CN113298587A

CN113298587A - Shop commodity information display method, electronic equipment and client

Info

Publication number: CN113298587A
Application number: CN202010451436.8A
Authority: CN
Inventors: 赵立冬; 黄梦华
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2020-05-25
Filing date: 2020-05-25
Publication date: 2021-08-24

Abstract

The embodiment of the specification provides a shop commodity information display method, electronic equipment and a client, the method determines a display strategy matched with behavior characteristics according to the behavior characteristics corresponding to the identity of a user, and displays commodities in a live shop, particularly commodity videos according to the corresponding display strategy. The method solves the technical problem that the commodity presentation mode on the shelf of the live broadcast shop is single, can present commodities according to different presentation strategies for users with different categories, has more diversified shop presentation modes, simultaneously supports second-level uploading of commodity videos, meets the personalized requirements of the users, and simultaneously improves the convenience of commodity release.

Description

Shop commodity information display method, electronic equipment and client

Technical Field

One or more embodiments of the present specification relate to the field of big data technologies, and in particular, to a method for displaying store commodity information, an electronic device, and a client.

Background

Live broadcasting has become a popular form of tape goods and is receiving increasing attention, which also brings more access flow to the opening of live stores. For example, a user may want to view other merchandise in the store while watching a live broadcast, and may enter the store to browse merchandise by clicking buttons such as "enter store and go shopping", "visit store", and the like.

However, the existing goods presentation mode on the shelves of some live stores is single, and the user access experience needs to be improved.

Disclosure of Invention

One or more embodiments of the specification describe a display method, electronic equipment and a client for store commodity information, which solve the technical problem of single commodity presentation mode on a shelf of a live broadcast store, can present commodities according to different display strategies for different users, and meet diversified requirements of the users.

According to a first aspect, the present specification provides a method of displaying store merchandise information, the method comprising:

acquiring an access request which is initiated by a user at a client and aims at a target shop, wherein the access request carries an identity of the user; determining behavior characteristics corresponding to the user based on the identity of the user; and determining a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

According to a second aspect, the present specification provides a method of displaying store merchandise information, the method comprising:

acquiring an access request which is initiated by a user when the user watches live broadcasting at a client and aims at a target live broadcasting shop, wherein the access request carries an identity of the user; determining behavior characteristics corresponding to the user based on the identity of the user; and determining a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target live broadcast shop based on the display strategy.

In one embodiment, determining a corresponding presentation policy based on the behavior feature includes:

determining the user class cluster to which the user belongs according to the similarity between the behavior characteristics and class cluster centers of a plurality of pre-formed user class clusters; the user clusters are obtained by performing cluster analysis on the behavior characteristics of a plurality of users, and one cluster center is a central point selected from one user cluster; and determining a display strategy corresponding to the user class cluster according to a preset mapping relation between the user class cluster and the display strategy.

In one embodiment, the plurality of user class clusters are obtained by:

acquiring historical behavior data corresponding to a plurality of users in a preset time period, acquiring behavior characteristics corresponding to each user respectively based on the historical behavior data, and performing cluster analysis by taking the behavior characteristics as cluster objects; and determining a plurality of user clusters and a cluster center corresponding to each user cluster according to the clustering result.

In one embodiment, determining the behavior feature corresponding to the user based on the identity of the user includes:

and acquiring historical behavior data of the user in a preset time period according to the identity of the user, and converting the historical behavior data into formatted data in a specified format to be used as behavior characteristics corresponding to the user.

In one embodiment, obtaining historical behavior data of the user over a predetermined period of time includes:

acquiring historical behavior data formed by the user at a first client and/or a second client; the first client and the second client run on the basis of different service platforms and both support live store broadcasting.

inputting the behavior characteristics into a pre-trained neural network model, and carrying out classification and identification on the behavior characteristics through the neural network model to obtain a display strategy corresponding to the user.

In one embodiment, the neural network model is trained by:

acquiring a training sample set, wherein each training sample comprises behavior characteristics of a sample user and an identifier of an applicable display strategy; and training the multi-classified neural network model by taking each display strategy as a classification class based on the training sample set.

In one embodiment, the method further comprises:

and after determining that a release instruction submitted by a user is received, releasing the commodity video selected by the user to a commodity shelf of the live broadcast shop so as to upgrade image-text detail pages of commodities on the commodity shelf to be presented as video detail pages.

In one embodiment, before determining that the issue instruction submitted by the user is received, the method further includes:

and after determining that a release request initiated by a user at a client is received, indicating the client to at least display a plurality of pre-made alternative commodity videos to the user.

In one embodiment, before instructing the client to display at least a plurality of pre-made video alternatives to the user, the method further includes:

and receiving the commodity video uploaded by the commodity video producer to serve as the alternative commodity video.

In a more specific embodiment, the video of the candidate commodity is provided with title information.

In a more specific embodiment, instructing the client to display at least a plurality of pre-made video of the candidate commodity to the user comprises: and instructing the client side to display a plurality of alternative commodity videos and the one-key publishing button to the user.

In a more specific embodiment, after determining that a publishing request initiated by a user based on a client is received, before instructing the client to display at least a plurality of pre-made video alternatives to the user, the method further includes:

determining store attributes of the live stores; and selecting a commodity video of the hot commodity matched with the store attribute from the commodity video big data as the alternative commodity video according to the store attribute, wherein the hot commodity is a commodity of which the transaction data in a preset time period reaches a preset standard.

In a more specific embodiment, publishing the user-selected merchandise video to a merchandise shelf of the live store includes:

classifying and identifying at least one commodity video selected by the user based on a commodity classification mode currently adopted by the live broadcast shop; and according to the identification result, the at least one commodity video is distributed to the commodity shelf of the corresponding category in a classified mode.

In a more specific embodiment, determining a corresponding presentation policy based on the behavior feature includes:

and determining a corresponding display strategy by combining each commodity video on the current commodity shelf of the live broadcast shop based on the behavior characteristics.

In a more specific embodiment, the display strategy at least comprises any one or any combination of the order of commodity presentation, page layout, page style, video tone, video background image and video background music.

According to a third aspect, the present specification also provides a store merchandise information distribution method including:

after a release instruction submitted by a user is determined to be received, the commodity video selected by the user is released to a commodity shelf of a live broadcast shop corresponding to the user, so that a client faces the user who accesses the shop, and the image-text detail pages of the commodities on the commodity shelf are upgraded to be displayed as video detail pages.

According to a fourth aspect, the present specification further provides a method for displaying store merchandise information, the method comprising:

acquiring an access request which is initiated by a user when the user watches live broadcasting at a first client and aims at a target live broadcasting shop, wherein the access request carries an identity of the user; acquiring historical behavior data formed by the user at least one client according to the identity, wherein the at least one client comprises a second client supporting store live broadcast, and the second client and the first client run on the basis of different service platforms and both support store live broadcast; determining behavior characteristics corresponding to the user according to the historical behavior data of the user; and determining a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

According to a fifth aspect, the present specification further provides a method for displaying store merchandise information, the method comprising:

sending an access request which is initiated by a user and aims at a target live broadcast shop to a server, wherein the access request carries an identity of the user, so that the server determines behavior characteristics corresponding to the user; receiving a display strategy issued by the server, wherein the display strategy corresponds to the behavior characteristics of the user; and presenting the commodities in the target live broadcast shop based on the display strategy.

According to a sixth aspect, the present specification further provides a store merchandise information distribution method including:

uploading a release instruction submitted by a user to a service end so that the service end releases the commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user, and the service end is used for upgrading image-text detail pages of commodities on the commodity shelf into video detail pages to be presented.

