CN109859008A - Method of Commodity Recommendation, device and terminal device based on user's portrait - Google Patents
Method of Commodity Recommendation, device and terminal device based on user's portrait Download PDFInfo
- Publication number
- CN109859008A CN109859008A CN201910039148.9A CN201910039148A CN109859008A CN 109859008 A CN109859008 A CN 109859008A CN 201910039148 A CN201910039148 A CN 201910039148A CN 109859008 A CN109859008 A CN 109859008A
- Authority
- CN
- China
- Prior art keywords
- commodity
- value
- evaluation
- subordinate
- user
- 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 57
- 238000011156 evaluation Methods 0.000 claims abstract description 158
- 239000013065 commercial product Substances 0.000 claims abstract description 57
- 238000004458 analytical method Methods 0.000 claims abstract description 51
- 230000000875 corresponding effect Effects 0.000 claims description 110
- 230000002596 correlated effect Effects 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000001419 dependent effect Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 abstract description 11
- 230000006870 function Effects 0.000 description 13
- 239000000047 product Substances 0.000 description 12
- 238000013527 convolutional neural network Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000008451 emotion Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000009193 crawling Effects 0.000 description 3
- 241000270322 Lepidosauria Species 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000002996 emotional effect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000005764 inhibitory process Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000003999 initiator Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 210000001938 protoplast Anatomy 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000000344 soap Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Abstract
The present invention is suitable for technical field of data processing, provides Method of Commodity Recommendation, device, terminal device and computer readable storage medium based on user's portrait, comprising: acquisition user images, and the user images are analyzed to obtain characteristics of human body's value;The commercial product recommending type selected of user is obtained, retrieves at least two subordinate's commodity under the commercial product recommending type in the database, and grabs at least two commodity evaluation corresponding with each subordinate's commodity;It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as objective appraisal, comprehensive analysis is carried out to all objective appraisals and obtains evaluation of estimate, and recommends subordinate's commodity corresponding with institute's evaluation values according to preset order.The present invention carries out commercial product recommending getting user images automatically, improves the convenience of commercial product recommending.
Description
Technical field
The invention belongs to technical field of data processing, more particularly to the Method of Commodity Recommendation based on user's portrait, device, end
End equipment and computer readable storage medium.
Background technique
With the fast development of information technology, current such as electric business website or service provider, supermarket all can be the use to be done shopping
Family provides the function of commercial product recommending, so that user can select in the commodity of recommendation, rather than aimlessly one
Part browses incoherent commodity, on the one hand commercial product recommending is that user brings convenience, on the other hand also improves what commodity were sold
Probability.
In the prior art, usually recommended according to data set by user in Recommendations related to user
Commodity.But actually user may be to the data that sets of needs and ignorant, such as in the feelings not measured in advance
Under condition, the data such as user and the waistline that can not know itself, therefore commercial product recommending by user's manual measurement and is set in the prior art
Fixed number evidence, convenience and applicability are low.
Summary of the invention
In view of this, the embodiment of the invention provides the Method of Commodity Recommendation drawn a portrait based on user, device, terminal device with
And computer readable storage medium is operated with needing user to manually set data when solving to carry out commercial product recommending in the prior art
The cumbersome problem of process.
The first aspect of the embodiment of the present invention provides a kind of Method of Commodity Recommendation based on user's portrait, comprising:
User images are acquired, and the user images are analyzed to obtain characteristics of human body's value;
The commercial product recommending type that user selectes is obtained, retrieves at least two under the commercial product recommending type in the database
Subordinate's commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as target and comments
Valence carries out comprehensive analysis to all objective appraisals and obtains evaluation of estimate, and recommends and institute's evaluation values pair according to preset order
The subordinate's commodity answered.
The second aspect of the embodiment of the present invention provides a kind of device for recommending the commodity based on user's portrait, comprising:
Acquisition unit for acquiring user images, and analyzes the user images to obtain characteristics of human body's value;
Picking unit, the commercial product recommending type selected for obtaining user, retrieves the commercial product recommending kind in the database
At least two subordinate's commodity under class, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
Recommendation unit, for by subordinate's commodity are corresponding and relevant to characteristics of human body value commodity evaluation
Be determined as objective appraisal, comprehensive analysis carried out to all objective appraisals and obtain evaluation of estimate, and according to preset order recommend with
The corresponding subordinate's commodity of institute's evaluation values.
The third aspect of the embodiment of the present invention provides a kind of terminal device, and the terminal device includes memory, processing
Device and storage in the memory and the computer program that can run on the processor, described in the processor execution
Following steps are realized when computer program:
User images are acquired, and the user images are analyzed to obtain characteristics of human body's value;
The commercial product recommending type that user selectes is obtained, retrieves at least two under the commercial product recommending type in the database
Subordinate's commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as target and comments
Valence carries out comprehensive analysis to all objective appraisals and obtains evaluation of estimate, and recommends and institute's evaluation values pair according to preset order
The subordinate's commodity answered.
The fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer program, and the computer program realizes following steps when being executed by processor:
User images are acquired, and the user images are analyzed to obtain characteristics of human body's value;
The commercial product recommending type that user selectes is obtained, retrieves at least two under the commercial product recommending type in the database
Subordinate's commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as target and comments
Valence carries out comprehensive analysis to all objective appraisals and obtains evaluation of estimate, and recommends and institute's evaluation values pair according to preset order
The subordinate's commodity answered.
Existing beneficial effect is the embodiment of the present invention compared with prior art:
The embodiment of the present invention analyzes user images to obtain characteristics of human body's value by acquisition user images, then
The commercial product recommending type (such as perfumed soap) selected according to user is searched at least two subordinate's commodity under the commercial product recommending type, is obtained
Objective appraisal under each subordinate's commodity and relevant to characteristics of human body's value is taken, to calculate evaluation according to all objective appraisals
Value determines the sequence of push subordinate's commodity according to the numerical values recited of evaluation of estimate, and the embodiment of the present invention is according to the user images of acquisition
Characteristics of human body's value is calculated, and automatic push subordinate commodity are worth according to characteristics of human body, user is avoided and manually sets data, improve
The convenience and applicability of commercial product recommending.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the implementation flow chart for the Method of Commodity Recommendation based on user's portrait that the embodiment of the present invention one provides;
Fig. 2 is the implementation flow chart of the Method of Commodity Recommendation provided by Embodiment 2 of the present invention based on user's portrait;
Fig. 3 is the implementation flow chart for the Method of Commodity Recommendation based on user's portrait that the embodiment of the present invention three provides;
Fig. 4 is the implementation flow chart for the Method of Commodity Recommendation based on user's portrait that the embodiment of the present invention four provides;
Fig. 5 is the implementation flow chart for the Method of Commodity Recommendation based on user's portrait that the embodiment of the present invention five provides;
Fig. 6 is the structural block diagram for the device for recommending the commodity based on user's portrait that the embodiment of the present invention six provides;
Fig. 7 is the schematic diagram of seven offer/terminal device of the embodiment of the present invention.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Fig. 1 shows the implementation process of the Method of Commodity Recommendation provided in an embodiment of the present invention based on user's portrait, is described in detail
It is as follows:
In S101, user images are acquired, and analyzed the user images to obtain characteristics of human body's value.
