CN115905681A - Shop recommendation method, server, display method, client and system - Google Patents
Shop recommendation method, server, display method, client and system Download PDFInfo
- Publication number
- CN115905681A CN115905681A CN202211268467.5A CN202211268467A CN115905681A CN 115905681 A CN115905681 A CN 115905681A CN 202211268467 A CN202211268467 A CN 202211268467A CN 115905681 A CN115905681 A CN 115905681A
- Authority
- CN
- China
- Prior art keywords
- target
- feature vector
- shop
- reason
- combination
- 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 64
- 239000013598 vector Substances 0.000 claims abstract description 224
- 238000012545 processing Methods 0.000 claims description 56
- 238000006243 chemical reaction Methods 0.000 claims description 24
- 238000004590 computer program Methods 0.000 claims description 17
- 230000004044 response Effects 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 description 41
- 238000012549 training Methods 0.000 description 21
- 238000010586 diagram Methods 0.000 description 18
- 238000004891 communication Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 14
- 230000000694 effects Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 8
- 230000014509 gene expression Effects 0.000 description 7
- 238000013507 mapping Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 239000004973 liquid crystal related substance Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 1
- 229910052802 copper Inorganic materials 0.000 description 1
- 239000010949 copper Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a shop recommendation method, a server, a display method, a client and a system, wherein the method comprises the following steps: responding to a shop recommendation request aiming at a target user sent by a client, and acquiring a target shop to be displayed and a plurality of recommendation reason and literature combinations of the target shop; wherein, each recommended reason and document combination comprises at least two recommended reason and documents with an arrangement sequence; acquiring a target user characteristic vector corresponding to the target user, a target shop characteristic vector corresponding to the target shop and a combined characteristic vector corresponding to each recommended reason and file combination; according to the target user feature vector, the target shop feature vector and the combined feature vector, determining a recommended reason and case combination matched with the target user and the target shop as a target case combination; and pushing the target file combination and the shop information of the target shop to a client side for the client side to display.
Description
Technical Field
The present disclosure relates to the field of recommendation technologies, and in particular, to a store recommendation method based on a recommendation reason, a server, a display method of store information, a client, and a store recommendation system based on a recommendation reason.
Background
With the development of the information-oriented society, people are more and more accustomed to using a network to search or select a consuming store, and some recommendation reasons are generally displayed to users together with the store, but in the prior art, a file with some recommendation reasons edited in advance can be used at the time of display.
In the prior art, a background server may push a plurality of recommendation reason documents of a store to a client for the store, and the client displays the received recommendation reason documents according to a preset sorting order. The recommendation reason document is mainly used for highlighting characteristics of merchants and assisting users in making consumption decisions.
However, the total length of the recommended reason documents may exceed the width of the interface of the client, and at least partially exceed the recommended reason documents of the interface, so that the client will not display the recommended reason documents, and the display area of the recommended reason documents is blank.
In the example shown in fig. 1a to 1c, the backend server pushes four recommendation reason documents including document 1, document 2, document 3, and document 4 to the client for one store. In the example shown in fig. 1a, the four recommended reason documents are sorted according to the sorting scheme 1, where the sorting scheme 1 may be document 1, document 2, document 3, and document 4, and then only document 1 and document 2 may be shown in the shop card of the client. In the example shown in fig. 1b, the four recommended reason documents are sorted according to the sorting scheme 2, the sorting scheme 2 can be document 1, document 2, document 4 and document 3, and then only document 1, document 2 and document 4 can be shown in the shop card of the client. In the example shown in fig. 1c, the four recommended reason documents are sorted according to the sorting scheme 3, the sorting scheme 3 can be the documents 2, 3, 4 and 1, and then only the documents 2, 3 and 4 can be shown in the shop card of the client.
Disclosure of Invention
An object of the present disclosure is to provide a new technical solution capable of improving the presentation effect of a recommended reason document.
According to a first aspect of the present disclosure, there is provided a store recommendation method based on a store recommendation reason, the method being used for a server, the method including:
responding to a shop recommendation request aiming at a target user sent by a client, and acquiring a target shop to be displayed and a plurality of recommendation reason and file combinations of the target shop; wherein, each recommended reason and document combination comprises at least two recommended reason and documents with an arrangement sequence;
acquiring a target user characteristic vector corresponding to the target user, a target shop characteristic vector corresponding to the target shop and a combined characteristic vector corresponding to each recommended reason and file combination;
according to the target user feature vector, the target shop feature vector and the combined feature vector, determining a recommended reason and case combination matched with the target user and the target shop as a target case combination;
and pushing the target file combination and the shop information of the target shop to a client side for the client side to display.
Optionally, obtaining a combination feature vector corresponding to the recommended reason and document combination includes:
obtaining the case characteristic vector of each recommended reason case in the recommended reason case combination;
and splicing the file characteristic vectors of the recommended reason files in the recommended reason file combination to obtain a combination characteristic vector corresponding to the recommended reason file combination.
Optionally, the obtaining of the plurality of recommended reason and literature combinations includes:
acquiring a plurality of recommended reason documents of the target shop;
and obtaining a plurality of recommended reason record combinations according to the interface size of the client and the length of each recommended reason record.
Optionally, the obtaining a plurality of recommended reason document combinations according to the interface size of the client and the length of each recommended reason document includes:
carrying out permutation and combination on the plurality of recommended reason and patterns to obtain a plurality of permutation and combination results;
and determining at least two recommended reason documents to be displayed in the interface of the client in each permutation and combination result as the recommended reason document combination according to the interface size of the client and the length of each recommended reason document.
Optionally, the dimensions of the target user feature vector and the target store feature vector are both target dimensions;
determining a recommended reason and case combination matched with the target user and the target store according to the target user feature vector, the target store feature vector and the combined feature vector, wherein the determining is used as a target case combination and comprises the following steps:
performing target dimension conversion processing on the combined feature vector to obtain a combined feature vector of the target dimension;
performing a point multiplication operation on the target user characteristic vector, the target store characteristic vector and the combined characteristic vector of the target dimension to obtain a prediction matching score of the target user, the target store and each recommended reason and literature combination;
and selecting a recommended reason and case combination with the maximum predicted matching score of the target user and the target shop as the target case combination.
