emotion；
Convolutional neural networks model is trained using the facial image sample set and sample type of emotion, until convergence
When, obtain micro- Expression Recognition model.
7. information recommendation method according to claim 6, which is characterized in that it is described using the facial image sample set and
After sample type of emotion is trained convolutional neural networks model, further includes:
Based on preset loss function, the loss of convolutional neural networks model is calculated；
When loss is higher than certain value, the weight parameter connected between each node in the convolutional neural networks model is adjusted, it is right
Convolutional neural networks model re -training, until obtaining optimal weight parameter.
8. a kind of information recommending apparatus characterized by comprising
Starting module is currently at display state when detecting for showing content information on the interface of customer service terminal device
Be specified content information when, start the camera of customer service terminal device；
Control module, the browsing before the display screen of the customer service terminal device is acquired for controlling the camera are described specified interior
Hold the facial image of information user, and extracts face characteristic information from the facial image；
Analysis module carries out micro- Expression analysis for face characteristic information to be inputted micro- Expression Recognition model, obtains the user
Type of emotion；Wherein, micro- Expression Recognition model is the qualified convolutional neural networks model of training；
Display module, for inquiring recommendation associated with the specified content information from database according to the type of emotion
The recommendation information is placed on the designated position at the interface of the customer service terminal device and shows by content information；
Wherein, the corresponding at least one recommendation information of each type of emotion.
9. a kind of terminal, which is characterized in that including memory and processor, computer-readable finger is stored in the memory
It enables, when the computer-readable instruction is executed by the processor, so that the processor is executed as appointed in claim 1 to 7
The step of information recommendation method described in one.
10. a kind of storage medium, is stored thereon with computer program, which is characterized in that the computer program is held by processor
When row, the information recommendation method as described in any one of claims 1 to 7 is realized.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910441455.XA CN110321477B (en) | 2019-05-24 | 2019-05-24 | Information recommendation method and device, terminal and storage medium |
| PCT/CN2019/115771 WO2020238023A1 (en) | 2019-05-24 | 2019-11-05 | Information recommendation method and apparatus, and terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910441455.XA CN110321477B (en) | 2019-05-24 | 2019-05-24 | Information recommendation method and device, terminal and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110321477A true CN110321477A (en) | 2019-10-11 |
| CN110321477B CN110321477B (en) | 2022-09-09 |
Family
ID=68119339
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910441455.XA Active CN110321477B (en) | 2019-05-24 | 2019-05-24 | Information recommendation method and device, terminal and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110321477B (en) |
| WO (1) | WO2020238023A1 (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110784763A (en) * | 2019-11-07 | 2020-02-11 | 深圳创维－Rgb电子有限公司 | Display terminal control method, display terminal and readable storage medium |
| CN111144906A (en) * | 2019-12-26 | 2020-05-12 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
| CN111177459A (en) * | 2019-12-14 | 2020-05-19 | 华为技术有限公司 | Information recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111291184A (en) * | 2020-01-20 | 2020-06-16 | 百度在线网络技术（北京）有限公司 | Expression recommendation method, device, equipment and storage medium |
| CN111708939A (en) * | 2020-05-29 | 2020-09-25 | 平安科技（深圳）有限公司 | Push method and device based on emotion recognition, computer equipment and storage medium |
| CN111951930A (en) * | 2020-08-19 | 2020-11-17 | 陈霄 | Emotion identification system based on big data |