In embodiments of the present invention, it for the ease of illustrating, is illustrated by terminal device of executing subject.Firstly, terminal
Equipment obtains user images, and user images can be the image comprising entire human body, be also possible to a part of area comprising human body
The image in domain is determined according to practical application scene.When obtaining user images, the picture that terminal device receives user can be
Selection instruction calls album picture corresponding with picture selection instruction as user images from photograph album;It is also possible to pass through end
The included camera of end equipment carries out Image Acquisition or video acquisition to the image within the scope of camera view, if what is carried out is
Image Acquisition, then directly using the image collected as user images, if what is carried out is video acquisition, according to preset point
Frame rate (for example each second is a frame) carries out sub-frame processing to the video collected, obtains at least two framed user's images.
For collected user images, which is analyzed to obtain characteristics of human body's value, wherein preset
Feature to be collected, and then analyze and obtain the region met in user images with feature to be collected, and according to multiple spies in the region
Characteristics of human body's value is calculated in value indicative, and characteristics of human body's value can be one or at least two in height values, brachium value and waistline value etc.
A combination.The embodiment of the present invention to the specific calculation of characteristics of human body's value and without limitation, such as can will containing meet to
The multiple images in the region of acquisition characteristics are trained convolutional neural networks as input parameter, meet especially by known
The characteristic value in the region of feature to be collected optimizes the classifier in convolutional neural networks, and convolutional neural networks training is completed
Afterwards, user images are inputted into convolutional neural networks, can be identified in user images by the classifier in convolutional neural networks
Meet the region of feature to be collected, and extract characteristics of human body's value in the region, or the human body analysis interface of open source can be called straight
It connects and user images is analyzed to obtain characteristics of human body's value etc..It is noted that at least two framed user's images if it exists, then exist
After the characteristics of human body's value for calculating every framed user's image, mean value computation, and the result that will be obtained are carried out to all characteristics of human body's values
As updated characteristics of human body's value.
Optionally, if detecting, there are human bodies within the scope of camera view, acquire user images.It is used to promote acquisition
The degree of automation of family image can set camera to long-time open state, and to camera in embodiments of the present invention
Image within the vision carries out human bioequivalence, if detecting, there are human bodies within the scope of camera view, by camera view
Image in range is saved as user images, is convenient for subsequent analysis;If detecting within the scope of camera view not
After there are human bodies, then abandon to the image within the scope of camera view or temporarily save, and setting does not carry out the image
Continuous analysis.The above method is applicable to the scenes such as supermarket gate inhibition, improves the convenience of acquisition user images.
In S102, the commercial product recommending type that user selectes is obtained, is retrieved under the commercial product recommending type in the database
At least two subordinate's commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity.
There are many type of merchandize possibility supported due to terminal device, such as jacket, shoes and lower dress etc., therefore in order to reduce quotient
The range that product are recommended obtains the commercial product recommending type that user selectes, specifically also for the accuracy recommended is promoted in this step
It can be by commodity selection instruction, text instruction or the phonetic order etc. of acquisition user, to analyze commercial product recommending type.True
After determining commercial product recommending type, at least two subordinate's commodity under commercial product recommending type are retrieved in the database, wherein subordinate's commodity
It is the specific commodity under commercial product recommending type, if commercial product recommending type is " cloth shoes ", then subordinate's commodity are some commodity product
The cloth shoes of some model of board.For each subordinate's commodity retrieved, corresponding at least two commodity evaluation, citing are grabbed
For, can by the electric business website where third party evaluation website or subordinate's commodity provide interface, using subordinate's commodity as
Keyword is searched at least two commodity relevant to subordinate's commodity by interface and is evaluated;In no interface or interface is not made
In the case where with permission, reptile instrument can also be write, and using subordinate's commodity as keyword, in electric business website or third party evaluation
At least two commodity evaluation relevant to subordinate's commodity is crawled in website.
Optionally, at least two subordinate's commodity are retrieved according to preset search condition, and is grabbed according to preset grasping condition
Take the corresponding at least two commodity evaluation of each subordinate's commodity.In embodiments of the present invention, the settable retrieval to subordinate's commodity
Condition and the grasping condition evaluated commodity, search condition and grasping condition can be customized by users setting, can also be set by terminal
It is standby to carry out unified setting in advance according to practical application scene, for example it is clear under only retrieval commercial product recommending type that search condition, which is arranged,
Number of looking at is more than subordinate's commodity of preset times, and to retrieve more popular subordinate's commodity, for example setting grasping condition is only
Subordinate's commodity are corresponding and creation time is later than the commodity evaluation of preset time and is passed through with grabbing newer commodity evaluation for crawl
The above method improves the applicability of commercial product recommending, and reduces the data volume got, improves computational efficiency.
It is subordinate's commodity are corresponding and relevant to characteristics of human body value commodity evaluation determines in S103
For objective appraisals, comprehensive analysis are carried out to all objective appraisals and obtain evaluation of estimate, and according to preset order recommend with it is described
The corresponding subordinate's commodity of evaluation of estimate.