Optionally, the obtaining of the user feature vector corresponding to the target user and the store feature vector corresponding to the target store includes:
acquiring a user identifier of the target user and a shop identifier of the target shop;
according to the user identification, obtaining a user characteristic vector corresponding to the target user from pre-stored user characteristic vectors of a plurality of users as the target user characteristic vector;
and acquiring a shop feature vector corresponding to the target shop from pre-stored shop feature vectors of a plurality of shops as the target shop feature vector according to the shop identification.
Optionally, before the obtaining of the user feature vector corresponding to the target user and the store feature vector corresponding to the target store, the method further includes:
acquiring a first feature vector of the target user, wherein the first feature vector comprises a plurality of user features reflecting the preference of the target user on stores and recommendation reason documents;
performing first dimension conversion processing on the first eigenvector to obtain a second eigenvector of a target dimension; performing second dimension conversion processing on the first feature vector to obtain a third feature vector of a target dimension;
summing the second feature vector and the third feature vector to obtain the feature vector of the target user;
acquiring a fourth feature vector of the target store, wherein the fourth feature vector comprises a plurality of store features influencing the preference of the target user to the target store;
performing third dimension conversion processing on the fourth feature vector to obtain a fifth feature vector of a target dimension; performing fourth dimension conversion processing on the fourth feature vector to obtain a sixth feature vector of a target dimension;
summing the fifth feature vector and the sixth feature vector to obtain the feature vector of the target shop;
and storing the target user characteristic vector and the target shop characteristic vector.
According to a second aspect of the present disclosure, there is provided a method for displaying store information, the method being used for a client, the method including:
responding to a target operation executed by a target user through the client, sending a shop recommendation request aiming at the target user to a server, so that the server determines a recommendation reason and literature combination matched with the target user and the target shop as a target literature combination in response to the shop recommendation request, and pushing the target literature combination and shop information of the target shop to the client; wherein, each recommended reason and document combination comprises at least two recommended reason and documents with an arrangement sequence;
receiving the target file combination and the shop information of the target shop;
and displaying the combination of the shop information of the target shop and the target file in the shop card corresponding to the target shop.
Optionally, the displaying the target pattern combination comprises:
and displaying the recommended reason file contained in the target file combination in a full frame in a recommended reason display frame of the shop card.
According to a third aspect of the present disclosure, there is provided a server comprising a first memory for storing an executable first computer program and a first processor; the first computer program is for controlling the first processor to perform the method according to the first aspect of the disclosure.
According to a fourth aspect of the present disclosure, there is provided a client comprising a second processor and a second memory for storing an executable second computer program; the second computer program is for controlling the second processor to perform the method according to the second aspect of the disclosure.
According to a fifth aspect of the present disclosure, there is provided a store recommendation system based on recommendation reasons, including the server according to the third aspect of the present disclosure, and the client according to the fourth aspect of the present disclosure.
Through the embodiment of the disclosure, the plurality of recommendation reason documents recommended to the target user can occupy the corresponding display positions in the interface of the client to the maximum extent, and the display effect of the recommendation reason documents is improved. In addition, when the target store has a plurality of recommended reason documents, the target user may be individually ranked in order of the recommended reason documents, and may be individually recommended in combination of the recommended reason documents.
Other features of the present disclosure and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1a is a diagram illustrating the display effect of the recommendation reason documents sorted by the sorting scheme 1 in the prior art in the client;
FIG. 1b is a diagram illustrating the display effect of the recommended reason documents sorted according to the sorting scheme 2 in the prior art on the client;
FIG. 1c is a diagram illustrating the display effect of the recommendation reason documents sorted according to the sorting scheme 3 in the prior art on the client;
FIG. 2 is a block diagram showing an example of a hardware configuration of a vehicle system that may be used to implement embodiments of the present disclosure;
FIG. 3 illustrates a flow chart of a store recommendation method based on a recommendation reason according to an embodiment of the disclosure;
FIG. 4 is a schematic diagram illustrating one example of a store recommendation method based on a recommendation reason according to an embodiment of the disclosure;
FIG. 5 shows a schematic block diagram of a server of an embodiment of the present disclosure;
FIG. 6 shows a flow chart of a method of displaying store information of an embodiment of the present disclosure;
fig. 7 shows a schematic block diagram of a client of an embodiment of the present disclosure;
FIG. 8 shows a schematic block diagram of a recommendation reason based store recommendation system of an embodiment of the present disclosure.
Detailed Description
Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as merely illustrative, and not limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be discussed further in subsequent figures.
< hardware configuration >
As shown in fig. 2, the recommendation system 100 includes a feature server 1000, a client 2000, a recommendation server 3000, and a network 4000.
In one example, feature server 1000 may be as shown in FIG. 2, including a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600. Although the server may also include speakers, microphones, and the like, these components are reasonably irrelevant to the disclosure and are omitted here.
The processor 1100 may be, for example, a central processing unit CPU, a microprocessor MCU, or the like. The memory 1200 includes, for example, a ROM (read only memory), a RAM (random access memory), a nonvolatile memory such as a hard disk, and the like. The interface device 1300 includes, for example, a USB interface, a serial interface, an infrared interface, and the like. Communication device 1400 is capable of wired or wireless communication, for example. The display device 1150 is, for example, a liquid crystal display panel, an LED display panel touch display panel, or the like. Input devices 1160 may include, for example, a touch screen, a keyboard, and the like.
In the present embodiment, the client 2000 is an electronic device having a communication function and a service processing function. The client 2000 may be a mobile terminal, such as a mobile phone, a laptop, a tablet, a palmtop, etc. In one example, the client 2000 is a device that performs management operations on the vehicle 3000, such as a cell phone with a designated Application (APP) installed.