In one embodiment, before uploading the issue instruction submitted by the user to the server, the method further includes:

and uploading the release request submitted by the user to a server, and at least displaying a plurality of pre-made alternative commodity videos for the user according to the indication of the server.

In one embodiment, after uploading the publishing request submitted by the user to the server, before displaying at least a plurality of pre-made candidate commodity videos to the user, the method further includes:

receiving hot commodity information issued by a server, and taking a commodity video corresponding to the hot commodity information as the alternative commodity video; the hot commodities are commodities which are selected by the server from the video big data of the commodities, matched with the store attributes of the live stores and have transaction data meeting a preset standard within a preset time period.

In a more specific embodiment, at least a plurality of pre-made video of the candidate commodities are presented to the user, and the method comprises the following steps:

and displaying a plurality of candidate commodity videos and the one-key publishing button to the user.

According to a seventh aspect, the present specification also provides an electronic device comprising:

the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is configured to acquire an access request which is initiated by a user at a client and aims at a target shop, and the access request carries an identity of the user; a first determining unit configured to determine a behavior feature corresponding to the user based on the identity of the user; the first display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

According to an eighth aspect, the present specification further provides an electronic device, wherein the electronic device comprises:

the second acquisition unit is configured to acquire an access request which is initiated by a user when the user watches live broadcasting at a client and aims at a target live broadcasting shop, wherein the access request carries the identity of the user; a second determining unit configured to determine a behavior feature corresponding to the user based on the identity of the user; and the second display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target live store based on the display strategy.

According to a ninth aspect, the present specification further provides an electronic device comprising:

the first publishing unit is configured to publish the commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user after determining that a publishing instruction submitted by the user is received, so that a client faces the user accessing the shop, and the image-text detail pages of the commodities on the commodity shelf are upgraded to be displayed as video detail pages.

According to a tenth aspect, the present specification also provides an electronic device comprising:

the third acquisition unit is configured to acquire an access request which is initiated by a user when the first client watches live broadcasting and aims at a target live broadcasting shop, wherein the access request carries the identity of the user; the acquisition unit is configured to acquire historical behavior data formed by the user at least one client according to the identity, the at least one client comprises a second client supporting store live broadcast, and the second client and the first client run on the basis of different service platforms and both support store live broadcast; a third determining unit, configured to determine a behavior feature corresponding to the user according to historical behavior data of the user; and the third display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

According to an eleventh aspect, the present specification further provides a client comprising:

the system comprises a first sending unit, a second sending unit and a third sending unit, wherein the first sending unit is configured to send an access request which is initiated by a user and aims at a target live broadcast shop to a server side, and the access request carries an identity of the user so that the server side can determine a behavior characteristic corresponding to the user; the first receiving unit is configured to receive a display strategy issued by the server, wherein the display strategy corresponds to the behavior characteristics of the user; and the fourth display unit is configured to display the commodities in the target live store based on the display strategy.

According to a twelfth aspect, the present specification further provides a client comprising:

the second issuing unit is configured to upload an issuing instruction submitted by a user to a server, so that the server issues a commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user, and the commodity video is used for upgrading an image-text detail page of a commodity on the commodity shelf to be a video detail page for presentation.

According to a thirteenth aspect, there is also provided a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method of any of the first to sixth aspects described above.

According to a fourteenth aspect, there is also provided a computing device, comprising a memory and a processor, wherein the memory stores executable code, and the processor executes the executable code to implement the method of any one of the first to sixth aspects.

To sum up, in the method disclosed in the embodiment of the present specification, a behavior feature corresponding to a user is determined according to an identity carried in an access request initiated by the user, then a display policy matched with the behavior feature is determined according to the behavior feature, and commodities in a target store are presented according to the display policy, so that the display policy is determined based on the behavior feature of the user, when the same store is visited by users with different behavior features, different commodity display modes can be presented, personalized requirements of the user can be better met, and user access experience is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic flow chart diagram illustrating one embodiment of a method for displaying store merchandise information provided herein;

FIG. 2 is a diagram showing an example of a system architecture in an application scenario of the method for displaying store commodity information according to the present specification;

FIG. 3 is a schematic flow chart diagram illustrating one embodiment of a method for displaying store merchandise information suitable for a live store provided by the present specification;

FIG. 4 is a schematic flow chart diagram illustrating one embodiment of a method for displaying store merchandise information provided herein;

FIG. 5 is a data table illustrating a mapping relationship between a user class cluster and a presentation policy in one embodiment of the present specification;

FIG. 6 is a schematic flow chart diagram illustrating one embodiment of a store merchandise information distribution method provided herein;

FIG. 7 is a schematic flow chart diagram illustrating a description from a client side perspective of one embodiment of a store merchandise information distribution method provided by the present specification;

FIG. 8 is a diagram illustrating a data interaction flow between a server and a first client in an embodiment of a method for displaying store merchandise information provided by the present specification;

FIG. 9 is a flowchart showing one embodiment of a store merchandise information distribution method provided by the present specification;

FIG. 10 shows a data interaction flow between the server and the second client in the process of executing the second-level uploading of the commodity video;

FIG. 11 is a system architecture diagram illustrating an example application scenario in which merchandise videos are published to shelves one time;

FIG. 12 is a flow chart illustrating one embodiment of a method for displaying store merchandise information provided by the present specification based on another application scenario;

FIG. 13 is a system architecture diagram illustrating an example of an application scenario of the method for displaying information about store merchandise provided by the present specification;

FIG. 14 is a schematic structural diagram of an electronic device for implementing a method for displaying information on store merchandise according to an embodiment of the present disclosure;

FIG. 15 is a schematic structural diagram illustrating an electronic device for applying a method for displaying store merchandise information to a live store according to an embodiment of the present disclosure;

FIG. 16 is a schematic structural diagram of a second client for implementing a store merchandise information distribution method according to an embodiment of the present specification;

FIG. 17 is a schematic structural diagram illustrating an electronic device for determining an own client exhibition policy according to behavior characteristics collected by a third client in one embodiment of the present specification;

fig. 18 is a schematic structural diagram illustrating a first client that uses behavior characteristics collected by a third client to determine an own client presentation policy in an embodiment of the present specification;

fig. 19 is a schematic structural diagram showing a client for implementing a store merchandise information distribution method according to an embodiment of the present specification.

Detailed Description

The scheme provided by the specification is described below with reference to the accompanying drawings.

One of the more typical application scenarios of the method for displaying store commodity information provided by the embodiment of the present specification is that a user initiates an access request for a store which opens live broadcasting, for example, the user initiates an access request for entering the store to access while watching the live broadcasting. The current live broadcast interface is generally provided with store icons to support store-in access, but the commodity display modes on some store shelves are fixed and single, the pertinence and the interestingness are lacked, and the interest degree of users accessing the stores to commodities is easily reduced.

In view of this, the present specification discloses a method for displaying store commodity information, which sets different store display strategies for different users, and further, the method can support second-level distribution of commodity videos, improve convenience of second users (store managers) in commodity distribution, distribute commodity details in a video form, and set more types of display strategies based on the commodity videos, so that commodities on store shelves can be presented more variously, and further, more different "faces" of stores can be displayed.

According to a first aspect of the present specification, embodiments of the present specification provide a method for displaying store merchandise information, which is described from the perspective of a server with reference to fig. 1, and may include the following processes:

s102, acquiring an access request which is initiated by a user at a client and aims at a target shop, wherein the access request carries an identity of the user; s104, determining behavior characteristics corresponding to the user based on the identity of the user; s106, determining a corresponding display strategy based on the behavior characteristics, and presenting the commodities in the target shop based on the display strategy.