Usually, the characteristic value of initiator of the commodity evaluation comprising commodity evaluation, such as buyer purchase in electric business website
It after buying dress, is evaluated if buyer issues according to the clothes of purchase, the subsequent evaluation is the buyer recorded with electric business website
Characteristic value (such as height and weight) constitute together commodity evaluation, to be issued and be shown.Therefore in the embodiment of the present invention
In, all commodity corresponding with subordinate's commodity grabbed are evaluated, the quotient wherein containing characteristics of human body's value is found out
Valence is judged, the commodity found out evaluation is determined as objective appraisal.For example, commodity evaluate evaluationAFor " height values
160 centimetres, evaluation: good ", commodity evaluate evaluationBFor " 170 centimetres of height values, evaluation: very poor ", commodity evaluation
evaluationCFor " 190 centimetres of height values, evaluation: can be with ", and characteristics of human body's value is 170 centimetres of height values, then comments commodity
Valence evaluationBIt is determined as objective appraisal.It in addition to this, in step s 102, can also be direct for optimization processing step
Using characteristics of human body's value and subordinate's commodity as the corresponding at least two commodity evaluation of keyword crawl subordinate's commodity, and will grab
Commodity evaluation be determined directly as objective appraisal.
It optionally, can be characteristics of human body's value based on default error range in order to promote the quantity of the objective appraisal determined
Corresponding characteristic error range is set, if the characteristic value in commodity evaluation is fallen within the scope of characteristic error, which is evaluated
It is determined as objective appraisal.For example, characteristics of human body's value is 170 centimetres of height values, and default error range be [- 10 centimetres, 10
Centimetre], then can calculate characteristic error range is [160 centimetres, 180 centimetres], with the citing of above-mentioned example, since commodity are evaluated
evaluationAIn characteristic value (160 centimetres of height values) drop within the scope of characteristic error, therefore commodity are evaluated
evaluationAEvaluation is evaluated with commodityBIt is determined as objective appraisal, wherein default error range can be carried out by user
Manual setting.The above method improves the quantity for the objective appraisal determined within an acceptable range, to improve subsequent meter
The reliability of the evaluation of estimate of calculating.
After obtaining objective appraisal, comprehensive analysis is carried out to all objective appraisals and obtains evaluation of estimate, wherein according to target
The format of evaluation is different, and there are different analysis modes in the embodiment of the present invention:
(1) if the format score of objective appraisal, the score in all objective appraisals is obtained, and all scores are carried out equal
Value handles to obtain evaluation of estimate;
(2) if the format of objective appraisal is text, a kind of mode be based on preset evaluation vocabulary to objective appraisal into
Row text analyzing obtains the Score on Prediction value of each objective appraisal, then carries out average value processing to all Score on Prediction values and commented
Value, for example the Score on Prediction value of the objective appraisal containing evaluation vocabulary " good " or " stick " is defined as 2, evaluation vocabulary will be contained
The Score on Prediction value of the objective appraisal of " poor " or " bad " is defined as 1;Another way is that the semantic sentiment analysis of open source is called to connect
Mouth carries out semantic sentiment analysis to each objective appraisal, and different emotional values is arranged for different emotion types in advance, right
After the semantic sentiment analysis of objective appraisal progress obtains emotion type, the corresponding emotion of emotion type is assigned for the objective appraisal
Value carries out average value processing to all emotional values and is evaluated finally after the completion of to the semantic sentiment analysis of all objective appraisals
Value.
After the completion of comprehensive analysis, the i.e. corresponding evaluation of estimate of subordinate's commodity, in order to realize under evaluation preferably
The preferential push for belonging to commodity recommends subordinate's commodity corresponding with evaluation of estimate according to preset order.Wherein, if evaluation of estimate gets over Gao Daibiao
Objective appraisal is better, then preset order is the sequence of evaluation of estimate from high to low;If evaluation of estimate is lower, to represent objective appraisal better,
Preset order is the sequence of evaluation of estimate from low to high.The embodiment of the present invention to the specific way of recommendation and without limitation, such as can
By tabular form realize preferential recommendation, i.e., evaluation of estimate it is higher represent objective appraisal it is better in the case where, if evaluation of estimate is higher,
Then the position of corresponding subordinate's commodity in lists is more forward.When recommending subordinate's commodity, if subordinate's commodity are wearable quotient
Product such as jacket or trousers etc., then can obtain subordinate's commodity businessman provide subordinate's commodity image dimensional parameters, and according to
The scaling of the dimensional parameters and user images, by the feature to be collected of meeting in the image and user images of subordinate's commodity
Region is overlapped, and is realized trying on online for subordinate's commodity, is promoted commercial product recommending effect.
By embodiment illustrated in fig. 1 it is found that in embodiments of the present invention, by analyzing collected user images
Characteristics of human body's value is obtained, the commercial product recommending type that user selectes is obtained, is retrieved under commercial product recommending type in the database at least
Two subordinate's commodity, and at least two commodity evaluation corresponding with each subordinate's commodity is grabbed, it finally will be with characteristics of human body's value phase
The commodity evaluation of pass is determined as objective appraisal, carries out comprehensive analysis to all objective appraisals and obtains evaluation of estimate, and is suitable according to presetting
Sequence recommends subordinate's commodity corresponding with evaluation of estimate, and the embodiment of the present invention reduces the work of user's manual setting parameter, and leads to
The mode for crossing Calculation Estimation value realizes the preferential recommendation of favorable comment commodity, improves the convenience and accuracy of commercial product recommending.
Shown in Fig. 2, be on the basis of the embodiment of the present invention one, and the quantity of objective appraisal be less than preset quantity, and
Commercial product recommending type and preset correlated characteristic there are on the basis of corresponding relationship, to by subordinate's commodity it is corresponding and with human body spy
Value indicative relevant commodity evaluation is determined as objective appraisal, to all objective appraisals carry out comprehensive analysis obtain the process of evaluation of estimate into
A kind of method obtained after row refinement.The embodiment of the invention provides the implementation processes for the Method of Commodity Recommendation drawn a portrait based on user
Figure, as shown in Fig. 2, the Method of Commodity Recommendation may comprise steps of:
In S201, associated eigenvalue corresponding with the correlated characteristic in characteristics of human body's value is extracted.