As shown in fig. 2, the client 2000 may include a processor 2100, a memory 2200, an interface device 2300, a communication device 2400, a display device 2500, an input device 2600, a speaker 2700, a microphone 2800, and the like. The processor 2100 may be a central processing unit CPU, a microprocessor MCU, or the like. The memory 2200 includes, for example, a ROM (read only memory), a RAM (random access memory), a nonvolatile memory such as a hard disk, and the like. The interface device 2300 includes, for example, a USB interface, a headphone interface, and the like. Communication device 2400 is capable of wired or wireless communication, for example. The display device 2500 is, for example, a liquid crystal display panel, a touch panel, or the like. The input device 2600 may include, for example, a touch screen, a keyboard, and the like. A user can input/output voice information through the speaker 2700 and the microphone 2800.
In one example, the recommendation server 3000 may include a processor 3100, a memory 3200, an interface device 3300, a communication device 3400, a display device 3500, and an input device 3600, as shown in fig. 2. Although the server may also include speakers, microphones, and the like, these components are reasonably irrelevant to the present disclosure and are omitted here.
The processor 3100 may be, for example, a central processing unit CPU, a microprocessor MCU, or the like. The memory 3200 includes, for example, a ROM (read only memory), a RAM (random access memory), a nonvolatile memory such as a hard disk, and the like. The interface device 3300 includes, for example, a USB interface, a serial interface, an infrared interface, and the like. The communication device 3400 can perform wired or wireless communication, for example. The display device 3500 is, for example, a liquid crystal display panel, an LED display panel touch display panel, or the like. The input device 3600 may include, for example, a touch screen, a keyboard, and the like.
The network 4000 may be a wireless communication network or a wired communication network, and may be a local area network or a wide area network. In the recommendation system shown in fig. 2, the recommendation server 3000 and the feature server 1000, and the client 2000 and the recommendation server 3000 may communicate with each other through the network 4000. The recommendation server 3000 may be the same as or different from the feature server 1000, and the network 4000 through which the client 2000 communicates with the recommendation server 3000.
It should be understood that although fig. 2 shows only one feature server 1000, client 2000, recommendation server 3000, it is not meant to limit the corresponding number, and multiple feature servers 1000, multiple clients 2000, multiple recommendation servers 3000 may be included in the recommendation system 100.
The recommendation system 100 shown in fig. 2 is merely illustrative and is in no way intended to limit the present disclosure, its applications, or uses.
In an embodiment of the present disclosure, the memory 1200 of the feature server 1000 is used for storing instructions for controlling the processor 1100 to operate so as to execute the method implemented by the feature server provided by the embodiment of the present disclosure.
Although a number of devices are shown for feature server 1000 in fig. 2, the present disclosure may refer to only some of the devices, for example, feature server 1000 refers to only memory 1200 and processor 1100.
In an embodiment of the present disclosure, the memory 2200 of the client 2000 is configured to store instructions for controlling the processor 2100 to execute the client 2000 to perform the method implemented by the client provided by the embodiment of the present disclosure.
Although multiple devices are shown in fig. 2 for client 2000, the disclosure may refer to only some of the devices, e.g., client 2000 refers only to memory 2200 and processor 2100.
In an embodiment of the present disclosure, the memory 3200 of the recommendation server 3000 is configured to store instructions for controlling the processor 3100 to operate so as to perform the method implemented by the recommendation server provided by the embodiment of the present disclosure.
Although a plurality of devices are shown for the recommendation server 3000 in fig. 2, the present disclosure may refer to only some of the devices, for example, the recommendation server 3000 refers to only the memory 3200 and the processor 3100.
In the above description, the skilled person can design the instructions according to the disclosed solution of the present disclosure. How the instructions control the operation of the processor is well known in the art and will not be described in detail here.
< first embodiment >
< method >
The method for recommending the store based on the store recommendation reason provided in the embodiment can be implemented by a server. In one example, the server may include recommendation server 3000 as shown in FIG. 2. In another example, the server may also include a feature server 1000 and a recommendation server 3000 as shown in FIG. 2.
As shown in fig. 3, the store recommendation method based on the store recommendation reason includes steps S3100 to S3400.
Step S3100, responding to a shop recommendation request aiming at a target user sent by a client, and acquiring a target shop to be displayed and a plurality of recommendation reason and file combinations of the target shop; wherein, each recommended reason and case combination comprises at least two recommended reason and cases with an arrangement sequence.
In one embodiment, the client may issue a store recommendation request for a target user to the server in response to the target user performing a target operation through the client.
Wherein the target user may be a user who logs in the target application through the client.
The target operation in this embodiment may be an operation of the target user opening the target application program, or an operation of the target user opening the target page of the target application program.
In one embodiment of the present disclosure, store feature vectors of a plurality of stores may be stored in advance in the feature server.
Further, the store recommendation request for the target user may carry a user identifier of the target user.
Then, in response to the store recommendation request, obtaining the target store to be exhibited may include: and sending the user identification of the target user to the feature server, and receiving the target shop returned by the feature server. The feature server may obtain a target user feature vector corresponding to the target user according to the user identifier sent by the recommendation server, obtain a target store to be displayed from the multiple stores according to the target user feature vector and pre-stored store feature vectors of the multiple stores, and return the target store to the recommendation server.
Alternatively, store feature vectors of a plurality of stores may be stored in advance in the recommendation server, the recommendation server may acquire a target user feature vector corresponding to a target user in response to a store recommendation request, and the target store to be displayed may be acquired from the plurality of stores based on the target user feature vector and the store feature vectors of the plurality of stores stored in advance.
In this embodiment, a preset store recommendation model may be used, and a store preference score of a target user for each store pair is obtained according to a target user feature vector and pre-stored store feature vectors of a plurality of stores; and selecting the shops with the descending sorting order of the shop preference scores according with the preset sorting range as target shops.
The target user is used for defining the preference degree of the target user to the shop, the shop with the descending sequence of the preference degrees of the shop according with the preset sequence range is selected as the target shop to be displayed, so that the target shop is the shop with the higher preference degree of the target user, the preference of the user to the shop can be accurately and effectively met, and the requirement of the user for obtaining the shop is actually met.
The preset sequencing range may be set according to a specific application scenario or application requirements, for example, the preset sequencing range may be set to 1-10.