It should be noted that the method may be applied to a live-broadcast network store, but does not exclude that the method is applied to other applicable objects, for example, for a store that is not opened live-broadcast, different display strategies may be set for different users by using the display method of store commodity information provided in the embodiments of the present specification, and a person skilled in the art may adaptively select an application scenario according to the scheme disclosed in the present specification, which is not listed in the present specification.

The following describes in detail a method for displaying store product information provided in the embodiments of the present specification, taking a typical application scenario of a live store as an example. Referring to fig. 2, fig. 2 shows an example of a system architecture in the application scenario, which includes a server 20 and a first client 10 a. When a first user watches the live broadcast of a certain live broadcast shop, a floating window shown as a first client 10a in fig. 2 is opened through a shop icon on a live broadcast interface, a link button for "shopping a shop" is arranged in the floating window, the first user clicks the "shopping a shop", the first client detects the clicking operation of the first user, and a corresponding access request is generated. After receiving the access request, the server 20 determines the behavior characteristics corresponding to the first user, determines the display policy in the corresponding category according to the category to which the behavior characteristics belong, and issues the display policy to the first client 10a, wherein the first client displays a plurality of commodities to the first user according to the corresponding display policy, and different display policies are correspondingly obtained when two first users with obviously different behavior characteristics visit the same live broadcast shop; and when two first users with closer behavior characteristics (for example, when the two first users belong to the same user class cluster based on cluster analysis) visit the same live store, the goods on the shelf can be displayed according to the same display strategy. In some embodiments, the item details on the shelves at the live store may be uploaded in a video format in advance.

For example, based on the analysis of the behavior characteristics, the current first user is interested in the skin care products sold on the network recently, so that the first user is recommended to be a video of the products of the currently sold skin care products, and a corresponding interactive interface is shown as 10 a' in fig. 2. And when another first user visits the same live shop, based on the behavior characteristic analysis of the another first user, the another first user is determined to be more concerned about the functional skincare products applicable to the sensitive muscles, and the another first user is shown according to different showing strategies, wherein a corresponding interactive interface is shown as 10 a' in fig. 2.

In the embodiment of the specification, the first user is an accessing party of the shop, namely a party sending an accessing request; the accessed stores in the currently received access request are live stores. The access request generally carries an identity of the first user and a live store identity, and the live store can be determined according to the live store identity. The first client is a client facing the first user and used by the first user, and the first client may be various clients supporting store-in access, such as a buyer client of panning.

The access request in the embodiment of the present specification includes an access request to a store and an access request to a certain product under the store. For example, the first user may enter the store by clicking a store icon on the live view screen, or may enter the store by clicking a push link of a certain product of the store, and the user may perform a click operation on the store or the product thereof as an access request.

Specifically, according to a second aspect, the present specification embodiment also provides a method for displaying store commodity information suitable for a live store. When described from the server perspective, the corresponding method flowchart is shown in fig. 3, and may include the following procedures:

s302, acquiring an access request which is initiated by a user when the user watches live broadcasting at a client and aims at a target live broadcasting shop, wherein the access request carries an identity of the user; s304, determining behavior characteristics corresponding to the user based on the identity of the user; s306, determining a corresponding display strategy based on the behavior characteristics, and presenting the behavior characteristics of the commodities in the target live broadcast shop based on the display strategy.

In some embodiments, the access request generally carries store identification information of the target live broadcast store. The access request is triggered based on the live broadcast interface, for example, the first user may open a corresponding floating window through a store icon on the live broadcast interface, and link buttons such as "shop shopping", "enter store shopping", and the like are displayed, and the first user clicks the link button to submit or initiate the access request.

In S302, the first client uploads an access request sent by the first user to the server, and after receiving the access request, the server reads the identity of the first user carried by the server, and as an implementable manner, also reads the identity of the live store.

Next, in S304, a behavior feature corresponding to the user is determined based on the identity of the user.

In this embodiment of the present specification, the behavior feature corresponding to the first user is obtained based on historical behavior data of the first user. The historical behavior data may be obtained from various APPs that support store live broadcasting, such as from an elutriation client, and particularly a buyer client. The historical behavior data may include various operation data formed when the first user uses each client, such as a search record, the number of times of watching a live broadcast, a time length, a live broadcast content, a visited shop type, visited commodity information, and the like, for example, the first user watches a full introduction about lipstick on the live broadcast room of the first client, but when the first user leaves the live broadcast room, the commodity is emptied, that is, the first user may be recorded as a kind of historical behavior data. For another example, the first user may pay attention to a live broadcast shop on the third client, and may be recorded as historical behavior data (on the premise that the first user is authorized or permitted) as a fan of the live broadcast shop.

In one embodiment, the behavior feature may be extracted based on the historical behavior data of the first user as follows: acquiring historical behavior data of the first user identity within a predetermined time period, wherein the predetermined time period can be set to 1-15 days, for example, acquiring the historical behavior data within 7 days before the current time, and then converting the acquired historical behavior data within 7 days into formatted data of a specified format, for example, converting the formatted data into a feature vector or a feature matrix, a two-dimensional array and the like of a specified dimension, that is, converting the historical behavior data into structured data with a uniform format, as a first behavior feature corresponding to the first user identity.

In some embodiments, the redundancy of the converted formatted data is still high, and in order to extract more refined behavior features to reduce the computation amount of subsequent data processing, the feature extraction may be performed on the formatted data by using characterization learning to obtain the behavior features.

Next, in S306, after the behavior feature of the first user is obtained, a display policy matching the behavior feature of the first user is determined.

The display strategy is a strategy for indicating how the live broadcast shop displays the goods on the shelf facing the first user. Due to the diversification of user requirements, the display strategy is also very diversified. In general, the category of the presentation policy may correspond to the category of the first user.

The user categories can be divided by adopting a cluster analysis algorithm or a neural network model and the like. For example, in some embodiments, the category of the first user may perform clustering analysis with the behavior feature of the first user as a clustering object, so as to obtain a clustering result that classifies a plurality of first users into different user clusters. The method comprises the steps of classifying a first user in advance, judging the category of the current first user based on the pre-classified category when receiving an access request of the first user, comparing the behavior characteristics (first behavior characteristics) of the current first user with the behavior characteristics (second behavior characteristics) of each classified category when judging, wherein the time overhead consumed by comparison operation can be borne, and compared with the step of re-classifying the category to formulate a corresponding display strategy, the method can respond to the access request more quickly. Next, to prevent confusion, the behavior feature corresponding to the currently accessing user determined in S304 is defined as a first behavior feature, and the cluster object of the cluster analysis is defined as a second behavior feature.

Specifically, in some embodiments, a cluster analysis algorithm is used to determine the category to which the first user belongs, and further determine the corresponding display policy:

referring to fig. 4, before performing S302 and S304, the method further includes S300: the method comprises the steps of obtaining historical behavior data corresponding to a plurality of sample users in a preset time period in advance, obtaining second behavior characteristics corresponding to each sample user based on the historical behavior data, performing clustering analysis by taking the second behavior characteristics as clustering objects, and determining a plurality of user clusters and a cluster center corresponding to each user cluster according to clustering results.

The second behavior feature and the first behavior feature have the same structured data format, and may be data which is extracted based on the historical behavior data of the first user and can be stored in a structured manner, except that the acquisition time of the historical behavior data for obtaining the second behavior feature is earlier than the acquisition time of the historical behavior data for obtaining the first behavior feature. The first behavior feature is typically derived based on the first user's behavior data over a recent period of time (e.g., within a week). For example, a first user may have the second behavior feature from historical behavior data in the past month, and the first behavior feature from historical behavior data 7-15 days before the first user visits the live store.