In embodiments of the present invention, there are the stature of user is more abnormal, and characteristics of human body's value be based on it is more to
Acquisition characteristics are analyzed the case where obtaining to user images, then the quantity for the objective appraisal determined according to characteristics of human body's value can
It can be less than preset quantity, lead to not calculate reliable and effective evaluation of estimate according to less objective appraisal.Such as spy to be collected
Sign includes height, brachium, waistline and shoulder breadth, then collected characteristics of human body's value includes height values, brachium value, waistline value and shoulder breadth
Value, and the brachium of user will be grown compared with ordinary person, then the quantity for being worth relevant commodity evaluation to the characteristics of human body of the user is possibly less than
Preset quantity.For above situation, different correlated characteristics can be set for different types of merchandize in advance, which is to use
Family selects an important factor for commodity under type of merchandize, for example can be the corresponding correlated characteristic of type of merchandize " jacket " setting is body
It is waistline that height, which is the corresponding correlated characteristic of type of merchandize " belt " setting, corresponding related special for type of merchandize " shoes " setting
Sign is subscript.After getting the commercial product recommending type that user selectes, the corresponding correlated characteristic of commercial product recommending type is obtained, and mention
Associated eigenvalue corresponding with the correlated characteristic in characteristics of human body's value is taken, it is worth noting that, following correlated characteristic all refers to
The corresponding correlated characteristic of commercial product recommending type.For example, characteristics of human body's value includes height values, brachium value, waistline value and shoulder breadth
Value, and the corresponding correlated characteristic of commercial product recommending type " jacket " is height, therefore extract number relevant to height in characteristics of human body's value
Value is used as associated eigenvalue, i.e. associated eigenvalue is the height values in protoplast's body characteristics value.
In S202, the commodity evaluation relevant to characteristics of human body's value is determined as first object evaluation, and will
The commodity evaluation relevant to the associated eigenvalue is determined as the second objective appraisal.
It, will be wherein relevant to characteristics of human body's value after all commodity evaluations corresponding for the subordinate's commodity grabbed
Commodity evaluation is determined as first object evaluation, and wherein the evaluation of relevant to associated eigenvalue commodity will be determined as the second target and comment
Valence, since associated eigenvalue is generally only a part of characteristics of human body's value, therefore the quantity for the first object evaluation determined is usual
Less than the second objective appraisal.
In S203, comprehensive analysis is carried out to all first object evaluations and obtains first object value, to all described
Second objective appraisal carries out comprehensive analysis and obtains the second target value, and carries out to the first object value and second target value
Weighted sum obtains institute's evaluation values.
According to same comprehensive analysis method in step S103, to all first objects evaluate carry out comprehensive analysis obtain
First object value carries out comprehensive analysis to all second objective appraisals and obtains the second target value.Since characteristics of human body's value is related to
Feature to be collected it is more, therefore calculated first object value is worth based on characteristics of human body and is actually more bonded user, therefore calculated
When final evaluation of estimate, traditional mean algorithm is not used, first object value and the second target value are weighted and are asked
With, wherein different weights is arranged to first object value and the second target value, and the weight of first object value is preferably than second
The weight of target value is bigger, and the weights sum of the weight of first object value and the second target value is 1.The specific value of weight can root
It is set according to practical application scene, for example can preset the weight of first object value is 60%, the second target value of setting
Weight is 40%, it is assumed that first object value is 4, and the second target value is 4.5, then being weighted the evaluation of estimate obtained after summation is 4*
60%+4.5*40%=4.2.
By embodiment illustrated in fig. 2 it is found that in embodiments of the present invention, by extract in characteristics of human body's value with correlated characteristic
Commodity evaluation relevant to characteristics of human body's value is determined as first object evaluation by corresponding associated eigenvalue, will be with correlated characteristic
It is worth relevant commodity evaluation and is determined as the second objective appraisal, progress comprehensive analysis is evaluated to all first objects and obtains first object
Value carries out comprehensive analysis to all second objective appraisals and obtains the second target value, finally to first object value and the second target value
It is weighted summation and obtains evaluation of estimate, the situation less in the objective appraisal obtained based on characteristics of human body's value of the embodiment of the present invention
Under, characteristics of human body's value is intercepted based on preset correlated characteristic, improves the quantity of the objective appraisal got, and mention
The reliability of calculated evaluation of estimate is risen.
It is on the basis of the embodiment of the present invention one, and in user images is in the user and camera phase shown in Fig. 3
Every preset acquisition apart from when be acquired on the basis of, user images will be analyzed obtain the process of characteristics of human body's value into
A kind of method obtained after row refinement.The embodiment of the invention provides the implementation processes for the Method of Commodity Recommendation drawn a portrait based on user
Figure, as shown in figure 3, the Method of Commodity Recommendation may comprise steps of:
In S301, obtain preset feature to be collected, and acquire in the user images with the feature pair to be collected
The image feature value answered.
User oriented image of the embodiment of the present invention be user and camera be separated by preset acquisition apart from when be acquired
Scene, for example install contact inductor in the fixed bit of supermarket gate inhibition, when user tramples contact inductor, triggering
Acquisition operation, using the image within the scope of current camera view as user images.For obtained user images, obtain pre-
If feature to be collected, and acquire image feature value corresponding with feature to be collected, image feature value herein in user images
It is the actual numerical value in feature to be collected shared region in user images.Wherein, feature to be collected is shared in user images
Region can be by inputting trained convolutional neural networks for user images or the human body analysis interface of open source being called to obtain, herein
It does not repeat them here.For example, the height of user images is 50 centimetres, and feature to be collected is height, and shared by feature to be collected
The height in region is the half of the height of user images, then can determine that image feature value includes height values, and the tool of height values
Body numerical value is 25 centimetres.
In S302, the acquisition is obtained apart from corresponding first scaling, based on described image characteristic value and described
Characteristics of human body's value is calculated in first scaling.
Under the premise of having determined that acquisition distance, then the first scaling between image and actual object shot is
Known (for example can be obtained by repeatedly shooting the image of the sample of known dimensions), and after obtaining image feature value, it will scheme
Characteristics of human body's value is obtained as characteristic value and the first scaling carry out product calculation.For example, if acquisition is apart from corresponding the
One scaling is 7:1, then in the case where image feature value is 25 centimetres, can calculate characteristics of human body's value is 25*7=175
Centimetre.
By embodiment illustrated in fig. 3 it is found that in embodiments of the present invention, by obtaining preset feature to be collected, and adopting
Collect image feature value corresponding with feature to be collected in user images, obtains acquisition apart from corresponding first scaling, will scheme
Characteristics of human body's value is obtained as characteristic value and the first scaling carry out product calculation, the embodiment of the present invention is fixed in acquisition distance
In the case of realize the calculating of characteristics of human body's value, improve the diversity of the calculation of characteristics of human body's value.
It is on the basis of the embodiment of the present invention one, and on the basis including at least two features to be collected shown in Fig. 4
On, a kind of method for being obtained after the process for being analyzed to obtain characteristics of human body's value to user images is refined.The present invention is real
It applies example and provides the implementation flow chart of the Method of Commodity Recommendation based on user's portrait, as shown in figure 4, the Method of Commodity Recommendation can be with
The following steps are included:
In S401, at least two image corresponding at least two features to be collected in the user images is acquired
Characteristic value, wherein there are corresponding relationships with a feature to be collected for each described image characteristic value.