In another embodiment of the disclosure, the obtaining of the target store to be exhibited in response to the store recommendation request may include at least one of:
selecting a first number of shops closest to a target user from the plurality of shops as target shops;
selecting a second quantity of shops with highest benefits in a first history time period from the multiple shops as target shops;
and selecting a third quantity of shops with the highest click rate in a second historical period from the multiple shops as target shops.
The first number, the second number, and the third number may be respectively set in advance according to an application scenario or a specific requirement, and the first number, the second number, and the third number may be equal or unequal, which is not limited herein. For example, the first number, the second number, and the third number may each be 10.
The first history time period and the second history time period in this embodiment may be preset history time periods according to application scenarios or specific requirements, and the first history time period and the second history time period may be the same or different. For example, the first historical period may be the past week and the second historical period may be the past 3 days.
In this embodiment, a plurality of recommended reason document combinations for each store may be stored in advance, and when a target store is acquired, a plurality of recommended reason document combinations for the target store may be acquired. Wherein, each recommendation reason document combination comprises at least two recommendation reason documents with an arrangement sequence.
In one embodiment of the disclosure, obtaining a plurality of recommended reason and document combinations of the target store may include: acquiring a plurality of recommended reason documents of a target store; and obtaining a plurality of recommended reason file combinations according to the interface size of the client and the length of each recommended reason file.
In this embodiment, all the recommendation reason documents in each recommendation reason document combination of the target store can be displayed on the client.
In one example, the length of the recommendation reason document may be determined based on the number of words of the recommendation reason document. Specifically, mapping data reflecting the mapping relationship between the word number of the recommended reason pattern and the length of the recommended reason pattern may be set in advance; the length of the recommended reason pattern can be obtained from the mapping data and the number of words of the recommended reason pattern. The mapping data may be a look-up table or a mapping function.
In one embodiment, obtaining a plurality of recommended reason document combinations according to the interface size of the client and the length of each recommended reason document may include: carrying out permutation and combination on the plurality of recommended reason documents to obtain a plurality of permutation and combination results; and determining at least two recommended reason documents to be displayed in the interface of the client in each arrangement combination result as recommended reason document combinations according to the interface size of the client and the length of each recommended reason document.
When the number of recommended reason documents of the target store is n, n! And (4) arranging and combining the results.
In this embodiment, the width of the display area of the recommended reason document in the interface of the client can be obtained according to the interface size of the client. Specifically, the width of the display area of the recommended reason document in the interface of the client may be obtained by storing the width ratio of the display area of the recommended reason document to the width of the interface of the client in advance, and obtaining the width of the display area of the recommended reason document in the interface of the client based on the interface size of the client and the width ratio.
Each recommendation reason document in the recommendation reason document combination in this embodiment can be completely displayed in the interface of the client. That is, for each permutation and combination, the first m recommendation reason documents can be completely displayed in the interface of the client, and the m-th to n-th recommendation reason documents cannot be displayed in the interface of the client. Then, the sum of the lengths of the first m recommended reason documents is less than or equal to the width of the display area of the recommended reason document in the interface of the client, and the sum of the lengths of the first m +1 recommended reason documents is greater than the width of the display area of the recommended reason document in the interface of the client.
For example, in the case where the recommended reason documents of the target store include document 1, document 2, document 3, and document 4, the ranking combination result may include: case 1, case 2, case 3, and case 4; case 1, case 3, case 2, and case 4; case 1, case 2, case 4 and case 3; case 1, case 4, case 3, and case 2; case 1, case 4, case 2 and case 3; case 2, case 1, case 3 and case 4; case 2, case 1, case 4 and case 3; case 2, case 3, case 1 and case 4; case 2, case 3, case 4, and case 1; case 2, case 4, case 1 and case 3; case 2, case 3, and case 1; case 3, case 1, case 2, and case 4; case 3, case 1, case 4 and case 2; case 3, case 2, case 4 and case 1; case 3, case 2, case 1 and case 4; case 3, case 4, case 2, and case 1; case 3, case 4, case 1 and case 2; case 4, case 3, case 2, and case 1; case 4, case 3, case 1 and case 2; case 4, case 2, case 3, and case 1; case 4, case 2, case 1 and case 3; case 4, case 1, case 2 and case 3; case 4, case 1, case 3 and case 2.
In the case where the result of the permutation and combination is the cases 1, 2, 3 and 4 as shown in fig. 1a, the recommended reason case combination obtained may include the cases 1 and 2. In the case of the case 1, case 2, case 4, and case 3 shown in fig. 1b as the result of the permutation and combination, the recommended reason case combination obtained may include the case 1, case 2, and case 4. In the case of the documents 2, 3, 4 and 1 shown in fig. 1c as the result of the permutation and combination, the obtained recommended reason document combination may include the documents 2, 3 and 4.
In another embodiment, obtaining a plurality of recommended reason document combinations according to the interface size of the client and the length of each recommended reason document may further include: and randomly and sequentially selecting at least two recommended reason documents with the sum of the lengths smaller than or equal to the width of a display area of the recommended reason documents in the interface of the client to obtain a recommended reason document combination.
In another embodiment of the present disclosure, all of the recommended reason documents for the targeted store may be included in each recommended reason document combination. Then, a plurality of recommended reason documents may be arranged and combined to obtain a plurality of arrangement and combination results, and each arrangement and combination result may be regarded as one recommended reason document combination.
Step S3200, obtain target user feature vector corresponding to the target user, target shop feature vector corresponding to the target shop, and combined feature vector corresponding to each recommended reason and literature combination.
In one embodiment of the present disclosure, user feature vectors of a plurality of users (including target users) and store feature vectors of a plurality of stores (including target stores) may be stored in the feature server or the recommendation server in advance. And each user feature vector has a unique user identifier corresponding to the user, and each shop feature vector has a unique shop identifier corresponding to the shop, so that the user feature vector of the corresponding user can be conveniently searched according to the user identifier, and the shop feature vector of the corresponding shop can be conveniently searched according to the shop identifier.
Then, obtaining a target user feature vector corresponding to the target user may include: and acquiring a user identifier of the target user, and searching a user characteristic vector corresponding to the target user from the characteristic server according to the user identifier of the target user to be used as the target user characteristic vector. The method for obtaining the target shop feature vector corresponding to the target shop may include: and acquiring a shop identification of the target shop, and searching a shop feature vector corresponding to the target shop from the feature server according to the shop identification of the target shop to be used as a target shop feature vector.