When the cluster analysis is executed, the similarity between the second behavior characteristics can represent the similarity between the past behavior habits or the recent behaviors of different sample users, and the cluster analysis which takes the second behavior characteristics as a clustering object is executed, which is equivalent to clustering the sample users with similar behaviors into a cluster. After the clustering result is obtained, the cluster determined by the clustering result can be directly used as a user cluster, and the cluster center of each user cluster can be regarded as a second behavior characteristic which can most represent the common characteristics of the user behaviors of all samples in the cluster.

In one embodiment, performing the clustering analysis may employ a k-means clustering algorithm (k-means clustering algorithm). For example, a K-means clustering algorithm is adopted to divide the second behavior features into K groups, then K second behavior features are randomly selected as initial clustering centers (i.e., cluster-like centers), then the distance between each object and each initial clustering center is calculated, and each second behavior feature is assigned to the closest clustering center, i.e., another closest second behavior feature. Each time a second behavior feature is assigned, the cluster center is recalculated based on the existing objects in the cluster. This process will be repeated until the termination condition is met. The termination condition may be that no (or a minimum number) of second behavior features are reassigned to different clusters, the cluster center is no longer changed, and the squared error sum is minimal, then the termination condition is considered to be reached.

The number of the first users can reach the hundred million level, the behavior data of the first users often have personalized differences, no obvious rules can be followed, and learning targets with reference values are difficult to determine, so that supervised machine learning is not suitable for class division of the first users, a k-means clustering algorithm is easy to implement, labels do not need to be set, the clustering process is relatively simple, and the execution efficiency is higher.

In one embodiment, the operation of obtaining the second behavior feature of each sample user in S300 is performed prior to the operation of obtaining the first behavior feature in S304 in the execution timing. The manner of obtaining the second behavior feature based on the historical behavior data in S300 may refer to obtaining the description of the first behavior feature in S304. The second behavior feature may also be obtained in other ways, for example, in another embodiment, the second behavior feature is extracted from historical behavior data through characterization learning.

As above, in general, the second behavior feature obtained in S300 and the behavior feature obtained in S304 should keep the same format, for example, as a feature vector, or as a two-dimensional array.

Since the classification of the historical behavior data of the first user belongs to big data processing analysis in most embodiments, it takes a certain time, and it is obviously infeasible to perform classification of the plurality of first users after receiving the access request of the first user, so that the plurality of first users should be classified at intervals of a preset time based on the behavior characteristics of the first user. That is, the category to which a first user belongs may be constantly changing, or the first user currently visiting does not belong to a user who has already performed cluster analysis, but when visiting the store, based on the division result of the user category of the latest time, the similarity between the first behavior feature of the current first user in the latest period of time and the second behavior feature of the center of each cluster may be calculated, and the cluster of the user to which the second behavior feature with the highest similarity to the first behavior feature is located may be used as the cluster to which the first user currently visiting belongs.

Next, the method further includes S301: and respectively generating different display strategies aiming at each user class cluster, enabling one user class cluster to correspond to at least one display strategy, and establishing a mapping relation between each user class cluster and each display strategy.

For example, in one embodiment, 50 user class clusters are determined according to the above cluster analysis, and then 50 display strategies may be generated correspondingly. Under the one-to-one mapping relation, when the number of the user class cluster is determined, the corresponding display strategy number can be clarified.

In another embodiment, one user class cluster may correspond to more than two presentation policies, and one presentation policy may also correspond to more than two user class clusters, which is not limited to one-to-one mapping, i.e. the number of user class clusters and the number of presentation policies may not be the same. In this case, the mapping relationship from the user class cluster to the presentation policy needs to be made explicit. For example, referring to FIG. 5, the determined 50 user class cluster numbers are C₁、C₂、C₃……C₅₀The generated n display strategy numbers are S respectively₁、S₂、S₃……S_nThe mapping relationship may be recorded by using a 50-column multi-row data table, the first row is written with 50 user class cluster numbers, and the display policy numbers corresponding to each user class cluster are sequentially written into other rows in the same column. For example wherein C₅₀Corresponds to S₅₀、S₄₈And S₄₅Three display strategies, when the class cluster to which the first user belongs is determined to be C₅₀Then can be from S₅₀、S₄₈And S₄₅One of them is randomly selected as the presentation strategy.

Next, the similarity between the first behavior feature and the center of each cluster class is calculated, respectively. And the cluster center is a second behavior feature which is selected as a cluster center point, the similarity between the first behavior feature and each cluster center is calculated, which is equivalent to calculating the similarity between the first behavior feature and a plurality of second behavior features in sequence, and the cluster with the cluster center with the highest similarity to the first behavior feature is taken as the user cluster to which the current first user belongs. The similarity calculation may be performed in various ways, and when the first behavior feature and the second behavior feature are both feature vectors, a distance between the two behavior feature vectors may be calculated as the similarity, for example, an euclidean distance.

Then, according to the mapping relation, a display strategy matched with the first behavior feature can be determined.

For example, through calculation, the first behavior feature is clustered with the user class C₂₅If the similarity of the cluster centers is the highest, the class to which the current first user belongs is determined to be C₂₅Referring to FIG. 5, C is read from the data table recording the mapping relationship₂₅The other rows in the column are numbered S₂₅And the number is the determined display strategy number.

The above description explains a way of determining the display policy based on the cluster analysis algorithm, and in another embodiment, the display policy may also be determined based on the neural network model, specifically as follows:

before S302 and S304, a training sample set is obtained, where each training sample in the training sample set includes behavior characteristics of a sample user and an identifier of an applicable presentation policy, and then based on the training sample set, a multi-class neural network model is trained with each presentation policy as a class.

The behavior characteristics of the sample user can be acquired by referring to the first behavior characteristics or the second behavior characteristics.

Then, executing S302 and S304 to obtain a first behavior feature, then inputting the first behavior feature into a pre-trained neural network model, and performing classification and identification on the first behavior feature through the neural network model to obtain a display strategy corresponding to the first user.

In this embodiment, the first behavior feature and the behavior feature of the sample user should adopt the same data format, for example, when the behavior feature of the sample user is a feature vector, the first behavior feature should also be a feature vector, and when the behavior feature of the sample user is a two-dimensional image, the first behavior feature should also adopt a two-dimensional image format.

Next, after determining the display policy based on the first behavior feature in the above manner, the server issues the determined display policy to the first client.

Wherein, in the show strategy, the live broadcast shop identification is also carried. And after receiving the display strategy, the first client indicates the first client to display the commodities in the live broadcast shop according to the display strategy according to the live broadcast shop identification.

In the embodiment of the present specification, the display policy includes a typesetting policy for displaying the product and a video playing policy for the product. The commodity video playing strategy comprises various video style designs, such as background music, background images of videos, color tones of the whole video picture and the like; the typesetting strategy comprises the sequence and layout of commodity video display and is compatible with the sequencing and layout of the picture and text commodities which are not off shelf.

According to a third aspect, embodiments of the present specification further provide a store merchandise information distribution method, as shown in fig. 6, the method including:

s601, determining that the issuing instruction submitted by the user is received.

S602, the commodity video selected by the user is released to a commodity shelf of a live broadcast shop corresponding to the user, so that the image-text detail page of the commodity on the commodity shelf is upgraded to be presented as a video detail page.

The user in S601 and S602 may be a second user, and the second user is a person having the authority to distribute the store product, such as a store owner or an authorized store manager.

The method of the first to third aspects may be performed on a server basis. With the continuous improvement of the computing power of the client, the method provided by the embodiments of the above aspects can also be executed directly on the client, that is, the execution subject of the method can be a server or a client. It should be noted that, when the client executes the method, the big data storage, for example, the steps of obtaining and storing the historical behavior data, etc. are still executed by the server, and those skilled in the art may adaptively determine the execution subject according to the above method provided by the embodiment of the present disclosure, which is not listed in this specification.