In embodiments of the present invention, at least two feature to be collected if it exists, then first analyze user images, adopt
Collect at least two image feature value corresponding at least two features to be collected in user images, wherein each image feature value
There are corresponding relationships with a feature to be collected.
In S402, the actual characteristic value corresponding with the feature to be collected of user input is obtained, and will
Ratio between actual characteristic value described image characteristic value corresponding with the feature to be collected is determined as the second pantograph ratio
Example.
There are at least two features to be collected, and user has known the actual characteristic value of one of them feature to be collected
When (actual numerical value), obtain the actual characteristic value corresponding with a feature to be collected of user's input, and by actual characteristic value with
Ratio between the corresponding image feature value of the feature to be collected is determined as the second scaling.For example, the reality of height
Characteristic value be for most people it is known, on this basis, if feature to be collected includes height, waistline and brachium,
The actual characteristic value corresponding with height of user's input can be obtained, so that it is determined that the second scaling, such as the corresponding figure of height
Picture characteristic value is 25 centimetres, and the actual characteristic value corresponding with height of the user's input got is 180 centimetres, then can be obtained
180/25=7.2 determines that the second scaling is 7.2:1.
In S403, all described image characteristic values are individually subjected to operation with second scaling, and by operation
All described image eigenvalue clusters afterwards are combined into characteristics of human body's value.
After obtaining the second scaling, all image feature values are individually subjected to product fortune with the second scaling
It calculates, and combines all image feature values after product calculation for characteristics of human body's value.For example, feature height to be collected is corresponding
Image feature value be 25 centimetres, the corresponding image feature value of feature waistline to be collected is 11 centimetres, feature shoulder breadth pair to be collected
The image feature value answered be 6 centimetres, the second scaling be 7.2:1, then by all image feature values individually with the second pantograph ratio
Characteristics of human body's value that example be combined into after operation includes height values 180 (25 × 7.2) centimetre, waistline value 79.2 (11 × 7.2)
Centimetre, shoulder breadth value 43.2 (6 × 7.2) centimetre.
By embodiment illustrated in fig. 4 it is found that in embodiments of the present invention, acquiring to be collected at least two in user images
Corresponding at least two image feature value of feature obtains the actual characteristic value corresponding with a feature to be collected of user's input,
And the ratio between actual characteristic value image feature value corresponding with the feature to be collected is determined as the second scaling, by institute
There is image feature value individually to carry out operation with the second scaling, and combines all image feature values after operation for human body spy
There is feature at least to be collected and user to the actual characteristic value of one of them feature to be collected in value indicative, the embodiment of the present invention
In known situation, characteristics of human body's value is obtained by calculating the second scaling, improves the calculation of characteristics of human body's value
Diversity, and the above method is not required to fixed acquisition distance, improves the flexibility for calculating characteristics of human body's value.
Shown in Fig. 5, be on the basis of the embodiment of the present invention one, and include at least two Brands on the basis of,
A kind of method obtained after the process for obtaining evaluation of estimate to all objective appraisals progress comprehensive analysis is refined.The present invention is real
It applies example and provides the implementation flow chart of the Method of Commodity Recommendation based on user's portrait, as shown in figure 5, the Method of Commodity Recommendation can be with
The following steps are included:
In S501, at least two browsings record of the user within a preset period of time is crawled, the institute crawled is analyzed
There is the Brand that occurs in the browsing record, and the Brand is shared in all browsing records
Ratio-dependent is brand coefficient.
User is crawled in embodiments of the present invention to more be bonded user for the concern situation of different Brands
At least two browsings record within a preset period of time, wherein preset time period can be configured according to practical application scene, than
One month before being current time such as settable preset time period.Crawl browsing record when, using the account name of user as
Keyword, where subordinate's commodity electric business website or other third party website in crawl, crawling operation can be based on electricity
The searching interface that quotient website or third party website provide can also be realized by writing reptile instrument such as python reptile instrument.For
Judge whether contain Brand in browsing record, the brand library for containing at least two Brand can be pre-established, and set
Setting the corresponding browsing time of each Brand is zero, and each browsing crawled is recorded, will be all in brand library
Brand is compared one by one with browsing record, if judging, browsing record includes some Brand, to the commodity
The corresponding browsing time of brand executes cumulative one operation.After being recorded and analyzed to all browsings, for each commodity product
Board calculates the ratio of the total number of the corresponding browsing time of Brand and browsing record, then to ratio execution plus an operation
Brand coefficient is obtained, there are corresponding relationships for a final Brand and a brand coefficient.For example, include in brand library
Brand BrandA, Brand BrandBWith Brand BrandC, crawl browsing record HistoryAFor
“BrandA", browsing record HistoryBFor " BrandB", browsing record HistoryCFor " BrandD", then it can calculate commodity product
Board BrandAWith Brand BrandBCorresponding browsing time is all 1, and due to Brand BrandDNot in brand library,
Therefore without record.In the above example, the total number for browsing record is 3, therefore can calculate Brand BrandAIt is corresponding
Brand coefficient is 1/3+1 ≈ 1.33, Brand BrandBCorresponding brand coefficient is 1/3+1 ≈ 1.33.It is noted that
Above-mentioned example, which has only been enumerated in a browsing record, only there is the example of a Brand, but should know, is actually answering
With in scene, at least two Brands are likely to occur in a browsing record.
Optionally, the modification interface that Brand in brand library is opened to user is based on modification interface according to user and issues
Instruction to brand library execute modification operation.Due to brand coefficient be calculated only for the Brand in brand library, therefore
In the embodiment of the present invention, the modification interface of Brand in brand library can be opened to user, and brand library is added by user
Brand and the operation such as Brand is deleted, promote user experience, is suitable for user only to some or certain several Brands
In the presence of the desired scene of browsing.
In S502, if subordinate's commodity are related to a Brand, to the subordinate quotient grabbed
All objective appraisals under product carry out comprehensive analysis and obtain evaluation of estimate, and are based on the corresponding brand of the Brand
Coefficient is weighted institute's evaluation values to obtain updated institute's evaluation values.
After retrieving subordinate's commodity, if subordinate's commodity and all Brands in brand library are uncorrelated, to grabbing
All objective appraisals under the subordinate's commodity got carry out comprehensive analysis and obtain evaluation of estimate;If in subordinate's commodity and brand library
Some Brand is related, then first, in accordance with the analysis mode in step S103 to all mesh under the subordinate's commodity grabbed
Mark evaluation carries out comprehensive analysis and obtains evaluation of estimate, and obtains corresponding to Brand (subordinate's commodity relevant Brand)
Brand coefficient is weighted operation (product operation) to evaluation of estimate based on the brand coefficient, updated evaluation of estimate is obtained, rear
Recommend subordinate's commodity corresponding with updated evaluation of estimate according to preset order in continuous operation.For example, if it is calculated with
The corresponding evaluation of estimate of subordinate's commodity is 4.5, and subordinate's commodity are related to a Brand in brand library, the Brand
Corresponding brand coefficient is 1.1, then updated evaluation of estimate is 4.5 × 1.1=4.95.
By embodiment illustrated in fig. 5 it is found that in embodiments of the present invention, by crawling user within a preset period of time extremely
Few two browsings record, analyzes the Brand occurred in all browsings record crawled, and by Brand all clear
Ratio-dependent shared in recording of looking at is brand coefficient, if subordinate's commodity are related to a Brand, under subordinate's commodity
All objective appraisals carry out comprehensive analysis and obtain evaluation of estimate, and evaluation of estimate is carried out based on Brand corresponding brand coefficient
Weighting obtains updated evaluation of estimate, the embodiment of the present invention by calculating brand coefficient, and according to brand coefficient to evaluation of estimate into
Row updates, so that the result of commercial product recommending is more bonded with the browsing of user habit, improves the accuracy of commercial product recommending.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Corresponding to the Method of Commodity Recommendation based on user's portrait described in foregoing embodiments, Fig. 6 shows implementation of the present invention
The structural block diagram for the device for recommending the commodity based on user's portrait that example provides, referring to Fig. 6, which includes:
Acquisition unit 61 for acquiring user images, and analyzes the user images to obtain characteristics of human body's value;
Picking unit 62, the commercial product recommending type selected for obtaining user, retrieves the commercial product recommending in the database
At least two subordinate's commodity under type, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
Recommendation unit 63, for subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity to be commented
Valence is determined as objective appraisal, carries out comprehensive analysis to all objective appraisals and obtains evaluation of estimate, and recommends according to preset order
Subordinate's commodity corresponding with institute's evaluation values.
Optionally, if the quantity of objective appraisal is less than preset quantity, and commercial product recommending type is deposited with preset correlated characteristic
In corresponding relationship, recommendation unit 63 includes:
Extraction unit, for extracting associated eigenvalue corresponding with the correlated characteristic in characteristics of human body's value;
Determination unit, for the commodity evaluation relevant to characteristics of human body's value to be determined as first object evaluation,
And the commodity evaluation relevant to the associated eigenvalue is determined as the second objective appraisal;
Weighted sum unit obtains first object value for carrying out comprehensive analysis to all first object evaluations, right
All second objective appraisals carry out comprehensive analysis and obtain the second target value, and to the first object value and second mesh
Scale value is weighted summation and obtains institute's evaluation values.
Optionally, user images user and camera be separated by preset acquisition apart from when be acquired, acquisition unit 61
Include:
Acquiring unit, for obtaining preset feature to be collected, and acquire in the user images with the spy to be collected
Levy corresponding image feature value;
Computing unit, for obtaining the acquisition apart from corresponding first scaling, based on described image characteristic value and
Characteristics of human body's value is calculated in first scaling.
Optionally, including at least two features to be collected, acquisition unit 61 include:
Collection apparatus unit, it is corresponding at least at least two features to be collected in the user images for acquiring
Two image feature values, wherein there are corresponding relationships with a feature to be collected for each described image characteristic value;
Ratio-dependent unit, for obtaining the actual characteristic corresponding with the feature to be collected of user input
Value, and the ratio between actual characteristic value described image characteristic value corresponding with the feature to be collected is determined as second
Scaling;
Assembled unit, for all described image characteristic values individually to be carried out operation with second scaling, and will
All described image eigenvalue clusters after operation are combined into characteristics of human body's value.
Optionally, including at least two Brands, recommendation unit 63 include:
Factor determination unit, for crawling at least two browsings record of the user within a preset period of time, analysis is climbed
The Brand occurred in all browsing records got, and the Brand is recorded in all browsings
In shared ratio-dependent be brand coefficient;
Weighted units, if related to a Brand for subordinate's commodity, under subordinate's commodity
All objective appraisals carry out comprehensive analysis and obtain evaluation of estimate, and be based on the corresponding brand coefficient of the Brand
Institute's evaluation values are weighted to obtain updated institute's evaluation values.
Therefore, the device for recommending the commodity provided in an embodiment of the present invention based on user's portrait is according to collected user images
It is automatic to carry out commercial product recommending, it avoids user and manually sets data, improve the convenience and applicability of commercial product recommending.
Fig. 7 is the schematic diagram of terminal device provided in an embodiment of the present invention.As shown in fig. 7, the terminal device 7 of the embodiment
Include: processor 70, memory 71 and is stored in the calculating that can be run in the memory 71 and on the processor 70
Machine program 72, such as the commercial product recommending program based on user's portrait.The processor 70 executes real when the computer program 72
Step in existing above-mentioned each Method of Commodity Recommendation embodiment based on user's portrait, such as step S101 shown in FIG. 1 is extremely
S103.Alternatively, the processor 70 realizes the above-mentioned commercial product recommending respectively based on user's portrait when executing the computer program 72
The function of each unit in Installation practice, such as the function of unit 61 to 63 shown in Fig. 6.
Illustratively, the computer program 72 can be divided into one or more units, one or more of
Unit is stored in the memory 71, and is executed by the processor 70, to complete the present invention.One or more of lists
Member can be the series of computation machine program instruction section that can complete specific function, and the instruction segment is for describing the computer journey
Implementation procedure of the sequence 72 in the terminal device 7.For example, the computer program 72 can be divided into acquisition unit, grab
Unit and recommendation unit are taken, each unit concrete function is as follows:
Acquisition unit for acquiring user images, and analyzes the user images to obtain characteristics of human body's value;
Picking unit, the commercial product recommending type selected for obtaining user, retrieves the commercial product recommending kind in the database
At least two subordinate's commodity under class, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
Recommendation unit, for by subordinate's commodity are corresponding and relevant to characteristics of human body value commodity evaluation
Be determined as objective appraisal, comprehensive analysis carried out to all objective appraisals and obtain evaluation of estimate, and according to preset order recommend with
The corresponding subordinate's commodity of institute's evaluation values.