The dimensions of the target user feature vector and the target shop feature vector in the embodiment are both target dimensions. The target dimension may be set in advance according to an application scenario or specific requirements, for example, the target dimension may be 32 dimensions.
In the embodiment where the recommendation server stores the user feature vectors of the multiple users and the store feature vectors of the multiple stores, before the step S3200 is executed to obtain the target user feature vector corresponding to the target user and the target store feature vector corresponding to the target store, the method may further include the step of generating and storing the target user feature vector and the target store feature vector, and specifically may include the following steps S4110 to S4190:
step S4110, a first feature vector of the target user is obtained. Wherein the first feature vector comprises a plurality of user features reflecting preferences of the target user for stores and recommendation reason documents.
The user characteristics in this embodiment may be set in advance according to an application scenario or specific requirements, and may include, for example, the age of the user, the gender of the user, a region to which the user belongs, a first set number of store types most frequently selected by the user, and a second set number of product types most frequently selected by the user.
The first feature vector X1 includes user features X1 reflecting the preferences of the target user for stores and recommendation reason documents j J takes a value from 1 to n, and n represents the total number of user features of the first feature vector X1.
In one example, the feature vector X1 may have 80 features, i.e., n =80, and at this time, the first feature vector X1 may be represented as X1= (X1) 1 ,x1 2 ,x1 3 ,……,x1 79 ,x1 80 )。
Step S4120, perform first dimension conversion processing on the first eigenvector to obtain a second eigenvector of the target dimension. As shown in fig. 4.
In this embodiment, a first dimension conversion process may be performed on the first feature vector based on the trained first machine learning model to obtain a second feature vector of the target dimension. The trained first machine learning model may be a first feature vector of n dimensions, which is converted into a second feature vector of a target dimension. For example, the trained first machine learning model may be a transform model.
Step S4130, perform second dimension conversion processing on the first feature vector to obtain a third feature vector of the target dimension. As shown in fig. 4.
In this embodiment, a second dimension conversion process may be performed on the first feature vector based on the trained second machine learning model to obtain a third feature vector of the target dimension. The second trained machine learning model may be a third feature vector that converts the n-dimensional first feature vector into a target dimension. Wherein the trained second machine learning model is not the same model as the trained first machine learning model. For example, the trained second machine learning model may be a deep fm model.
And step S4140, summing the second feature vector and the third feature vector to obtain a target user feature vector. As shown in fig. 4.
Step S4150, a fourth feature vector of the target store is obtained, where the fourth feature vector includes a plurality of store features that influence the preference of the target user for the target store.
The store characteristics in the present embodiment may be set in advance according to an application scenario or specific requirements, and may include, for example, a store type, a commodity type of a third set number of commodities having a highest sales volume in the store, a location of the store, a commodity type of a fourth set number of commodities having a highest evaluation by a user in the store, and the like.
The fourth feature vector X4 includes a store feature X4 that affects the target user's preference for the target store i The value of i is a natural number from 1 to m, and m represents the total number of store features of the fourth feature vector X4.
In one example, the feature vector X4 may have 30 features, i.e., m =30, and at this time, the fourth feature vector X4 may be represented as X4= (X4) 1 ,x4 2 ,x4 3 ,……,x4 29 ,x4 30 )。
Step S4160, perform third dimension conversion processing on the fourth feature vector to obtain a fifth feature vector of the target dimension. As shown in fig. 4.
In this embodiment, a third dimension conversion process may be performed on the fourth feature vector based on a trained third machine learning model to obtain a fifth feature vector of the target dimension. The trained third machine learning model may be a fifth feature vector that converts the m-dimensional fourth feature vector into the target dimension. For example, the trained third machine learning model may be a Transformer model.
Step S4170, based on the trained fourth machine learning model, perform fourth dimension conversion processing on the fourth feature vector to obtain a sixth feature vector of the target dimension. As shown in fig. 4.
In this embodiment, a fourth dimension conversion process may be performed on the fourth feature vector based on a trained fourth machine learning model, so as to obtain a sixth feature vector of the target dimension. The trained fourth machine learning model may be a sixth feature vector of the target dimension by converting the fourth feature vector of the m dimension. Wherein the trained fourth machine learning model and the trained third machine learning model are not the same model. For example, the trained fourth machine learning model may be a deep fm model.
Further, the trained third machine learning model is not the same as the trained first machine learning model, and the trained fourth machine learning model is not the same as the trained second machine learning model.
And step S4180, summing the fifth feature vector and the sixth feature vector to obtain a target shop feature vector. As shown in fig. 4.
Step S4190 stores the target user feature vector and the target store feature vector.
In this embodiment, the first feature vector is subjected to first dimension conversion processing and second dimension conversion processing, so as to obtain a second feature vector and a third feature vector of a corresponding target dimension, and then the second feature vector and the third feature vector of the target dimension are summed, so as to obtain a target user feature vector of the target dimension; and performing third-dimension conversion processing and fourth-dimension conversion processing on the fourth feature vector respectively to obtain a fifth feature vector and a sixth feature vector of a corresponding target dimension, and summing the fifth feature vector and the sixth feature vector of the target dimension to obtain a target shop feature vector of the target dimension, so that when a target case selected based on the target user feature vector and the target shop feature vector is subsequently combined, the obtained target case combination has a better effect, and the accuracy of the recommendation reason case pushed to the client is improved.
In addition, the present embodiment generates and stores the target user feature vector and the target shop vector in advance, and calls them directly when step S3200 is executed, so that the recommendation calculation amount can be reduced, the recommendation speed can be increased, and the calculation pressure of the server can be reduced.
In an embodiment of the present disclosure, before performing step S3200 to obtain the target user feature vector corresponding to the target user and the target store feature vector corresponding to the target store, the method may further include steps S4210 to S4280 as follows:
step S4210, a plurality of training samples are obtained, each training sample including a first feature vector of a sample user, a fourth feature vector of a sample store, and a combined feature vector of a recommended reason and literature combination of the sample store.