According to a fourth aspect, embodiments of the present specification further provide a method for displaying store merchandise information, which is described from the perspective of the first client, and shown in fig. 7, the method may include the following processes:

s701, sending an access request which is initiated by a user and aims at a target live broadcast shop to a server, wherein the access request carries an identity of the user; s702, receiving a display strategy issued by the server, wherein the display strategy corresponds to the behavior characteristics of the user; s703, displaying the commodities in the target live broadcast shop based on the display strategy.

The first client is a client which is provided for the first user to use and supports shop live broadcast.

Referring to fig. 8, fig. 8 shows a data interaction flow between a server and a first client in an embodiment of the method for displaying store merchandise information provided by the present specification:

the first user U1 clicks the store visit button;

s801, a first client generates an access request based on clicking of a user;

s802, the first client sends the access request to a server;

s803, the server receives the access request and determines the identity of the first user;

s804, the server acquires the corresponding behavior characteristics of the first user based on the identity of the first user;

s805, the server determines a corresponding display strategy based on the behavior characteristics;

s806, the server sends a display strategy to the first client;

and S807, the first client displays each commodity on the commodity shelf in a video form according to the received display strategy.

One feature of live tape is real-time, and the user may sometimes miss the item of interest for some reason and then view it as another item, at which time the user may wish to obtain further information about the item of interest by visiting the store. In the existing live-broadcast network shop, most of the commodities on the shelf introduce commodity details in the form of pictures and texts, and display strategies which can be designed aiming at the pictures and texts are limited, so that the layout or the sequencing of the pictures and texts of the commodities can be changed only and the commodities are not vivid enough.

In addition, the threshold of releasing commodities in the current shops is high, commodities on a shelf of the shop are limited to be in an image-text form, releasing steps are relatively complex, the styles of the commodities on the shelf of the shop are limited, hot commodities in the live broadcasting process are not introduced into the shop in a corresponding detail mode, and flow matching efficiency brought by live broadcasting is greatly improved.

Therefore, in order to enrich the presentation modes of stores, enable the stores to have more display strategies and improve the commodity distribution efficiency, the display method of the store commodity information disclosed by the specification further supports the uploading of commodity details in a video form.

In some embodiments, the image-text detail pages of the commodities in the live stores are upgraded into video detail pages, the links for releasing the commodities are changed, the commodities can be released by one key, and the video second-level uploading of the commodities is realized.

According to a fifth aspect of the present specification, embodiments of the present specification further provide a store merchandise information distribution method, which is described from the perspective of the second client, and the method may include the following processes:

Referring to fig. 9, fig. 9 is a system architecture diagram illustrating an example of an application scenario in which a commodity video is published to a shelf by one key. The system architecture includes at least a server 20 and a second client 10 b. The second client 10b sends a publishing instruction submitted by the second user to the server, and the server 20 publishes the commodity video selected by the second user to the commodity shelf of the live broadcast shop to which the second user belongs by one key.

In this embodiment of the present disclosure, the second client is a client for a store owner or a manager to log in the store to manage, edit, or publish the product, and is a client facing the second user, that is, a client provided for a seller to use, such as an APP (APP i cat i on, APP) of a Taobao seller, an APP of a thousand Newton seller, and so on. The second client may be a separate APP, or may be a functional module on the APP, such as a seller center of the treasure. The second user is a seller user or a merchant user, and generally includes a store owner or a manager.

In this specification, the video producer includes various video providers that upload videos of the product, for example, the video producer may be a supplier or a third-party video producing organization that cooperates with the supplier. In some embodiments, the commodity video recommended to the second user is a template video that has been produced by the supplier and is agreed by the supplier to be shared by the live stores.

In the embodiment of the present specification, the store attribute may be used to describe the type of the main commercial product of the store, for example, if the commercial product on the current shelf of a live store is mainly a clothing set (the percentage of all commercial products exceeds a predetermined threshold), the store attribute is clothing; if the current goods on the shelf of the other live broadcast shop are mainly tea leaves, the properties of the shop are tea or medium drinks and the like.

In the embodiment of the specification, in the process of selecting and publishing the commodity by the second user, the commodity video displayed for the second user generally has a title, so that the second user can quickly distinguish the main content of the video and the main selling point information of the commodity. The title may be filled in by the video producer at the time the video is produced. Or can be edited by the second user to embody the unique selling point information which the second user needs to show.

Specifically, in some embodiments, after determining that the publishing request submitted by the second user based on the second client is received, the server instructs the second client to display at least a plurality of pre-made candidate commodity videos to the second user, for example, instructs the second client to display the plurality of candidate commodity videos and the one-key publishing button to the second user. And then, judging whether the user sends an issuing instruction or not according to the next operation executed by the second user through the second client.

And after determining that a release instruction submitted by the second user is received, releasing the commodity video selected by the second user to a commodity shelf of the live broadcast shop by one key.

In some embodiments, the release instruction carries the video information of the commodity selected by the second user, and the release instruction is submitted by the second user clicking a one-key release button displayed on the interactive interface of the second client.

And then, the server distributes at least one commodity video which is determined to be selected by the second user to a commodity shelf of the live broadcast shop corresponding to the second user according to the distribution instruction.

For example, when the second user selects 3 commodity videos from the candidate commodity videos, the 3 commodity videos are added to a commodity shelf of a live broadcast shop corresponding to the second user together, so that the second-level synchronous release of the 3 commodity videos is realized. When the server receives an access request aiming at the live store, displaying a commodity shelf added with 3 commodity videos to a first user accessed in the store, wherein the newly added 3 commodities are displayed in a video format. In one embodiment, the original commodity on the commodity shelf can be displayed in a mode of keeping the image-text details until the second user deletes the original commodity with the image-text details.

Under the original commodity publishing mode, the video uploading can be supported while the commodity image and text are supported to be published, when a second user publishes a commodity, the picture, the text and/or the video of the commodity are uploaded in multiple steps according to a fixed service link, but the service link for uploading the video needs the second user to sequentially execute multiple operation steps, the commodity video is a local video, and the service link cannot realize the one-key second-level publishing of the commodity video at present.

Therefore, in some embodiments of the present specification, in order to ensure second-level publication of a commodity video, the commodity video is not uploaded according to an original service link any more, but a dedicated service link for one-key publication of the commodity video is correspondingly set based on a change of a commercial link acquired by the commodity video, so as to support second-level uploading of the commodity video. The original business link for publishing the image-text details and the video of the commodity can be reserved.

It should be noted that, in this embodiment of the present specification, the commodity video uploaded by the second client is not limited to only the commodity video provided by the video producer and having been shot, but in some embodiments, when the commodity video is published, the publishing of the local commodity video shot by the second user is also supported.

In some embodiments, when the commodity videos are distributed to the commodity shelf, the at least one commodity video determined to be selected is classified and identified, and the at least one commodity video is distributed to the commodity shelf of the live broadcast shop according to the identified category. Generally, the commodity video uploaded by the second user has a title, and the commodity category can be determined by identifying title keywords, or the voice information in the video is acquired, and the commodity category is determined by voice identification. And then storing the commodity video into a storage space corresponding to a shelf under the corresponding category of the live broadcast shop, namely classifying the commodities and then putting the classified commodities into the shop shelves.

For example, in a live-broadcast shop for home-run china tableware, the commodity shelves include a suit shelf and an accessory shelf such as a bowl shelf, a plate shelf, and a spoon, and when the second user selects the bone china plate recommended by the server, the server classifies the commodity as a plate and issues the commodity onto the corresponding plate shelf. In an actual application scenario, the classification of the released commodity video can be specifically set according to the classification condition of the current commodity shelf of the corresponding live broadcast shop. In the embodiment of the present specification, the classification of the commodities on the shelves may be performed according to a conventional commodity intrinsic classification manner, for example, the tea leaves may be classified into black tea, oolong tea, green tea, and the like, and when a corresponding tea leaf video is uploaded, the tea leaves may be automatically classified according to the attributes of the tea leaves and be correspondingly distributed on the shelves of the corresponding category.