The terminal device 7 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The terminal device may include, but be not limited only to, processor 70, memory 71.It will be understood by those skilled in the art that Fig. 7
The only example of terminal device 7 does not constitute the restriction to terminal device 7, may include than illustrating more or fewer portions
Part perhaps combines certain components or different components, such as the terminal device can also include input-output equipment, net
Network access device, bus etc..
Alleged processor 70 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 71 can be the internal storage unit of the terminal device 7, such as the hard disk or interior of terminal device 7
It deposits.The memory 71 is also possible to the External memory equipment of the terminal device 7, such as be equipped on the terminal device 7
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card dodge
Deposit card (Flash Card) etc..Further, the memory 71 can also both include the storage inside list of the terminal device 7
Member also includes External memory equipment.The memory 71 is for storing needed for the computer program and the terminal device
Other programs and data.The memory 71 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit division progress for example, in practical application, can according to need and by above-mentioned function distribution by different functions
Unit is completed, i.e., the internal structure of the terminal device is divided into different functional units, to complete whole described above
Or partial function.Each functional unit in embodiment can integrate in one processing unit, be also possible to each unit list
It is solely physically present, can also be integrated in one unit with two or more units, above-mentioned integrated unit can both use
Formal implementation of hardware can also be realized in the form of software functional units.In addition, the specific name of each functional unit also only
It is the protection scope that is not intended to limit this application for the ease of mutually distinguishing.The specific work process of unit in above system,
It can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed terminal device and method can pass through it
Its mode is realized.For example, terminal device embodiment described above is only schematical, for example, the unit is drawn
Point, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components can
To combine or be desirably integrated into another system, or some features can be ignored or not executed.Another point, it is shown or beg for
The mutual coupling or direct-coupling or communication connection of opinion can be through some interfaces, the INDIRECT COUPLING of device or unit
Or communication connection, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, the present invention realizes above-described embodiment side
All or part of the process in method can also instruct relevant hardware to complete, the computer by computer program
Program can be stored in a computer readable storage medium, and the computer program is when being executed by processor, it can be achieved that above-mentioned each
The step of a embodiment of the method.Wherein, the computer program includes computer program code, and the computer program code can
Think source code form, object identification code form, executable file or certain intermediate forms etc..The computer-readable medium can be with
It include: any entity or device, recording medium, USB flash disk, mobile hard disk, magnetic disk, light that can carry the computer program code
Disk, computer storage, read-only memory (Read-Only Memory, ROM), random access memory (Random Access
Memory, RAM), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that described computer-readable
The content that medium includes can carry out increase and decrease appropriate according to the requirement made laws in jurisdiction with patent practice, such as at certain
A little jurisdictions do not include electric carrier signal and telecommunication signal according to legislation and patent practice, computer-readable medium.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of Method of Commodity Recommendation based on user's portrait characterized by comprising
User images are acquired, and the user images are analyzed to obtain characteristics of human body's value;
The commercial product recommending type that user selectes is obtained, retrieves at least two subordinaties under the commercial product recommending type in the database
Commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as objective appraisal, it is right
All objective appraisals carry out comprehensive analysis and obtain evaluation of estimate, and recommend institute corresponding with institute's evaluation values according to preset order
State subordinate's commodity.
2. Method of Commodity Recommendation as described in claim 1, which is characterized in that if the quantity of the objective appraisal is less than present count
Amount, and the commercial product recommending type and preset correlated characteristic be there are corresponding relationship, it is described by subordinate's commodity it is corresponding and
The commodity evaluation relevant to characteristics of human body's value is determined as objective appraisal, carries out comprehensive point to all objective appraisals
Analysis obtains evaluation of estimate, comprising:
Extract associated eigenvalue corresponding with the correlated characteristic in characteristics of human body's value;
The commodity evaluation relevant to characteristics of human body's value is determined as first object evaluation, and will be with the correlated characteristic
It is worth the relevant commodity evaluation and is determined as the second objective appraisal;
Comprehensive analysis are carried out to all first objects evaluations and obtain first object value, to all second objective appraisals into
Row comprehensive analysis obtains the second target value, and is weighted summation to the first object value and second target value and obtains institute
Evaluation values.
3. Method of Commodity Recommendation as described in claim 1, which is characterized in that the user images are in the user and camera
Be separated by it is preset acquisition apart from when be acquired, it is described that the user images are analyzed to obtain characteristics of human body's value, comprising:
Preset feature to be collected is obtained, and acquires characteristics of image corresponding with the feature to be collected in the user images
Value;
The acquisition is obtained apart from corresponding first scaling, based on described image characteristic value and first scaling
Calculation obtains characteristics of human body's value.
4. Method of Commodity Recommendation as described in claim 1, which is characterized in that described right including at least two features to be collected
The user images are analyzed to obtain characteristics of human body's value, comprising:
Acquire at least two image feature value corresponding at least two features to be collected in the user images, wherein
There are corresponding relationships with a feature to be collected for each described image characteristic value;
Obtain the actual characteristic value corresponding with the feature to be collected of user input, and by the actual characteristic value
Ratio between described image characteristic value corresponding with the feature to be collected is determined as the second scaling;
All described image characteristic values are individually subjected to operation with second scaling, and by all figures after operation
As eigenvalue cluster is combined into characteristics of human body's value.
5. Method of Commodity Recommendation as described in claim 1, which is characterized in that described to institute including at least two Brands
There is the objective appraisal to carry out comprehensive analysis and obtain evaluation of estimate, comprising:
At least two browsings record of the user within a preset period of time is crawled, all browsing records crawled are analyzed
The Brand of middle appearance, and be brand by Brand ratio-dependent shared in all browsing records
Coefficient;
If subordinate's commodity are related to a Brand, to all described under the subordinate's commodity grabbed
Objective appraisal carries out comprehensive analysis and obtains evaluation of estimate, and based on the corresponding brand coefficient of the Brand to the evaluation
Value is weighted to obtain updated institute's evaluation values.