Step S4220, acquiring an actual matching score of each training sample, wherein the actual matching score is the click rate of the sample user on the recommended reason and case combination of the sample store, or the income displayed by the recommended reason and case combination of every thousand times of the sample store.
In the embodiment that the actual matching score is the click rate of the sample user on the recommended reason and literature combination of the sample store, the method may include the steps of obtaining browsing click log data of the sample user in advance, determining the display times of the recommended reason and literature combination of the sample store to the sample user according to the browsing click log data of the sample user, clicking the click times of the recommended reason and literature combination of the sample store by the sample user, and calculating a quotient obtained by dividing the click times by the display times, wherein the quotient is the click rate of the recommended reason and literature combination of the sample store by the sample user. The number of clicks of the sample user for clicking on the recommended reason and literature combination of the sample store may be the sum of the number of clicks of all recommended reason and literature combinations of the sample user for clicking on the recommended reason and literature combination of the sample store.
Step S4230, processing the first feature vector of the training sample according to the initial first machine learning model to obtain a first processing result.
In this embodiment, the first to-be-determined coefficient of the initial first machine learning model may be used as a variable, and the processing expression of the training sample may be determined as the first processing result according to the first feature vector of the training sample.
Step S4240, processing the first feature vector of the training sample according to the initial second machine learning model to obtain a second processing result.
In this embodiment, the second predetermined coefficient of the initial second machine learning model may be used as a variable, and the processing expression of the training sample may be determined as the second processing result according to the first feature vector of the training sample.
And step S4250, processing the fourth feature vector of the training sample according to the initial third machine learning model to obtain a third processing result.
In this embodiment, the third predetermined coefficient of the initial third machine learning model may be used as a variable, and the processing expression of the training sample may be determined as the third processing result according to the fourth feature vector of the training sample.
Step S4260, processing a fourth feature vector of the training sample according to the initial fourth machine learning model to obtain a fourth processing result.
In this embodiment, a fourth pending coefficient of the initial fourth machine learning model may be used as a variable, and a processing expression of the training sample may be determined as a fourth processing result according to a fourth feature vector of the training sample.
Step S4270, summing the first processing result and the second processing result to obtain a fifth processing result; and summing the third processing result and the fourth processing result to obtain a sixth processing result.
Step S4280, training the initial first machine learning model, the initial second machine learning model, the initial third machine learning model and the initial fourth machine learning model according to the fifth processing result, the sixth processing result, the combined feature vector and the actual matching score of the training sample to obtain the trained first machine learning model, the trained second machine learning model, the trained third machine learning model and the trained fourth machine learning model.
In this embodiment, a target dimension conversion process may be performed on the combined feature vector of the training sample to obtain a combined feature vector of a target dimension, and then a point multiplication operation is performed on the fifth processing result, the sixth processing result, and the combined feature vector of the target dimension of the training sample to obtain a matching score expression of the training sample; constructing a loss function according to the matching score expression and the actual matching score of the training sample; and solving the loss function, and determining a first undetermined coefficient, a second undetermined coefficient, a third undetermined coefficient and a fourth undetermined coefficient to obtain a trained first machine learning model, a trained second machine learning model, a trained third machine learning model and a trained fourth machine learning model.
In an embodiment of the present disclosure, each recommendation reason document of the target store may have a corresponding feature vector, and the combined feature vector corresponding to each recommendation reason document combination may be obtained by concatenating the feature vectors corresponding to the recommendation reason documents included in the corresponding recommendation reason document combination in the order of arrangement of the recommendation reason documents in the recommendation reason document combination.
For example, if the recommended reason document combination includes document 1, document 2, and document 4, the feature vector corresponding to document 1 is X71, the feature vector corresponding to document 2 is X72, and the feature vector corresponding to document 4 is X74, then the combined feature vector corresponding to the recommended reason document combination can be expressed as { X71, X72, X74}.
And step S3300, determining a recommended reason and case combination matched with the target user and the target shop as a target case combination according to the target user feature vector, the target shop feature vector and the combined feature vector.
The target file combination selected by the embodiment can best meet the preference of the target user on the recommended files of the target shop, and can occupy the corresponding display position in the interface of the client to the maximum extent, so that the display effect of the target file combination is improved.
In an embodiment of the present disclosure, determining a recommended reason document combination matching the target user and the target store according to the target user feature vector, the target store feature vector, and the combined feature vector may include, as the target document combination, steps S3310 to S3330 as follows:
step S3310, perform target dimension conversion processing on the combined feature vector to obtain a combined feature vector of a target dimension. As shown in fig. 4.
In an embodiment of the present disclosure, the combined feature vector may be mapped to a space of a target dimension, so as to obtain a combined feature vector of the target dimension.
Step S3320, performing dot product operation on the target user characteristic vector, the target shop characteristic vector and the combined characteristic vector of the target dimension to obtain the prediction matching scores of the target user, the target shop and each recommended reason and literature combination. As shown in fig. 4.
Step S3330, selecting the recommended reason and literature combination with the maximum predicted matching score with the target user and the target shop as the target literature combination.
For example, if the predicted match score for the recommended reason portfolio shown in FIG. 1a may be 0.6, the predicted match score for the recommended reason portfolio shown in FIG. 1b may be 0.7, the predicted match score for the recommended reason portfolio shown in FIG. 1c may be 0.65, and then the recommended reason portfolio shown in FIG. 1b may be combined as the target portfolio.
By the method of the embodiment, the calculation amount of the selected target file combination can be reduced, and the calculation speed of the selected target file combination can be increased.
And step S3400, pushing the target file combination and the shop information of the target shop to the client so as to be displayed by the client.
In this embodiment, the recommendation server pushes the target pattern combination and the store information of the target store to the client, the client may be a store card showing the target store in the interface, and the store card of the target store may be shown with the store information of the target store and all recommendation reason patterns in the target pattern combination to provide decision information for the target user and highlight the store characteristics of the target store.
Through the embodiment of the disclosure, the plurality of recommendation reason documents recommended to the target user can occupy the corresponding display positions in the interface of the client to the maximum extent, and the display effect of the recommendation reason documents is improved. In addition, when the target store has a plurality of recommended reason documents, the target user may be individually ranked in order of the recommended reason documents, and may be individually recommended in combination of the recommended reason documents.
< Server >
In the present embodiment, there is also provided a server 5000, as shown in fig. 5, including a first memory 5100 and a first processor 5200.
The first memory 5100 for storing an executable first computer program; the first computer program is used to control the first processor 5200 to execute any one of the recommendation-reason-based store recommendation methods provided in the present embodiment.
< second embodiment >
< method >
The method for displaying the store information provided in the embodiment may be implemented by a client. In one example, the client may comprise client 2000 as shown in FIG. 2.
As shown in fig. 6, the method of displaying store information includes steps S6100 to S6300.
Step S6100, responding to the target operation executed by the target user through the client, sending a shop recommendation request aiming at the target user to the server, so that the server responds to the shop recommendation request, determines a recommendation reason and literature combination matched with the target user and the target shop as a target literature combination, and pushes the target literature combination and the shop information of the target shop to the client; wherein, each recommended reason and case combination comprises at least two recommended reason and cases with an arrangement sequence.
Wherein the target user may be a user who logs in the target application through the client.
The target operation in this embodiment may be an operation of the target user opening the target application program, or an operation of the target user opening the target page of the target application program.
The method includes that a server executes a shop recommendation method based on shop recommendation reasons, wherein the shop recommendation method is described in the first embodiment, a plurality of recommendation reason and literature combinations of a target shop and the target shop to be displayed are obtained in response to the shop recommendation request, a target user feature vector corresponding to a target user, a target shop feature vector corresponding to the target shop and a combination feature vector corresponding to each recommendation reason and literature combination are obtained, a recommendation reason and literature combination matched with the target user and the target shop is determined according to the target user feature vector, the target shop feature vector and the combination feature vector to serve as a target literature combination, and shop information of the target literature combination and the target shop is pushed to a client; wherein, each recommended reason and case combination comprises at least two recommended reason and cases with an arrangement sequence. Reference may be made to the first embodiment, which is not described herein again.
In step S6200, the target pattern combination and the store information of the target store are received.
In step S6300, the combination of the store information and the target pattern of the target store is displayed on the store card corresponding to the target store.
The client can be used for displaying the shop card of the target shop in the interface, and the shop card of the target shop can be displayed with the shop information of the target shop and all the recommended reason documents in the target document combination so as to provide decision information for the target user and highlight the shop characteristics of the target shop.
Through the embodiment of the disclosure, the plurality of recommendation reason documents recommended to the target user can occupy the corresponding display positions in the interface of the client to the maximum extent, and the display effect of the recommendation reason documents is improved. In addition, when the target store has a plurality of recommendation reason documents, the target user can be subjected to personalized ranking of the recommendation reason documents, and further personalized recommendation of the recommendation reason document combination is carried out on the target user, so that the store information of the target store in the store card and all the recommendation reason documents in the target document combination can be more prominent, and the decision-making information can be provided for the target user.
In one embodiment of the present disclosure, presenting the target paperwork combination may comprise: the recommendation reason documents included in the target document combination are displayed in a full frame in the recommendation reason display frame of the shop card.
The recommendation reason document included in the target document set can be displayed in the recommendation reason display frame of the shop card in a full frame, so that the recommendation reason document included in the target document set can be filled in the recommendation reason display frame. The recommendation reason display frame may be a display frame for displaying the recommendation reason document included in the target document combination, and the client may display the frame line of the recommendation reason display frame in an arbitrary color or may not display the frame line of the recommendation reason display frame.
Thus, the effect of displaying the recommended reason documents can be improved.
< client >
In this embodiment, there is also provided a client 7000, as shown in fig. 7, comprising a second memory 7100 and a second processor 7200.
The second memory 7100 for storing a second executable computer program; the second computer program is used to control the second processor 7200 to execute the store information display method provided in this embodiment.
< third embodiment >
< System >
In this embodiment, a store recommendation system 8000 based on a recommendation reason is further provided, as shown in fig. 8, including the server 5000 according to the first embodiment and the client 7000 according to the second embodiment.
Through the embodiment of the disclosure, the plurality of recommendation reason documents recommended to the target user can occupy the corresponding display positions in the interface of the client to the maximum extent, and the display effect of the recommendation reason documents is improved. In addition, when the target store has a plurality of recommended reason documents, the target user may be individually ranked in order of the recommended reason documents, and may be individually recommended in combination of the recommended reason documents.
The present disclosure may be systems, methods, and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for causing a processor to implement various aspects of the present disclosure.
The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be interpreted as a transitory signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or an electrical signal transmitted through an electrical wire.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry that can execute the computer-readable program instructions implements aspects of the present disclosure by utilizing the state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer-readable program instructions may also be stored in a computer-readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer-readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are equivalent.
Having described embodiments of the present disclosure, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen in order to best explain the principles of the embodiments, the practical application, or improvements made to the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the present disclosure is defined by the appended claims.
Claims (10)
1. A shop recommendation method based on a shop recommendation reason is characterized in that the method is used for a server, and comprises the following steps:
responding to a shop recommendation request aiming at a target user sent by a client, and acquiring a target shop to be displayed and a plurality of recommendation reason and literature combinations of the target shop; wherein, each recommendation reason and case combination comprises at least two recommendation reason and cases with a ranking order;
acquiring a target user characteristic vector corresponding to the target user, a target shop characteristic vector corresponding to the target shop and a combined characteristic vector corresponding to each recommended reason and file combination;
according to the target user feature vector, the target shop feature vector and the combined feature vector, determining a recommended reason and file combination matched with the target user and the target shop to serve as a target file combination;
and pushing the target file combination and the shop information of the target shop to a client so as to be displayed by the client.
2. The method of claim 1, wherein obtaining the plurality of recommended reason document combinations comprises:
acquiring a plurality of recommended reason documents of the target shop;
and obtaining a plurality of recommended reason file combinations according to the interface size of the client and the length of each recommended reason file.
3. The method of claim 2, wherein obtaining a plurality of recommended reason document combinations according to the interface size of the client and the length of each recommended reason document comprises:
carrying out permutation and combination on the plurality of recommended reason and patterns to obtain a plurality of permutation and combination results;
and determining at least two recommended reason documents to be displayed in the interface of the client in each permutation and combination result as the recommended reason document combination according to the interface size of the client and the length of each recommended reason document.
4. The method of claim 1, wherein the dimensions of the target user feature vector and the target store feature vector are both target dimensions;
determining a recommended reason and case combination matched with the target user and the target store according to the target user feature vector, the target store feature vector and the combined feature vector, wherein the determining is used as a target case combination and comprises the following steps:
performing target dimension conversion processing on the combined feature vector to obtain a combined feature vector of the target dimension;
performing point multiplication operation on the target user feature vector, the target shop feature vector and the combined feature vector of the target dimension to obtain a prediction matching score of the target user, the target shop and each recommended reason and literature combination;
and selecting a recommended reason and file combination with the maximum predicted matching score of the target user and the target shop as the target file combination.
5. The method according to claim 4, wherein before the obtaining of the target user feature vector corresponding to the target user and the target store feature vector corresponding to the target store, the method further comprises:
acquiring a first feature vector of the target user, wherein the first feature vector comprises a plurality of user features reflecting the preference of the target user on stores and recommendation reason documents;
performing first dimension conversion processing on the first eigenvector to obtain a second eigenvector of a target dimension; performing second dimension conversion processing on the first feature vector to obtain a third feature vector of a target dimension;
summing the second feature vector and the third feature vector to obtain the target user feature vector;
obtaining a fourth feature vector of the target store, wherein the fourth feature vector comprises a plurality of store features influencing the preference of the target user for the target store;
performing third dimension conversion processing on the fourth feature vector to obtain a fifth feature vector of a target dimension; performing fourth dimension conversion processing on the fourth feature vector to obtain a sixth feature vector of a target dimension;
summing the fifth feature vector and the sixth feature vector to obtain the feature vector of the target shop;
and storing the target user characteristic vector and the target shop characteristic vector.
6. A method for displaying store information is used for a client and comprises the following steps:
responding to a target operation executed by a target user through the client, sending a shop recommendation request aiming at the target user to a server, so that the server determines a recommendation reason and literature combination matched with the target user and the target shop as a target literature combination in response to the shop recommendation request, and pushing the target literature combination and shop information of the target shop to the client; wherein, each recommended reason and document combination comprises at least two recommended reason and documents with an arrangement sequence;
receiving the target file combination and the store information of the target store;
and displaying the combination of the shop information of the target shop and the target file in the shop card corresponding to the target shop.
7. The method of claim 6, wherein presenting the target portfolio comprises:
and displaying the recommended reason file contained in the target file combination in a full frame in a recommended reason display frame of the shop card.
8. A server comprising a first processor and a first memory for storing an executable first computer program; the first computer program is for controlling the first processor to perform the method of any one of claims 1 to 5.
9. A client comprising a second processor and a second memory, said second memory for storing a second computer program executable; the second computer program is for controlling the second processor to perform the method of claim 6 or 7.
10. A store recommendation system based on a recommendation reason, comprising the server according to claim 8 and the client according to claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268467.5A CN115905681A (en) | 2022-10-17 | 2022-10-17 | Shop recommendation method, server, display method, client and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211268467.5A CN115905681A (en) | 2022-10-17 | 2022-10-17 | Shop recommendation method, server, display method, client and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115905681A true CN115905681A (en) | 2023-04-04 |
Family
ID=86477373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211268467.5A Pending CN115905681A (en) | 2022-10-17 | 2022-10-17 | Shop recommendation method, server, display method, client and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115905681A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116579812A (en) * | 2023-07-12 | 2023-08-11 | 迎动科技发展(成都)有限公司 | Comprehensive ordering method, system, equipment and medium for automobile repair shops |
-
2022
- 2022-10-17 CN CN202211268467.5A patent/CN115905681A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116579812A (en) * | 2023-07-12 | 2023-08-11 | 迎动科技发展(成都)有限公司 | Comprehensive ordering method, system, equipment and medium for automobile repair shops |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10650311B2 (en) | Suggesting resources using context hashing | |
| CN109522483B (en) | Method and device for pushing information | |
| CN106874467B (en) | Method and apparatus for providing search results | |
| US10242093B2 (en) | Method and system for performing a probabilistic topic analysis of search queries for a customer support system | |
| US10748157B1 (en) | Method and system for determining levels of search sophistication for users of a customer self-help system to personalize a content search user experience provided to the users and to increase a likelihood of user satisfaction with the search experience | |
| KR102087807B1 (en) | Character inputting method and apparatus | |
| CN108280200B (en) | Method and device for pushing information | |
| US11151618B2 (en) | Retrieving reviews based on user profile information | |
| US10599760B2 (en) | Intelligent form creation | |
| US20220121668A1 (en) | Method for recommending document, electronic device and storage medium | |
| CN110059172B (en) | Method and device for recommending answers based on natural language understanding | |
| CN110232920B (en) | Voice processing method and device | |
| JP6648283B2 (en) | Note providing method and apparatus using artificial intelligence based relevance calculation | |
| CN105824436A (en) | Character input method and input method system | |
| US20170186102A1 (en) | Network-based publications using feature engineering | |
| US20240187358A1 (en) | Intelligent Assistant Content Generation | |
| US20210049665A1 (en) | Deep cognitive constrained filtering for product recommendation | |
| CN115905681A (en) | Shop recommendation method, server, display method, client and system | |
| CN112116411B (en) | Training method, device and system for commodity recommendation ordering model | |
| US20170193452A1 (en) | Job referral system | |
| CN109685091B (en) | Digital experience targeting using bayesian approach | |
| US20180039944A1 (en) | Job referral system | |
| CN114817716A (en) | Method, device, equipment and medium for predicting user conversion behaviors and training model | |
| US20210166282A1 (en) | Personalized Dynamic Sub-Topic Category Rating from Review Data | |
| CN115994807A (en) | Material recommendation method, device and system |
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 |