In some embodiments, the server selects video information of the hot commodity matched with the attributes of the shop from the commodity video big data as a plurality of candidate commodity videos displayed for the second user. For example, if the live broadcast store is a store mainly selling office supplies, the current hot office supplies are selected from the commodities of the large quantity of office supplies based on the big data processing, and recommended to the second user. The hot commodities are commodities of which the transaction data meet a preset standard within a preset time period, for example, commodities of which the transaction amount exceeds 1 ten thousand and the goodness of evaluation reaches more than 85% within the past week, or commodities of which the transaction amount per day exceeds 5000 in a live broadcast room of other live broadcast shops, and the like.

Whether the commodity is a hot commodity or not is determined according to the nature of the commodity, and the transaction amount is not used as a unique reference standard, for example, when luxury goods are transacted, the hotspot degree of the current commodity is comprehensively judged by combining the transaction amount, the transaction difficulty degree and other factors.

In one embodiment of the store commodity information publishing method provided by the specification, from the perspective of the second client, the second client recommends the candidate commodity video to the user on an interactive interface facing the second user. The candidate commodity videos comprise hotspot commodity videos. For example, after the second user logs in the live store, the second client recommends the relevant hot goods on the interactive interface facing the second user in the live store according to the instruction of the service.

In a more specific embodiment, as mentioned above, the second client further presents a button for uploading the video of the merchandise in an interactive interface towards the second user. For example, buttons such as "upload by one key" and "distribute by one key" are displayed on the product distribution interface of the live broadcast shop. In another embodiment, a title input field may be displayed in the same interface as the button, and "please input the title of the product here" is prompted in the field; or setting a title input field on the next interface or the previous interface of the second user for submitting the commodity video. When the video producer has provided the title and the second user considers that the title does not need to be modified, the second user can skip the link of title input and directly submit the issuing instruction. And then uploading the commodity video information with the title to a server.

In a specific embodiment, referring to fig. 10, in the process of executing the second-level uploading of the commodity video, a data interaction flow between the server and the second client is as follows:

and S1000, the server receives the manufactured commodity video provided by the video manufacturer.

S1001, if it is determined that the second user U2 submits the publish request, the publish request is sent to the server.

The release request generally carries a store identity, such as a store id (identity document), of the live store corresponding to the second user.

S1002, the server specifies the store attribute of the live store corresponding to the store I D carried in the distribution request.

S1003, the server screens hot commodities matched with the store attributes according to the store attributes.

In the embodiment, the server screens out the video information of the hot commodities which are matched with the shop attributes from the commodity video big data according to the shop attributes. The embodiment is only an example, and in other embodiments, the video of the candidate commodity recommended by the server to the second user may have recommendation reference factors of multiple dimensions, and is not limited to recommending a hot commodity.

S1004, the server sends a display instruction to the second client.

The display instruction carries video information of the hot commodity as a candidate commodity video displayed for a second user.

And S1005, the second client displays the commodity video and the one-key release button of the hot commodity to the second user.

And S1006, the second client sends the issuing instruction to the server after receiving the issuing instruction submitted by the second user.

In this embodiment, the second user clicks the one-click issue button, and the issue command is considered to be submitted.

And S1007, distributing the commodity video selected by the U2 to a corresponding commodity shelf.

And S1008, feeding back a message to the second client to inform the second user of the result.

S1009, the second client displays the publishing result message, for example, displays a message that the publishing is successful, on the interactive interface facing the second user.

Referring to fig. 11, in a specific embodiment, fig. 11 is a system architecture diagram illustrating an example of an application scenario in which a commodity video is published to a shelf by one key. In this embodiment, the second user clicks on the "second-order published items" or the like on the seller's interactive interface to trigger the region of the publication request, then, after receiving the issue request, the server issues a display instruction to the second client 10b, so that the second client 10b displays a plurality of candidate commodity videos and one-key release buttons, for example, "Tieguanyin" and "jasmine tea" shown in 10 b' in fig. 11 are all candidate commodity videos, the second user selects "Tieguanyin" and "jasmine tea", clicking 'one-key issue', the second client sends the issue instruction to the server, after the server receives the issue instruction, releasing the Tieguanyin and the jasmine tea to corresponding shelves in second grade, feeding back information to a second client after successful release, so that the second client presents to the second user a prompt message "publish successful!as shown in 10 b! The goods have been released to the shelves ".

Wherein, in a further embodiment, the Tieguanyin is released on a oolong tea shelf and the jasmine tea is released on a scented tea shelf.

Therefore, after the second user determines the commodity video to be published, the commodity is published only by selecting the alternative commodity video and clicking the one-key publishing button, the operation is more convenient and fast, the one-key publishing can be realized, and the second-level uploading is realized, so that the waiting time of the second user is reduced.

Therefore, the commodity details in the video uploading mode are supported, compared with a picture and text expression mode, the commodity details are more visual, the information is richer, the expression modes are more various, and a more diversified presentation mode can be designed when the shop displays the commodities.

In some embodiments of the present specification, the display policy is not only adapted to the behavior characteristics of the first user who accesses the live store, but also determined by combining video information of various commodities on the current shelf of the live store, and the videotext information of commodities which are not off-shelf can be compatible, so as to obtain a comprehensive display policy.

For example, if one of the current behavior characteristics of the first user is a preference for chinese wind or a chinese classical element, the corresponding presentation policy should be set as: the method comprises the steps of preferentially displaying a commodity video with Chinese wind elements selected from a shop shelf, and playing a classical music piece of Chinese wind as background music in the process of playing the commodity video or also supporting the process of browsing image-text details, wherein for example, as an implementable mode, the background sound of the commodity video can be turned off and replaced by the classical music piece of Chinese wind.

Or, in some embodiments, the commodity videos uploaded by the second user are processed in advance, and multiple commodity videos of different styles are generated for the same commodity, that is, multiple backup videos are pre-made for the same commodity, for example, the same video is processed into multiple backup videos of different color tones, different background images, and different background music, and different backup videos are displayed for different users.

The above embodiments illustrate the display method and the distribution method of the store commodity information provided by the embodiments of the present specification in various aspects.

According to a sixth aspect, embodiments of the present specification further provide a method for displaying store commodity information, which is applicable to the following application scenarios: based on the historical behavior data of the user obtained from the second client to determine the behavior characteristics, applying the obtained corresponding display strategy to the commodity display of the first client; or acquiring the behavior characteristics of the user from the second client, and presenting the corresponding strategy through the first client. For example, a user views a live broadcast using a first client and initiates an access request, and the first client may determine a corresponding presentation policy based on behavior characteristics obtained from a second client. The first client and the second client are clients supporting live broadcasting of shops, and the second client and the first client can be two different APPs running on the basis of different platforms. Specifically, referring to fig. 12, the method includes:

s1201, acquiring an access request for a target live broadcast shop initiated when a user watches live broadcast at a first client, wherein the access request carries an identity of the user; s1202, acquiring historical behavior data formed by the user at least one client according to the identity; s1203, determining behavior characteristics corresponding to the user according to the historical behavior data of the user; s1204, determining a corresponding display strategy based on the behavior characteristics, and presenting the commodities in the target shop based on the display strategy.

Wherein, as in the above embodiment, the access request is submitted by the first user for the target live store while watching the live broadcast. The at least one live store-enabled client comprises a second client.