| CN112016001A (en) * | 2020-08-17 | 2020-12-01 | 上海掌门科技有限公司 | Friend recommendation method and device and computer readable 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 |
| CN112214667A (en) * | 2020-09-18 | 2021-01-12 | 建信金融科技有限责任公司 | Information pushing method, device and equipment based on three-dimensional model and storage medium |
| CN112328813A (en) * | 2020-10-30 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Recommendation information generation method and device based on AI and computer equipment |
| CN112464025A (en) * | 2020-12-17 | 2021-03-09 | 当趣网络科技(杭州)有限公司 | Video recommendation method and device, electronic equipment and medium |
| CN112685596A (en) * | 2019-10-18 | 2021-04-20 | 中移(苏州)软件技术有限公司 | Video recommendation method and device, terminal and storage medium |
| CN112912915A (en) * | 2020-03-31 | 2021-06-04 | 浙江核新同花顺网络信息股份有限公司 | Method and system for refreshing current page information |
| CN112966128A (en) * | 2021-02-23 | 2021-06-15 | 武汉大学 | Self-media content recommendation method based on real-time emotion recognition |
| WO2021135197A1 (en) * | 2019-12-31 | 2021-07-08 | 北京市商汤科技开发有限公司 | State recognition method and apparatus, electronic device, and storage medium |
| CN113472834A (en) * | 2020-04-27 | 2021-10-01 | 海信集团有限公司 | Object pushing method and device |
| CN113515335A (en) * | 2021-05-20 | 2021-10-19 | 口碑(上海)信息技术有限公司 | Method and device for determining service information display form and electronic equipment |
| 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 |
| CN113724544A (en) * | 2021-08-30 | 2021-11-30 | 安徽淘云科技股份有限公司 | Playing method and related equipment thereof |
| CN113742599A (en) * | 2021-11-05 | 2021-12-03 | 太平金融科技服务(上海)有限公司深圳分公司 | Content recommendation method, device, equipment and computer readable storage medium |
| CN113794803A (en) * | 2021-08-06 | 2021-12-14 | 维沃移动通信（杭州）有限公司 | Unread message prompting method, unread message prompting device, electronic equipment and unread message prompting medium |
| CN113992745A (en) * | 2021-10-20 | 2022-01-28 | 平安银行股份有限公司 | Activity information pushing method and device, electronic equipment and storage medium |
| CN114443182A (en) * | 2020-10-30 | 2022-05-06 | 深圳Tcl数字技术有限公司 | Interface switching method, storage medium and terminal equipment |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112732953B (en) * | 2020-12-30 | 2024-04-12 | 上海众源网络有限公司 | Recommendation method, sample analysis device, electronic equipment and storage medium |
| CN112948691B (en) * | 2021-03-29 | 2022-12-13 | 建信金融科技有限责任公司 | Method and device for calculating experience index of entity place |
| CN113315876B (en) * | 2021-05-27 | 2023-01-20 | 中国银行股份有限公司 | Telephone bank service control method, device, server and storage medium |
| CN113239233B (en) * | 2021-05-31 | 2024-06-25 | 平安科技（深圳）有限公司 | Personalized image recommendation method, device, equipment and storage medium |
| CN113469704A (en) * | 2021-06-10 | 2021-10-01 | 云南电网有限责任公司 | System and method for automatically recommending customer service strategy |
| CN113469785A (en) * | 2021-06-29 | 2021-10-01 | 深圳市点购电子商务控股股份有限公司 | Method and device for establishing multi-person session and computer equipment |
| CN113643047B (en) * | 2021-08-17 | 2024-05-10 | 中国平安人寿保险股份有限公司 | Recommendation method, device, equipment and storage medium of virtual reality control strategy |
| TR2021013962A2 (en) * | 2021-09-06 | 2021-09-21 | Inc Yazilim Coezuemleri Ticaret Ltd Sirketi | A SYSTEM THAT DOES CHARACTER ANALYSIS |
| CN114170356B (en) * | 2021-12-09 | 2022-09-30 | 米奥兰特(浙江)网络科技有限公司 | Online route performance method and device, electronic equipment and storage medium |
| CN114287938B (en) * | 2021-12-13 | 2024-02-13 | 重庆大学 | Method and equipment for obtaining safety interval of human body parameters in building environment |
| CN117499477B (en) * | 2023-11-16 | 2024-06-07 | 北京易华录信息技术股份有限公司 | Information pushing method and system based on large model training |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150254447A1 (en) * | 2014-03-10 | 2015-09-10 | FaceToFace Biometrics, Inc. | Expression recognition in messaging systems |
| CN107169002A (en) * | 2017-03-31 | 2017-09-15 | 咪咕数字传媒有限公司 | A kind of personalized interface method for pushing and device recognized based on face |
| CN109583970A (en) * | 2018-12-14 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Advertisement placement method, device, computer equipment and storage medium |