6. a kind of device for recommending the commodity based on user's portrait characterized by comprising
Acquisition unit for acquiring user images, and analyzes the user images to obtain characteristics of human body's value;
Picking unit, the commercial product recommending type selected for obtaining user, is retrieved in the database under the commercial product recommending type
At least two subordinate's commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
Recommendation unit, for by subordinate's commodity are corresponding and relevant to characteristics of human body value commodity evaluation determination
For objective appraisals, comprehensive analysis are carried out to all objective appraisals and obtain evaluation of estimate, and according to preset order recommend with it is described
The corresponding subordinate's commodity of evaluation of estimate.
7. a kind of terminal device, which is characterized in that the terminal device includes memory, processor and is stored in the storage
In device and the computer program that can run on the processor, the processor are realized as follows when executing the computer program
Step:
User images are acquired, and the user images are analyzed to obtain characteristics of human body's value;
The commercial product recommending type that user selectes is obtained, retrieves at least two subordinaties under the commercial product recommending type in the database
Commodity, and grab at least two commodity evaluation corresponding with each subordinate's commodity;
It is subordinate's commodity are corresponding and relevant to characteristics of human body's value commodity evaluation is determined as objective appraisal, it is right
All objective appraisals carry out comprehensive analysis and obtain evaluation of estimate, and recommend institute corresponding with institute's evaluation values according to preset order
State subordinate's commodity.
8. terminal device as claimed in claim 7, which is characterized in that if the quantity of the objective appraisal is less than preset quantity,
And the commercial product recommending type and preset correlated characteristic be there are corresponding relationship, it is described by subordinate's commodity it is corresponding and with institute
It states the relevant commodity evaluation of characteristics of human body's value and is determined as objective appraisal, comprehensive analysis is carried out to all objective appraisals and is obtained
To evaluation of estimate, comprising:
Extract associated eigenvalue corresponding with the correlated characteristic in characteristics of human body's value;
The commodity evaluation relevant to characteristics of human body's value is determined as first object evaluation, and will be with the correlated characteristic
It is worth the relevant commodity evaluation and is determined as the second objective appraisal;
Comprehensive analysis are carried out to all first objects evaluations and obtain first object value, to all second objective appraisals into
Row comprehensive analysis obtains the second target value, and is weighted summation to the first object value and second target value and obtains institute
Evaluation values.
9. terminal device as claimed in claim 7, which is characterized in that including at least two Brands, described pair of all institute
It states objective appraisal progress comprehensive analysis and obtains evaluation of estimate, comprising:
At least two browsings record of the user within a preset period of time is crawled, all browsing records crawled are analyzed
The Brand of middle appearance, and be brand by Brand ratio-dependent shared in all browsing records
Coefficient;
If subordinate's commodity are related to a Brand, to all objective appraisals under subordinate's commodity
It carries out comprehensive analysis and obtains evaluation of estimate, and institute's evaluation values are added based on the Brand corresponding brand coefficient
Power obtains updated institute's evaluation values.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization Method of Commodity Recommendation as described in any one of claim 1 to 5 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910039148.9A CN109859008A (en) | 2019-01-16 | 2019-01-16 | Method of Commodity Recommendation, device and terminal device based on user's portrait |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910039148.9A CN109859008A (en) | 2019-01-16 | 2019-01-16 | Method of Commodity Recommendation, device and terminal device based on user's portrait |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109859008A true CN109859008A (en) | 2019-06-07 |
Family
ID=66894872
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910039148.9A Pending CN109859008A (en) | 2019-01-16 | 2019-01-16 | Method of Commodity Recommendation, device and terminal device based on user's portrait |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109859008A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113469790A (en) * | 2021-07-05 | 2021-10-01 | 深圳市点购电子商务控股股份有限公司 | Evaluation order recommendation method and device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105447058A (en) * | 2014-09-29 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Auxiliary information offering method and device |
CN106709781A (en) * | 2016-12-05 | 2017-05-24 | 姚震亚 | Personal image design and collocation purchasing device and method |
US20190238488A1 (en) * | 2016-08-04 | 2019-08-01 | Tencent Technology (Shenzhen) Company Limited | Information processing method, apparatus, device and storage medium |
-
2019
- 2019-01-16 CN CN201910039148.9A patent/CN109859008A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105447058A (en) * | 2014-09-29 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Auxiliary information offering method and device |
US20190238488A1 (en) * | 2016-08-04 | 2019-08-01 | Tencent Technology (Shenzhen) Company Limited | Information processing method, apparatus, device and storage medium |
CN106709781A (en) * | 2016-12-05 | 2017-05-24 | 姚震亚 | Personal image design and collocation purchasing device and method |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113469790A (en) * | 2021-07-05 | 2021-10-01 | 深圳市点购电子商务控股股份有限公司 | Evaluation order recommendation method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109685631B (en) | Personalized recommendation method based on big data user behavior analysis | |
CN109241415B (en) | Project recommendation method and device, computer equipment and storage medium | |
CN109189991A (en) | Repeat video frequency identifying method, device, terminal and computer readable storage medium | |
CN109559208A (en) | A kind of information recommendation method, server and computer-readable medium | |
CN110325983A (en) | Image retrieving apparatus and image search method | |
CN111709810A (en) | Object recommendation method and device based on recommendation model | |
CN107958007A (en) | Case information search method and device | |
US11386301B2 (en) | Cluster and image-based feedback system | |
CN105243124B (en) | Combination of resources processing method and processing device | |
CN108632311A (en) | Information-pushing method and device | |
CN109447729A (en) | A kind of recommended method of product, terminal device and computer readable storage medium | |
CN110111167A (en) | A kind of method and apparatus of determining recommended | |
CN110458360A (en) | Prediction technique, device, equipment and the storage medium of hot resource | |
CN110516163B (en) | Commodity ordering method and system based on user behavior data | |
CN107153656A (en) | A kind of information search method and device | |
CN111949887A (en) | Item recommendation method and device and computer-readable storage medium | |
CN112417294A (en) | Intelligent business recommendation method based on neural network mining model | |
CN113761359B (en) | Data packet recommendation method, device, electronic equipment and storage medium | |
CN108416627A (en) | A kind of brand influence force monitoring method and system based on internet data | |
CN110191183A (en) | Accurate intelligent method for pushing, system, device and computer readable storage medium | |
CN112380449B (en) | Information recommendation method, model training method and related device | |
CN108664515A (en) | A kind of searching method and device, electronic equipment | |
CN113689259A (en) | Commodity personalized recommendation method and system based on user behaviors | |
CN106202216A (en) | A kind of data processing method and data handling system | |
CN111666513A (en) | Page processing method and device, electronic equipment and readable 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |
|
RJ01 | Rejection of invention patent application after publication |