In the application scenario corresponding to the sixth aspect, the behavior feature for determining the display policy may be obtained based on another APP different from the first client, and the obtained display policy may be applied to merchandise display in a live store of the first client. A corresponding system architecture diagram for this application scenario is shown in fig. 13.

According to the seventh aspect, embodiments of the present specification further provide an electronic device, and referring to fig. 14, the electronic device 1400 includes:

a first obtaining unit 1401, configured to obtain an access request, which is initiated by a user at a client and is addressed to a target store, where the access request carries an identity of the user.

A first determining unit 1402 configured to determine a behavior feature corresponding to the user based on the identity of the user.

A first display unit 1403 configured to determine, based on the behavior characteristics, a corresponding display policy to present the goods in the target store based on the display policy.

According to an eighth aspect, embodiments of the present specification further provide an electronic device, and referring to fig. 15, the electronic device 1500 includes:

a second obtaining unit 1501, configured to obtain an access request for a target live broadcast shop, which is initiated when a user watches live broadcast at a client, where the access request carries an identity of the user;

a second determining unit 1502 configured to determine a behavior feature corresponding to the user based on the identity of the user;

a second presentation unit 1503 configured to determine, based on the behavior characteristics, a corresponding presentation policy to present the goods in the target live store based on the presentation policy.

In one embodiment, the second presentation unit 1503 is configured to: determining a first user class cluster to which a first user belongs according to the similarity between the behavior characteristics and class cluster centers of a plurality of pre-formed user class clusters; and determining the display strategy corresponding to the first user class cluster according to the preset mapping relation between the user class clusters and the display strategy. The plurality of user clusters are obtained by performing cluster analysis on the behavior characteristics of the plurality of first users, and the cluster center is the center point of the user cluster.

In one embodiment, the server further comprises a cluster analysis unit 1504 configured to:

acquiring historical behavior data corresponding to a plurality of users in a preset time period, acquiring behavior characteristics corresponding to the users respectively based on the historical behavior data, and performing clustering analysis by taking the behavior characteristics as clustering objects; and determining a plurality of user clusters and a cluster center corresponding to each user cluster according to the clustering result.

In one embodiment, the second determining unit 1502 is configured to: and acquiring historical behavior data of the user in a preset time period according to the identity of the user, and converting the historical behavior data into formatted data in a specified format to be used as behavior characteristics corresponding to the user.

In one embodiment, the second determining unit 1502 is further configured to:

In another embodiment, the second presentation unit 1503 is configured to: and inputting the behavior characteristics into a pre-trained neural network model, and classifying and identifying the behavior characteristics through the neural network model to obtain a display strategy corresponding to the first user.

In another embodiment, the server further comprises a training unit 1505 configured to: inputting the behavior characteristics into a pre-trained neural network model, and carrying out classification and identification on the behavior characteristics through the neural network model to obtain a display strategy corresponding to the user.

The neural network model can be obtained by training in the following way: acquiring a training sample set, wherein each training sample comprises behavior characteristics of a sample user and an identifier of an applicable display strategy; and training the multi-classified neural network model by taking each display strategy as a classification class based on the training sample set.

In a more specific embodiment, the server further includes a first publishing unit 1506 configured to: and after determining that a release instruction submitted by the user is received, releasing the commodity video selected by the user to a commodity shelf of the live broadcast shop by one key so as to upgrade the image-text detail page of the commodity on the commodity shelf into a video detail page for displaying.

In a more specific embodiment, the first publishing unit 1506 is configured to: and after determining that a release request initiated by a user at a client is received, indicating the client to at least display a plurality of pre-made alternative commodity videos to the user.

In a more specific embodiment, the first publishing unit 1506 is further configured to: and receiving the commodity video uploaded by the commodity video producer to serve as the alternative commodity video.

In a more specific embodiment, the first publishing unit 1506 may be further configured to: and instructing the client side to display a plurality of alternative commodity videos and the one-key publishing button to the user.

In a more specific embodiment, the first publishing unit 1506 is further configured to: determining the store attributes of the live stores; selecting video information of hot commodities matched with the store attributes from the video big data of the commodities according to the store attributes; the hot commodity is a commodity of which the transaction data reaches a preset standard in a preset time period.

In a more specific embodiment, the first publishing unit 1506 is further configured to:

In a more specific embodiment, the second presentation unit 1503 is further configured to: and determining a corresponding display strategy by combining each commodity video on the current shelf of the live broadcast shop based on the behavior characteristics.

According to a ninth aspect, embodiments of the present specification further provide an electronic device, and referring to fig. 16, the electronic device 1600 includes:

the first publishing unit 1601 is configured to, after determining that a publishing instruction submitted by a user is received, publish the commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user, so that a client faces the user visiting the shop, and upgrade image-text detail pages of commodities on the commodity shelf into video detail pages for display.

According to a tenth aspect, embodiments of the present specification further provide an electronic device, and referring to fig. 17, the electronic device 1700 includes:

a third obtaining unit 1701, configured to obtain an access request for a target live broadcast shop initiated by a user while a first client watches live broadcast, where the access request carries an identity of the user;

a collecting unit 1702, configured to collect, according to the identity, historical behavior data formed by the user at least one client, where the at least one client includes a second client that supports store live broadcast, and the second client and the first client operate based on different service platforms and both support store live broadcast;

a third determining unit 1703 configured to determine a behavior feature corresponding to the user according to the historical behavior data of the user;

a third display unit 1704 configured to determine a corresponding display policy based on the behavior feature to present the goods in the target store based on the display policy.

According to the eleventh aspect, embodiments of the present specification further provide a client, and referring to fig. 18, the client 1800 includes:

a first sending unit 1801, configured to send, to a server, an access request initiated by a user and directed to a target live broadcast store, where the access request carries an identity of the user, so that the server determines a behavior feature corresponding to the user;

a first receiving unit 1802, configured to receive a display policy issued by the server, where the display policy corresponds to behavior characteristics of the user;

a fourth presentation unit 1803, configured to present the goods in the target live store based on the presentation policy.

According to a twelfth aspect, embodiments of the present specification further provide a client, and referring to fig. 19, a client 1900 includes:

a second issuing unit 1901, configured to upload an issuing instruction submitted by a user to a server, so that the server issues a commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user, and is used to upgrade an image-text detail page of a commodity on the commodity shelf into a video detail page for presentation.

In one embodiment, the second publication unit 1901 is configured to: and uploading the release request submitted by the user to a server, and at least displaying a plurality of pre-made alternative commodity videos for the user according to the indication of the server.

In one embodiment, the second publication unit 1901 is further configured to: receiving hot commodity information issued by a server, and taking a commodity video corresponding to the hot commodity information as the alternative commodity video; the hot commodities are commodities which are selected by the server from the video big data of the commodities, matched with the store attributes of the live stores and have transaction data meeting a preset standard within a preset time period.

In one embodiment, the second publication unit 1901 is configured to: and the user displays a plurality of candidate commodity videos and the one-key publishing button.

According to a thirteenth aspect of the present specification, there is also provided a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the method of any of the first to sixth aspects described above.

According to a fourteenth aspect of the present specification, there is also provided a computing device comprising a memory and a processor, wherein the memory stores executable code, and the processor executes the executable code to implement the method of any one of the first to sixth aspects.

In summary, the display method for store commodity information provided by the embodiment of the present specification displays commodities facing different types of users according to different display strategies, and meets different requirements of different users in a targeted manner; the novel content E-commerce goods shelf is provided, an original goods uploading link is changed, goods information directly uses big data, selling point information is uploaded by one key, goods are put on shelf at the second level, original image-text detail pages are upgraded into video detail pages, the interest of goods display is increased, and meanwhile the problems that the threshold of goods released by a merchant at a grade opening is high and the matching efficiency of the flow of a live broadcast shop is improved are solved. In summary, a live store shelf of "thousands of people and thousands of faces" is provided to a first user accessing the store by watching live access, resulting in a richer and personalized access experience for the user.

Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in this invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.

The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the present invention should be included in the scope of the present invention.

Claims

1. A method for displaying store merchandise information, the method comprising:

acquiring an access request which is initiated by a user at a client and aims at a target shop, wherein the access request carries an identity of the user;

determining behavior characteristics corresponding to the user based on the identity of the user;

and determining a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

2. A display method of store commodity information, wherein the method comprises the following steps:

acquiring an access request which is initiated by a user when the user watches live broadcasting at a client and aims at a target live broadcasting shop, wherein the access request carries an identity of the user;

and determining a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target live broadcast shop based on the display strategy.

3. The method of claim 1 or 2, wherein determining, based on the behavioral characteristics, a corresponding presentation policy comprises:

determining the user class cluster to which the user belongs according to the similarity between the behavior characteristics and class cluster centers of a plurality of pre-formed user class clusters; the user clusters are obtained by performing cluster analysis on the behavior characteristics of a plurality of users, and one cluster center is a central point selected from one user cluster;

and determining a display strategy corresponding to the user class cluster according to a preset mapping relation between the user class cluster and the display strategy.

4. The method of claim 3, wherein the plurality of user class clusters are derived by:

acquiring historical behavior data corresponding to a plurality of users in a preset time period, acquiring behavior characteristics corresponding to each user respectively based on the historical behavior data, and performing cluster analysis by taking the behavior characteristics as cluster objects;

and determining a plurality of user clusters and a cluster center corresponding to each user cluster according to the clustering result.

5. The method of claim 1 or 2, wherein determining the behavioral characteristic corresponding to the user based on the identity of the user comprises:

6. The method of claim 5, wherein obtaining historical behavior data of the user over a predetermined period of time comprises:

7. The method of claim 1 or 2, wherein determining, based on the behavioral characteristics, a corresponding presentation policy comprises:

8. The method of claim 7, wherein the neural network model is trained by:

acquiring a training sample set, wherein each training sample comprises behavior characteristics of a sample user and an identifier of an applicable display strategy;

and training the multi-classified neural network model by taking each display strategy as a classification class based on the training sample set.

9. The method according to claim 1 or 2, wherein the method further comprises:

10. The method of claim 9, wherein determining that the user-submitted issuance instruction was received further comprises:

11. The method of claim 10, wherein, prior to instructing the client to present at least a plurality of pre-made videos of items of merchandise to the user, further comprising:

12. The method of claim 10, wherein the video of the item of merchandise is provided with title information.

13. The method of claim 10, wherein instructing the client to present at least a plurality of pre-made video of items of merchandise to a user comprises:

and instructing the client side to display a plurality of alternative commodity videos and the one-key publishing button to the user.

14. The method of claim 10, wherein determining that after receiving a user-initiated publication request based on a client, before instructing the client to display at least a plurality of pre-made video alternatives to the user, further comprises:

determining store attributes of the live stores;

and selecting a commodity video of the hot commodity matched with the store attribute from the commodity video big data as the alternative commodity video according to the store attribute, wherein the hot commodity is a commodity of which the transaction data in a preset time period reaches a preset standard.

15. The method of claim 9, wherein publishing the user-selected merchandise video to a merchandise shelf of the live store comprises:

classifying and identifying at least one commodity video selected by the user based on a commodity classification mode currently adopted by the live broadcast shop;

and according to the identification result, the at least one commodity video is distributed to the commodity shelf of the corresponding category in a classified mode.

16. The method of any of claims 10-15, wherein determining, based on the behavioral characteristics, a corresponding presentation policy comprises:

17. The method of claim 16, wherein the display strategy comprises at least any one or any combination of order of commodity presentation, page layout, page style, video tone, video background image, video background music.

18. A store commodity information distribution method, wherein the method comprises:

and after determining that a release instruction submitted by a user is received, releasing the commodity video selected by the user to a commodity shelf of a live broadcast shop corresponding to the user so as to upgrade image-text detail pages of commodities on the commodity shelf into video detail pages for presentation.

19. A display method of store commodity information, wherein the method comprises the following steps:

acquiring an access request which is initiated by a user when the user watches live broadcasting at a first client and aims at a target live broadcasting shop, wherein the access request carries an identity of the user;

acquiring historical behavior data formed by the user at least one client according to the identity, wherein the at least one client comprises a second client supporting store live broadcast, and the second client and the first client run on the basis of different service platforms and both support store live broadcast;

determining behavior characteristics corresponding to the user according to the historical behavior data of the user;

20. A display method of store commodity information, wherein the method comprises the following steps:

sending an access request which is initiated by a user and aims at a target live broadcast shop to a server, wherein the access request carries an identity of the user, so that the server determines behavior characteristics corresponding to the user;

receiving a display strategy issued by the server, wherein the display strategy corresponds to the behavior characteristics of the user;

and presenting the commodities in the target live broadcast shop based on the display strategy.

21. A store commodity information distribution method, wherein the method comprises:

22. The method of claim 21, wherein before uploading the issue instruction submitted by the user to the server, further comprising:

23. The method of claim 22, wherein after uploading the publishing request submitted by the user to a server, before displaying at least a plurality of pre-made video alternatives to the user, further comprising:

24. The method of claim 22, wherein presenting at least a plurality of pre-made videos of candidate items to the user comprises:

25. An electronic device, wherein the electronic device comprises:

the system comprises a first acquisition unit, a second acquisition unit and a third acquisition unit, wherein the first acquisition unit is configured to acquire an access request which is initiated by a user at a client and aims at a target shop, and the access request carries an identity of the user;

a first determining unit configured to determine a behavior feature corresponding to the user based on the identity of the user;

the first display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

26. An electronic device, wherein the electronic device comprises:

the second acquisition unit is configured to acquire an access request which is initiated by a user when the user watches live broadcasting at a client and aims at a target live broadcasting shop, wherein the access request carries the identity of the user;

a second determining unit configured to determine a behavior feature corresponding to the user based on the identity of the user;

and the second display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target live store based on the display strategy.

27. An electronic device, wherein the electronic device comprises:

28. An electronic device, wherein the electronic device comprises:

the third acquisition unit is configured to acquire an access request which is initiated by a user when the first client watches live broadcasting and aims at a target live broadcasting shop, wherein the access request carries the identity of the user;

the acquisition unit is configured to acquire historical behavior data formed by the user at least one client according to the identity, the at least one client comprises a second client supporting store live broadcast, and the second client and the first client run on the basis of different service platforms and both support store live broadcast;

a third determining unit, configured to determine a behavior feature corresponding to the user according to historical behavior data of the user;

and the third display unit is configured to determine a corresponding display strategy based on the behavior characteristics so as to present the commodities in the target shop based on the display strategy.

29. A client, wherein the client comprises:

the system comprises a first sending unit, a second sending unit and a third sending unit, wherein the first sending unit is configured to send an access request which is initiated by a user and aims at a target live broadcast shop to a server side, and the access request carries an identity of the user so that the server side can determine a behavior characteristic corresponding to the user;

the first receiving unit is configured to receive a display strategy issued by the server, wherein the display strategy corresponds to the behavior characteristics of the user;

and the fourth display unit is configured to display the commodities in the target live store based on the display strategy.

30. A client, wherein the client comprises:

31. A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed in a computer, causes the computer to perform the method of any of claims 1-24.

32. A computing device comprising a memory and a processor, wherein the memory has stored therein executable code that when executed by the processor implements the method of any of claims 1-24.