| US20190122071A1 (en) * | 2017-10-24 | 2019-04-25 | International Business Machines Corporation | Emotion classification based on expression variations associated with same or similar emotions |
| CN109767261A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, computer equipment and storage medium |
| CN109785066A (en) * | 2019-01-17 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, equipment and storage medium based on micro- expression |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321477B (en) * | 2019-05-24 | 2022-09-09 | 平安科技（深圳）有限公司 | Information recommendation method and device, terminal and storage medium |
-
2019
- 2019-05-24 CN CN201910441455.XA patent/CN110321477B/en active Active
- 2019-11-05 WO PCT/CN2019/115771 patent/WO2020238023A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150254447A1 (en) * | 2014-03-10 | 2015-09-10 | FaceToFace Biometrics, Inc. | Expression recognition in messaging systems |
| CN107169002A (en) * | 2017-03-31 | 2017-09-15 | 咪咕数字传媒有限公司 | A kind of personalized interface method for pushing and device recognized based on face |
| US20190122071A1 (en) * | 2017-10-24 | 2019-04-25 | International Business Machines Corporation | Emotion classification based on expression variations associated with same or similar emotions |
| CN109583970A (en) * | 2018-12-14 | 2019-04-05 | 深圳壹账通智能科技有限公司 | Advertisement placement method, device, computer equipment and storage medium |
| CN109767261A (en) * | 2018-12-18 | 2019-05-17 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, computer equipment and storage medium |
| CN109785066A (en) * | 2019-01-17 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, equipment and storage medium based on micro- expression |
Non-Patent Citations (1)
| Title |
|---|
| 唐爽: "基于深度神经网络的微表情识别", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑(2018年第07期)》 * |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN112685596A (en) * | 2019-10-18 | 2021-04-20 | 中移(苏州)软件技术有限公司 | Video recommendation method and device, terminal and storage medium |
| CN112685596B (en) * | 2019-10-18 | 2023-04-14 | 中移(苏州)软件技术有限公司 | Video recommendation method and device, terminal and storage medium |
| CN110784763A (en) * | 2019-11-07 | 2020-02-11 | 深圳创维－Rgb电子有限公司 | Display terminal control method, display terminal and readable storage medium |
| CN110784763B (en) * | 2019-11-07 | 2021-11-02 | 深圳创维－Rgb电子有限公司 | Display terminal control method, display terminal and readable storage medium |
| CN111177459A (en) * | 2019-12-14 | 2020-05-19 | 华为技术有限公司 | Information recommendation method and device, electronic equipment and computer-readable storage medium |
| WO2021114936A1 (en) * | 2019-12-14 | 2021-06-17 | 华为技术有限公司 | Information recommendation method and apparatus, electronic device and computer readable storage medium |
| CN111144906A (en) * | 2019-12-26 | 2020-05-12 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
| WO2021135197A1 (en) * | 2019-12-31 | 2021-07-08 | 北京市商汤科技开发有限公司 | State recognition method and apparatus, electronic device, and storage medium |
| CN111291184A (en) * | 2020-01-20 | 2020-06-16 | 百度在线网络技术（北京）有限公司 | Expression recommendation method, device, equipment and storage medium |
| CN112912915A (en) * | 2020-03-31 | 2021-06-04 | 浙江核新同花顺网络信息股份有限公司 | Method and system for refreshing current page information |
| US11971934B2 (en) | 2020-03-31 | 2024-04-30 | Hithink Royalflush Information Network Co., Ltd. | Methods and systems for refreshing current page information |
| CN113472834A (en) * | 2020-04-27 | 2021-10-01 | 海信集团有限公司 | Object pushing method and device |
| CN111708939B (en) * | 2020-05-29 | 2024-04-16 | 平安科技（深圳）有限公司 | Emotion recognition-based pushing method and device, computer equipment and storage medium |
| CN111708939A (en) * | 2020-05-29 | 2020-09-25 | 平安科技（深圳）有限公司 | Push method and device based on emotion recognition, computer equipment and storage medium |
| CN112016001A (en) * | 2020-08-17 | 2020-12-01 | 上海掌门科技有限公司 | Friend recommendation method and device and computer readable medium |
| CN111951930A (en) * | 2020-08-19 | 2020-11-17 | 陈霄 | Emotion identification system based on big data |
| CN112214667A (en) * | 2020-09-18 | 2021-01-12 | 建信金融科技有限责任公司 | Information pushing method, device and equipment based on three-dimensional model and storage medium |
| CN112328813A (en) * | 2020-10-30 | 2021-02-05 | 中国平安人寿保险股份有限公司 | Recommendation information generation method and device based on AI and computer equipment |
| CN112328813B (en) * | 2020-10-30 | 2023-08-25 | 中国平安人寿保险股份有限公司 | Recommendation information generation method and device based on AI and computer equipment |
| CN114443182A (en) * | 2020-10-30 | 2022-05-06 | 深圳Tcl数字技术有限公司 | Interface switching method, storage medium and terminal equipment |
| CN112464025A (en) * | 2020-12-17 | 2021-03-09 | 当趣网络科技(杭州)有限公司 | Video recommendation method and device, electronic equipment and medium |
| CN112464025B (en) * | 2020-12-17 | 2023-08-01 | 当趣网络科技(杭州)有限公司 | Video recommendation method and device, electronic equipment and medium |
| CN112966128A (en) * | 2021-02-23 | 2021-06-15 | 武汉大学 | Self-media content recommendation method based on real-time emotion recognition |
| CN113515335A (en) * | 2021-05-20 | 2021-10-19 | 口碑(上海)信息技术有限公司 | Method and device for determining service information display form and electronic equipment |
| CN113794803A (en) * | 2021-08-06 | 2021-12-14 | 维沃移动通信（杭州）有限公司 | Unread message prompting method, unread message prompting device, electronic equipment and unread message prompting 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 |
| CN113724544B (en) * | 2021-08-30 | 2023-08-22 | 安徽淘云科技股份有限公司 | Playing method and related equipment thereof |
| CN113724544A (en) * | 2021-08-30 | 2021-11-30 | 安徽淘云科技股份有限公司 | Playing method and related equipment thereof |
| CN113992745A (en) * | 2021-10-20 | 2022-01-28 | 平安银行股份有限公司 | Activity information pushing method and device, electronic equipment and storage medium |
| CN113992745B (en) * | 2021-10-20 | 2024-03-22 | 平安银行股份有限公司 | Activity information pushing method and device, electronic equipment and storage medium |
| CN113742599A (en) * | 2021-11-05 | 2021-12-03 | 太平金融科技服务(上海)有限公司深圳分公司 | Content recommendation method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020238023A1 (en) | 2020-12-03 |
| CN110321477B (en) | 2022-09-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321477A (en) | Information recommendation method, device, terminal and storage medium | |
| Schulz et al. | A design space of visualization tasks | |
| CN110046304A (en) | A kind of user's recommended method and device | |
| CN109558535B (en) | Personalized article pushing method and system based on face recognition | |
| Tkalcic et al. | Affective labeling in a content-based recommender system for images | |
| Liu et al. | Fineaction: A fine-grained video dataset for temporal action localization | |
| Fan et al. | Facial action unit intensity estimation via semantic correspondence learning with dynamic graph convolution | |
| CN110334689A (en) | Video classification methods and device | |
| CN110298249A (en) | Face identification method, device, terminal and storage medium | |
| CN108874992A (en) | The analysis of public opinion method, system, computer equipment and storage medium | |
| CN107203897A (en) | A kind of evaluation method of Products Show degree, apparatus and system | |
| CN110580489B (en) | Data object classification system, method and equipment | |
| CN108205768A (en) | Database building method and data recommendation method and device, equipment and storage medium | |
| CN109658188A (en) | Source of houses recommended method, device, equipment and storage medium based on big data analysis | |
| CN110472082A (en) | Data processing method, device, storage medium and electronic equipment | |
| CN110458360A (en) | Prediction technique, device, equipment and the storage medium of hot resource | |
| CN110287313A (en) | A kind of the determination method and server of risk subject | |
| Zhang et al. | Guided attention in cnns for occluded pedestrian detection and re-identification | |
| CN109961077A (en) | Gender prediction's method, apparatus, storage medium and electronic equipment | |
| CN108062416B (en) | Method and apparatus for generating label on map | |
| CN110413894A (en) | The training method of commending contents model, the method for commending contents and relevant apparatus | |
| CN108229527A (en) | Training and video analysis method and apparatus, electronic equipment, storage medium, program | |
| CN106537387B (en) | Retrieval/storage image associated with event | |
| CN109376598A (en) | Facial expression image processing method, device, computer equipment and storage medium | |
| CN110347781A (en) | Article falls discharge method, article recommended method, device, 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |