CN114245165B

CN114245165B - Live broadcast recommendation method and device, electronic equipment and storage medium

Info

Publication number: CN114245165B
Application number: CN202111543204.6A
Authority: CN
Inventors: 吕梦天; 甄云启; 郭树礼
Original assignee: Beijing Dajia Internet Information Technology Co Ltd
Current assignee: Beijing Dajia Internet Information Technology Co Ltd
Priority date: 2021-12-16
Filing date: 2021-12-16
Publication date: 2023-04-25
Anticipated expiration: 2041-12-16
Also published as: CN114245165A

Abstract

The disclosure relates to a live broadcast recommendation method, a live broadcast recommendation device, electronic equipment and a storage medium, and belongs to the technical field of live broadcast. The method comprises the following steps: after article information of articles in the living broadcast room and article demand information of the plurality of accounts are obtained according to a living broadcast plan of the living broadcast room and image data of the plurality of accounts, at least one target account is determined from the plurality of accounts based on the matching degree between the two types of information, and the living broadcast room is recommended to the at least one target account. In the process, the release plan of the articles in the living broadcast room is obtained in advance according to the living broadcast plan, so that the article demand information of a plurality of accounts can be combined, target accounts which are likely to transfer resources based on the articles in the living broadcast room can be screened out from the plurality of accounts, the living broadcast room is recommended to the target accounts in a targeted and targeted manner, the accuracy of living broadcast recommendation is improved, and the conversion rate of the accounts in the living broadcast room is further effectively improved.

Description

Live broadcast recommendation method and device, electronic equipment and storage medium

Technical Field

The disclosure relates to the technical field of live broadcasting, in particular to a live broadcasting recommendation method, a live broadcasting recommendation device, electronic equipment and a storage medium.

Background

With the continuous development of live broadcast and electronic commerce, the combination of live broadcast and electronic commerce is also more and more compact. The anchor issues a live broadcast forecast for the items in the live broadcast room by creating a live broadcast plan, and introduces the items according to the live broadcast plan so that the audience can watch the live broadcast at any time to purchase the items in the live broadcast room. In the live broadcast scene of the electronic commerce, how to pertinently recommend live broadcast to the live broadcast room so as to improve the conversion rate of audience in the live broadcast room becomes a problem to be solved.

Disclosure of Invention

The invention provides a live broadcast recommendation method, a live broadcast recommendation device, electronic equipment and a storage medium, which can improve the accuracy of live broadcast recommendation and further effectively improve the conversion rate of accounts in a live broadcast room. The technical scheme of the present disclosure is as follows:

according to a first aspect of an embodiment of the present disclosure, there is provided a live recommendation method, including:

determining item information of an item in a target living broadcast room based on a living broadcast plan of the target living broadcast room, wherein the living broadcast plan indicates a release plan of the item in the target living broadcast room;

determining item demand information for a plurality of accounts based on representation data of the plurality of accounts, the item demand information indicating demand information for items by the accounts;

At least one target account is determined from the plurality of accounts based on a degree of matching between the item information and item demand information for the plurality of accounts, and the target live room is recommended to the at least one target account.

According to the live broadcast recommendation method, after article information of articles in a live broadcast room and article demand information of a plurality of accounts are obtained according to a live broadcast plan of the live broadcast room and image data of the plurality of accounts, at least one target account is determined from the plurality of accounts based on matching degree between the two types of information, and the live broadcast room is recommended to the at least one target account. In the process, the release plan of the articles in the living broadcast room is obtained in advance according to the living broadcast plan, so that the article demand information of a plurality of accounts can be combined, target accounts which are likely to transfer resources based on the articles in the living broadcast room can be screened out from the plurality of accounts, the living broadcast room is recommended to the target accounts in a targeted and targeted manner, the accuracy of living broadcast recommendation is improved, and the conversion rate of the accounts in the living broadcast room is further effectively improved.

In some embodiments, the determining item information for items within the target living room based on the living program of the target living room includes:

And extracting a first keyword for describing each item based on the live broadcast plan to obtain item information of each item, wherein the first keyword indicates basic information of the item and a resource transfer form for transferring resources based on the item.

By means of the method for reflecting the article information of the articles by the keywords, basic information and resource transfer forms of the articles can be intuitively and conveniently known, and data support is provided for matching degree between the article information of the articles and the article demand information of the account in the follow-up measurement.

In some embodiments, the determining the item demand information for the plurality of accounts based on the representation data for the plurality of accounts includes:

and extracting a second keyword for describing each account based on the image data of each account to obtain the article demand information of each account, wherein the second keyword indicates the basic information of the articles required by the account and the demand information of the account for different resource transfer forms.

By means of the method that the article demand information of the account is embodied through the keywords, basic information of articles required by the account and demand information of different resource transfer forms can be intuitively and conveniently known, and data support is provided for matching degree between the article information of the articles and the article demand information of the account in the follow-up measurement.

In some embodiments, the determining at least one target account from the plurality of accounts based on the degree of matching between the item information and the item demand information for the plurality of accounts includes:

acquiring first reference information of each account based on the matching degree between the item information and the item demand information of each account, wherein the first reference information indicates the probability of the account performing resource transfer based on the item in the target live broadcasting room;

and determining an account, of the plurality of accounts, for which the first reference information meets a first target condition as the target account.

By means of the method, the account which is likely to transfer resources based on any item in the target living broadcast room is determined to be the target account by combining the item information of each item in the target living broadcast room and the item demand information of each account in the plurality of accounts, and after the target living broadcast room is recommended to the target accounts, the target account is likely to transfer resources based on the item in the target living broadcast room, so that the accuracy of living broadcast recommendation is improved, and the time for watching living broadcast by the account in the target living broadcast room and the conversion rate of the account are further improved.

Acquiring second reference information of each account based on the matching degree between the item information and the item demand information of each account, wherein the second reference information indicates the probability of the account performing resource transfer based on each item in the target live broadcasting room;

and determining an account, of the plurality of accounts, for which the second reference information meets a second target condition as the target account.

By means of the method, the account which is likely to transfer resources based on an item in the target living broadcast room is determined to be the target account by combining the item information of each item in the target living broadcast room and the item demand information of each account in the plurality of accounts, and after the target living broadcast room is recommended to the target accounts, the target account is likely to transfer resources based on the item in the target living broadcast room, so that the accuracy of living broadcast recommendation is improved, and the conversion rate of the account in the target living broadcast room is improved in a targeted manner.

In some embodiments, the recommending the target live room to the at least one target account includes:

recommending the target live room to the at least one target account before the live start time of the target live room; and/or

And recommending the target live room to the at least one target account after the live start time of the target live room.

By recommending the target live broadcasting room before the live broadcasting start time of the target live broadcasting room, the target live broadcasting room can be timely recommended to the target account before the target live broadcasting room starts live broadcasting, so that the target account is guided to reserve the target live broadcasting room or can be timely watched when the target live broadcasting room starts live broadcasting, and more accounts are attracted to enter the target live broadcasting room. By recommending the target live broadcasting room after the live broadcasting start time of the target live broadcasting room, more accounts can be pertinently attracted to enter the target live broadcasting room in the live broadcasting process of the target live broadcasting room, the accuracy of live broadcasting recommendation is improved, and the conversion rate of the accounts in the target live broadcasting room is further improved. Of course, the target live broadcasting room can be recommended to at least one target account before and after the live broadcasting starting time of the target live broadcasting room, so that the account is attracted to enter the live broadcasting room to the maximum extent, the accuracy of live broadcasting recommendation is greatly improved, and the conversion rate of the account in the target live broadcasting room is effectively improved.

And sending a first play prompt to the terminal where the at least one target account is located, wherein the first play prompt indicates the play information of the target live broadcasting room.

By the method, the target account can be timely prompted to watch the target live broadcasting room, so that the purpose of the account in the live broadcasting room after the target live broadcasting room is opened is stronger, the accuracy of live broadcasting recommendation is improved, and the conversion rate of the account in the live broadcasting room and the time for the account to watch the live broadcasting room are further improved.

In some embodiments, the method further comprises:

determining a target item corresponding to the item demand of each target account from at least one item of the target living room based on the item demand information of each target account;

the recommending the target live room to the at least one target account includes:

and sending a second play prompt to the terminal where the at least one target account is located, wherein the second play prompt indicates the play information of the target object in the target live broadcasting room.

By the method, the target account can be prompted to view the target live broadcasting room in a targeted manner based on the object demand of the target account, the purpose of the account in the live broadcasting room after the target live broadcasting room is opened is further improved, the accuracy of live broadcasting recommendation is improved, and the conversion rate of the account in the live broadcasting room and the time for viewing the live broadcasting room by the account are further effectively improved.

According to a second aspect of the embodiments of the present disclosure, there is provided a live recommendation apparatus, including:

a first determining unit configured to execute a live program based on a target live room, determining item information of an item in the target live room, the live program indicating a distribution plan of the item in the target live room;

a second determining unit configured to perform determining item demand information of a plurality of accounts based on image data of the plurality of accounts, the item demand information indicating demand information of the account for items;

and a recommending unit configured to perform a recommendation of the target live room to at least one target account from among the plurality of accounts based on the matching degree between the item information and the item demand information of the plurality of accounts.

In some embodiments, the first determining unit is configured to perform:

In some embodiments, the second determining unit is configured to perform:

In some embodiments, the recommendation unit is configured to perform:

In some embodiments, the apparatus further comprises:

a third determining unit configured to perform determining a target item corresponding to an item demand of each target account from at least one item of the target living room based on the item demand information of each target account;

the recommendation unit is configured to perform:

According to a third aspect of embodiments of the present disclosure, there is provided an electronic device comprising:

One or more processors;

a memory for storing the processor-executable program code;

wherein the processor is configured to execute the program code to implement the live recommendation method described above.

According to a fourth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium comprising: the program code in the computer readable storage medium, when executed by a processor of an electronic device, enables the electronic device to perform the live recommendation method described above.

According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product comprising one or more instructions that are executable by one or more processors of an electronic device to enable the electronic device to perform the live recommendation method described above.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure and do not constitute an undue limitation on the disclosure.

FIG. 1 is a schematic illustration of an implementation environment for a live recommendation method, according to an example embodiment;

FIG. 2 is a flowchart illustrating a live recommendation method according to an example embodiment;

FIG. 3 is a flowchart illustrating a live recommendation method according to an exemplary embodiment;

FIG. 4 is a schematic diagram of a live schedule shown in accordance with an exemplary embodiment;

FIG. 5 is a schematic diagram illustrating a live recommendation method according to an example embodiment;

FIG. 6 is a block diagram illustrating a live recommender in accordance with an exemplary embodiment;

fig. 7 is a block diagram of a server, according to an example embodiment.

Detailed Description

In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the foregoing figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as detailed in the accompanying claims.

The data referred to in this disclosure may be data authorized by the user or sufficiently authorized by the parties.

Fig. 1 is a schematic view of an implementation environment of a live recommendation method according to an exemplary embodiment, and referring to fig. 1, the implementation environment includes: a terminal 101 and a server 102.

The terminal 101 may be at least one of a smart phone, a smart watch, a desktop computer, a laptop computer, a virtual reality terminal, an augmented reality terminal, a wireless terminal, and a laptop portable computer. The terminal 101 has a communication function and can access a wired network or a wireless network. Terminal 101 may be referred to generally as one of a plurality of terminals, with embodiments of the present disclosure being illustrated only by terminal 101. Those skilled in the art will recognize that the number of terminals may be greater or lesser. In the disclosed embodiment, the terminal 101 includes a first terminal 1011 and a second terminal 1012. The first terminal 1011 is a terminal used by the first account, and illustratively, a live broadcast client is operated on the first terminal 1011 and is used for providing a creation function of a live broadcast program for a live broadcast room of the first account, before the first account is live broadcast, the live broadcast program of the live broadcast room is created by the first account through the live broadcast client, and then the items are introduced according to the created live broadcast program in the live broadcast process. In some embodiments, the first account is a main account of the living room, and in other embodiments, the first account is another staff member of the living room, such as a merchant, etc., which is not limited by the disclosed embodiments. The second terminal 1012 is a terminal used by other accounts (such as an account of watching a live broadcast room), and illustratively, a live broadcast client is operated on the second terminal 1012 and is used for providing a live broadcast function of a target live broadcast room, and other accounts can watch and enter the corresponding live broadcast room to watch and broadcast according to own requirements through the live broadcast client. In some embodiments, the second terminal 1012 is also used to prompt other accounts to view the live room of the first account.

The server 102 may be an independent physical server, a server cluster or a distributed file system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, a content delivery network (Content Delivery Network, CDN), basic cloud computing services such as big data and an artificial intelligence platform. The server 102 and the terminal 101 may be directly or indirectly connected through wired or wireless communication, which is not limited by the embodiments of the present disclosure. Illustratively, server 102 provides background services for live management clients running on terminal 1011 and live application clients running on terminal 1012. Alternatively, the number of servers 102 may be greater or lesser, which is not limited by the disclosed embodiments. Of course, the server 102 may also include other functional servers to provide more comprehensive and diverse services.

Fig. 2 is a flowchart illustrating a live recommendation method, as shown in fig. 2, performed by a server, according to an exemplary embodiment, comprising the steps of:

In step 201, the server determines item information for items within the target living room based on a living broadcast schedule of the target living room, the living broadcast schedule indicating a distribution schedule of items within the target living room.

In the embodiment of the disclosure, the target living room is any living room, and at least one article (such as an article, including a virtual article and a physical article, etc. without limitation) can be released, so that an account watching the target living room can transfer resources (such as purchasing articles, etc.) in the target living room based on the released article by releasing the article in the target living room. The live broadcast schedule indicates a release schedule of the articles in the target live broadcast room, and the release schedule of the articles illustratively comprises release time, attribute information, resource transfer forms and the like of the articles, wherein the release time of the articles indicates the release sequence, release time and the like of the articles. The attribute information of the item indicates an item name of the item, an item picture, an item inventory, an item price, and the like. The resource transfer forms of the items indicate the transaction forms (e.g., funding pre-sales, full payments, pay-before-use, etc.) and marketing campaign forms (e.g., coupons, discounts, seconds kills, etc.) of the items, which are not limited by embodiments of the present disclosure.

In some embodiments, for any item within the target living room, the item information for that item is embodied by keywords that describe the item. For example, taking the article a as a blower, the article information of the article a includes: blower, household appliances, price XX, inventory YY, discount amount ZZ, etc., which are not limited by the disclosed embodiments.

In step 202, the server determines item demand information for a plurality of accounts based on the representation data for the plurality of accounts, the item demand information indicating demand information for items for the accounts.

In the embodiment of the disclosure, the accounts are any accounts, and the portrait data is used for characterizing the attribute, behavior habit and the like of the accounts. In some embodiments, the portrait data includes a plurality of different types of data, such as, for example, an account registered for user a for an account, including gender, age group, region, type of item frequently purchased, price range of item frequently purchased, type of direct broadcast room of interest, type of anchor of interest, and sensitivity to item preference information, etc., which are not limiting embodiments of the present disclosure.

In some embodiments, for any one of the plurality of accounts, the item demand information for that account is embodied by keywords that describe the account. For example, taking account a as user a, the item demand information of account a includes: the electrical appliance, price zone XX-XX, office of interest, major broadside effort of interest are all greater than YY, and large coupons of interest, etc., which are not limited by the disclosed embodiments.

In step 203, the server determines at least one target account from the plurality of accounts based on the matching degree between the item information and the item demand information of the plurality of accounts, and recommends the target live room to the at least one target account.

In the embodiment of the disclosure, the matching degree between the item information and the item demand information of a plurality of accounts is used for measuring the probability that a certain account performs resource transfer based on the item (such as the probability of purchasing goods) in the target living broadcast room, and a reference value is provided for determining a target account, that is, the matching degree between the item information and the item demand information of the certain account is used for measuring the probability that the account performs resource transfer based on the item in the target living broadcast room. Illustratively, the higher the degree of match between the item information and the account for any account, the higher the likelihood that the target account will transfer resources based on items within the target living room.

In the live broadcast recommendation method provided by the embodiment of the disclosure, after article information of articles in a live broadcast room and article demand information of a plurality of accounts are obtained according to a live broadcast plan of the live broadcast room and image data of the plurality of accounts, at least one target account is determined from the plurality of accounts based on matching degree between the two types of information, and the live broadcast room is recommended to the at least one target account. In the process, the release plan of the articles in the living broadcast room is obtained in advance according to the living broadcast plan, so that the article demand information of a plurality of accounts can be combined, target accounts which are likely to transfer resources based on the articles in the living broadcast room can be screened out from the plurality of accounts, the living broadcast room is recommended to the target accounts in a targeted and targeted manner, the accuracy of living broadcast recommendation is improved, and the conversion rate of the accounts in the living broadcast room is further effectively improved.

The foregoing fig. 2 is merely a basic flow of the disclosure, and the scheme provided in the disclosure is further described below based on a specific embodiment, and fig. 3 is a flowchart illustrating a live recommendation method according to an exemplary embodiment, where the embodiment is executed by a server, and referring to fig. 3, the method includes:

In step 301, a server obtains a live schedule for a target live room that indicates a distribution schedule for items within the target live room.

In an embodiment of the disclosure, a live program of a target live room is published on a server by a first account. In some embodiments, the first account is a hosting account of the target live room, or other staff members of the target live room (e.g., merchants, etc.), to which embodiments of the present disclosure are not limited. The method comprises the steps that a live broadcast client is operated on a first terminal, a first account is a main broadcast account, a live broadcast plan of a target live broadcast room is created through the main broadcast client, the first terminal responds to a live broadcast plan release operation of the main broadcast account for the live broadcast plan, a live broadcast plan release request is sent to a server, and the server acquires the live broadcast plan of the target live broadcast room when receiving the live broadcast plan release request. By the method, when the server has the release requirement of the live broadcast program in each live broadcast room, the live broadcast program in the live broadcast room can be immediately acquired, and the efficiency of acquiring the live broadcast program by the server is improved.

In some embodiments, the server obtains a live program for each live room in the target platform every first time period. The first duration is a preset duration, for example, 10 minutes, which is not limited thereto. The target platform is any platform, for example, the target platform is a live application program, or a shopping application program, etc., which is not limited thereto. By the method, the server can update the live broadcast plans of all live broadcast rooms in the platform at regular time, the live broadcast recommendation range is enlarged, corresponding live broadcast recommendation services can be provided for all live broadcast rooms in the platform, and the coverage rate of live broadcast recommendation is improved.

In some embodiments, the server obtains a live schedule for a target live room based on a live recommendation request for the target live room. The live broadcast recommendation request is sent by the first terminal and used for requesting recommendation of the target live broadcast room to a plurality of accounts. The method comprises the steps that a live broadcast client is operated on a first terminal, a first account is a main broadcast account, after the main broadcast account creates a live broadcast plan of a target live broadcast room through the live broadcast client, the first terminal responds to live broadcast recommendation operation of the main broadcast account on the target live broadcast room, a live broadcast recommendation request is sent to a server, and the server acquires the live broadcast plan of the target live broadcast room when receiving the live broadcast recommendation request. By the method, the server can acquire the live program of the live broadcasting room when the live broadcasting recommendation needs exist in each live broadcasting room, so that recommendation can be performed pertinently according to the recommendation needs of the live broadcasting room, and personalized requirements of the live broadcasting room recommendation are met.

The contents of the live schedule of the target live room are described below.

Illustratively, the live schedule of the target live room includes release opportunities, attribute information, resource transfer forms, and the like of the items in the target live room. Wherein the release timing of the articles indicates the release order, release time, etc. of the articles. The attribute information of the item indicates an item name of the item, an item picture, an item inventory, an item price, and the like. The resource transfer forms of the items indicate the transaction forms (e.g., funding pre-sales, full payments, pay-before-use, etc.) and marketing campaign forms (e.g., coupons, discounts, kills for seconds, etc.) of the items. In some embodiments, the live schedule further includes attribute information of the target live room, such as a live title, a live start time, live content, and the like. In other embodiments, the living broadcast program further includes a target living broadcast room resource transfer form that indicates a living broadcast room transaction form (e.g., payment channels including A and B, payment platforms including C and D, etc.) and a marketing campaign form (e.g., living broadcast room coupons, living broadcast room benefits, etc.). It should be noted that the foregoing content about the live program is merely illustrative, and in some embodiments, the content of the live program can be adjusted according to actual requirements, which is not limited by the embodiments of the present disclosure. In addition, through the mode of creating the live broadcast plan in advance, the first account can release articles according to the content indicated by the live broadcast plan, so that the labor investment in the live broadcast process is reduced, the operation cost is reduced, the operation pressure in the live broadcast process is reduced, and the man-machine interaction efficiency is improved.

In step 302, the server determines item information for items within the target living room based on the living schedule of the target living room.

In the embodiment of the disclosure, after acquiring the live broadcast plan of the target live broadcast room, the server processes the live broadcast plan to obtain the article information of the articles in the target live broadcast room. For any item, the item information for that item can characterize the underlying information for that item and the form of resource transfer based on that item. Illustratively, the article is taken as an example of the commodity, and the article information of the article is that related to the selling of the article (such as commodity class, price, preferential and the like, which is not limited thereto).

In some embodiments, the item information of an item is embodied by keywords that describe the item. Illustratively, the server extracts a first keyword describing each item based on the live schedule to obtain item information of each item, the first keyword indicating basic information of the item. In some embodiments, the first keyword further indicates a resource transfer form for transferring resources based on the item. Schematically, after acquiring a live broadcast plan of a target live broadcast room, a server extracts text content of the live broadcast plan, performs text recognition on the text content, and performs keyword extraction on basic information of recognized objects to obtain a first keyword for describing each object. In some embodiments, the server performs keyword recognition on the recognized basic information of the items and the content related to the resource transfer form of the resource transfer based on the items, and obtains a first keyword for describing each item. In addition, it should be noted that the number of the first keywords is one or more, which is not limited in the embodiments of the present disclosure. For example, taking the article a as a blower, the article information of the article a includes: blowers, household appliances, price XX, inventory YY, marketing campaigns as large coupons, live house coupons may be used, etc., as embodiments of the present disclosure are not limited in this regard. By means of the method for reflecting the article information of the articles by the keywords, basic information and resource transfer forms of the articles can be intuitively and conveniently known, and data support is provided for matching degree between the article information of the articles and the article demand information of the account in the follow-up measurement.

In some embodiments, the item information of the item is embodied by a tag of the item. Illustratively, the server extracts the item characteristics of each item based on the live schedule, generates a first tag for each item based on the item characteristics of each item to obtain item information for each item, the first tag indicating base information for the item. In some embodiments, the first tag further indicates a resource transfer form for transferring resources based on the item. After acquiring a live broadcast plan of a target live broadcast room, the server extracts text content of the live broadcast plan, performs text recognition on the text content, performs feature extraction on basic information of recognized objects to obtain object features of each object, and generates a first label of each object based on the object features of each object. In some embodiments, the server performs feature extraction on the identified basic information of the items and the content related to the resource transfer form based on the resource transfer of the items to obtain item features of each item, and generates a first tag of each item based on the item features of each item. In addition, the first label is formed by combining any of letters, numbers and symbols. For example, taking the article a as an example of a blower, the article information of the article a is represented by a tag 01a5060yhq, where 01 represents the blower, a represents the household small electric appliance, 50 represents the price zone, 60 represents the inventory, yhq represents the marketing campaign as a coupon, and the embodiment of the present disclosure is not limited thereto. It should be noted that, in some embodiments, the form of the first tag can be adjusted according to actual requirements, so long as the basic information and the resource transfer form of the article can be represented. By means of the method for reflecting the article information of the articles through the labels, the data volume of the article information can be simplified on the basis of the basic information and the resource transfer form of the characterization articles, the data processing volume of matching degree between the article information of the articles and the article demand information of the account can be reduced, and the live broadcast recommendation efficiency can be improved.

Through the

steps

301 and 302, the server obtains the item information of the items in the target living broadcast room from the living broadcast plan side of the target living broadcast room, so as to sense the distribution plan of the items in the living broadcast (such as which items distributed in the living broadcast room and which marketing activities in the living broadcast room exist) in advance before the living broadcast starts, and provide data support for screening target accounts from a plurality of accounts.

In step 303, the server obtains portrait data for a plurality of accounts.

In the embodiment of the present disclosure, the plurality of accounts are arbitrary accounts, and a terminal used by the plurality of accounts is referred to as a second terminal. In some embodiments, the plurality of accounts are all accounts logged on to the target platform. By acquiring the portrait data of all the accounts on the target platform, the subsequent screening of the target accounts can be performed based on all the accounts on the target platform, so that the coverage rate of live broadcast recommendation is improved. In some embodiments, the plurality of accounts are partial accounts on the target platform. Illustratively, the plurality of accounts are accounts for opening a live broadcast recommendation function, for example, a live broadcast application client is operated on the second terminal, the live broadcast application client provides the live broadcast recommendation function, and the accounts can open the live broadcast recommendation function through the live broadcast application client. By the method, the account with the live broadcast recommendation requirement can be subjected to live broadcast recommendation in a targeted manner, and the accuracy of live broadcast recommendation is improved on the basis of meeting the personalized requirement of live broadcast recommendation, so that the conversion rate of the account in a live broadcast room is improved.

In some embodiments, the server stores portrait data for all accounts registered on the target platform. Illustratively, the server updates the representation data of the account according to the attributes and behavior habits of the account, etc., every second time period (e.g., 1 day). When the plurality of accounts are part of the accounts on the target platform, the server acquires the portrait data of the plurality of accounts from the portrait data of all the accounts by using the identifiers (such as account ids) of the plurality of accounts as indexes. By the method, the server stores the portrait data of the account, signaling interaction between the server and other data centers can be reduced, and live broadcast recommendation efficiency is improved.

In some embodiments, the server is associated with a data center for storing representation data of the account. In some embodiments, the server sends a portrait data acquisition request to the data center based on account identifications of the plurality of accounts, and the data center feeds back portrait data of the plurality of accounts to the server after receiving the portrait data acquisition request. By the method, the server pulls the portrait data of the account to the data center when needed, so that the storage space of the server can be saved, and the service life of the server can be prolonged.

In step 304, the server determines item demand information for the plurality of accounts based on the representation data for the plurality of accounts.

In the embodiment of the disclosure, after the server acquires the portrait data of the plurality of accounts, the portrait data is processed to obtain the article demand information of the plurality of accounts. For any account, the article requirement information of the account can represent basic information of articles required by the account, taking articles as commodities as examples, and the article requirement information of the account can represent commodity classes, prices and the like which are purchased frequently by the account. In some embodiments, the item demand information of the account can also characterize the demand information of the account for different resource transfer forms, taking items as goods as examples, the item demand information of the account can characterize the sensitivity of the account to different marketing campaigns, and so on.

In some embodiments, the item demand information for an account is embodied by keywords that describe the account. Illustratively, the server extracts a second keyword for describing each account based on the image data of each account, and obtains the item demand information of each account, wherein the second keyword indicates the basic information of the items required by the account. In some embodiments, the second key further indicates demand information for the account for different resource transfer forms. Schematically, after the server acquires the portrait data of the account, the server extracts the text content of the portrait data, performs text recognition on the text content, and performs keyword extraction on the recognized content related to the basic information of the required object of the account to obtain a second keyword for describing the account. In some embodiments, the server performs keyword recognition on the recognized content related to the basic information of the required items of the account and the requirement information of the account for different resource transfer forms, and obtains a second keyword for describing each account. In addition, the number of the second keywords is one or more, which is not limited in the embodiments of the present disclosure. For example, taking account a as user a, the item demand information of account a includes: the embodiments of the present disclosure are not limited to such items as appliances, price intervals XX-XX, items released by regular purchases of official living broadcast rooms, items released by regular purchases of anchor fans greater than YY's anchor, regular use of large coupons, and the like. By means of the method that the article demand information of the account is embodied through the keywords, basic information of articles required by the account and demand information of different resource transfer forms can be intuitively and conveniently known, and data support is provided for matching degree between the article information of the articles and the article demand information of the account in the follow-up measurement.

In some embodiments, the item demand information for the account is embodied by a tag for the account. Schematically, the server extracts the demand characteristics of the articles of each account based on the portrait data of the accounts, and generates a second label of each account based on the demand characteristics of the articles of each account to obtain the demand information of the articles of each account, wherein the second label indicates the basic information of the articles. In some embodiments, the second tag further indicates a resource transfer form for transferring resources based on the item. After the server acquires the portrait data of the accounts, extracting text content of the portrait data, carrying out text recognition on the text content, carrying out feature extraction on the recognized content related to basic information of the required articles of the accounts to obtain the article demand features of each account, and generating a second label of each account based on the article demand features of each account. In some embodiments, the server performs feature extraction on the identified basic information of the required items of the accounts and the content related to the demand information of the accounts for different resource transfer forms to obtain the item demand features of each account, and generates a second label of each account based on the item demand features of each account. In addition, it should be noted that in some embodiments, the second label is formed by any combination of letters, numbers and symbols. For example, taking account a as user a, the article requirement information of the account a is represented by a label a50GFdVyhq, where a represents a household small appliance, 50 represents a price interval, GF represents an article issued by an official direct broadcast room frequently purchased by the account, dV represents an article issued by a host who frequently purchased a fan number of the host more than YY by the account, yhq represents a large coupon frequently used by the account, which is not limited by the embodiments of the present disclosure. It should be noted that, in some embodiments, the form of the second tag can be adjusted according to actual requirements, so long as the basic information of the required items of the account and the requirement information of different resource transfer forms can be represented. By means of the method for representing the article demand information of the account through the label, the data size of the article demand information can be simplified on the basis of basic information representing the articles required by the account and demand information for different resource transfer forms, the data processing size of matching degree between the article information of the articles and the article demand information of the account can be reduced, and live broadcast recommendation efficiency can be improved.

Through the

steps

303 and 304, the server obtains the article requirement information of the accounts from the portrait data side of the accounts, so as to obtain the basic information of the articles required by the accounts and the requirement information of different resource transfer forms (such as which articles the accounts want to purchase and which marketing activities tend to purchase articles, etc.), and provide data support for the screening target accounts from the accounts.

It should be noted that, in some embodiments, the server performs

steps

301 and 302 after performing

steps

303 and 304. In other embodiments, the server performs step 303 and step 304 synchronously when performing step 301 and step 302, which is not limited by the embodiments of the present disclosure.

In step 305, the server determines at least one target account from the plurality of accounts based on the degree of matching between the item information and the item demand information for the plurality of accounts.

In the embodiment of the disclosure, after acquiring the item information of the item in the target living broadcast room and the item demand information of the plurality of accounts, the server calculates the matching degree between the item information and the item demand information of the plurality of accounts, and determines the account meeting the requirements in the plurality of accounts as the target account. In some embodiments, the degree of matching is embodied by a percentage, and in some embodiments, the degree of matching is embodied by a natural number, which is not limited by the embodiments of the present disclosure.

The process of calculating the degree of matching between the item information and the item demand information of the plurality of accounts by the server is described below.

In some embodiments, in a case where the item information of the item and the item demand information of the account are both embodied by keywords, for any one account, the server obtains, based on text similarity between a plurality of first keywords included in the item information of each item and a plurality of second keywords included in the item demand information of the account, a degree of matching between the item information of each item and the item demand information of the account. For example, the item information of item a includes a plurality of first keywords: blowers, household appliances, prices 50, inventory 60, and marketing campaigns are large coupons; the item demand information for account a includes a plurality of second keywords: the electronic products, the price interval 50-55, the products released by the common purchase official living broadcast room, the products released by the common purchase of the main broadcasting with the amount of vermicelli larger than 1 ten thousand and the common use of the large coupon. The server calculates the text similarity between the plurality of first keywords and the plurality of second keywords to be 50%, and correspondingly, the matching degree between the article a and the account a to be 50%.

In some embodiments, in a case where the item information of the item and the item demand information of the account are both embodied by keywords, for any one account, the server obtains the degree of matching between the item information of each item and the item demand information of the account based on the number of keywords matched between the plurality of first keywords included in the item information of each item and the plurality of second keywords included in the item demand information of the account. For example, the item information of item a includes a plurality of first keywords: blowers, household appliances, prices 50, inventory 60, and marketing campaigns are large coupons; the item demand information for account a includes a plurality of second keywords: the electronic products, the price interval 50-55, the products released by the common purchase official living broadcast room, the products released by the common purchase of the main broadcasting with the amount of vermicelli larger than 1 ten thousand and the common use of the large coupon. The number of keywords matched between the first keywords and the second keywords is 3, and then the matching degree between the article A and the account A is 3.

In some embodiments, where the item information for the item and the item demand information for the account are embodied by tags, for any one account, the server obtains a degree of matching between the item information for each item and the item demand information for the account based on a degree of tag similarity between the item information for each item and the item demand information for the account. For example, the article information of the article a is represented by the tag 01a5060yhq, the article demand information of the account a is represented by the tag a50GFdVyhq, and the server calculates the similarity of the tags between the two tags (for example, the number of the same characters is a percentage of the total characters), so as to obtain the matching degree between the article information of the article a and the article demand information of the account a is 60%.

It should be noted that the above expression forms of the several matching degrees and the process of calculating the matching degrees are merely illustrative, and in some embodiments, the server may also apply other calculation manners to calculate the matching degrees between the item information and the item requirement information of the multiple accounts, which is not limited in this embodiment of the disclosure.

In the following, based on the above-described process of calculating the matching degree, several cases in which the server determines at least one target account from the plurality of accounts will be described.

The first, server determines a target account based on the probability that the account will transfer resources within the target living room based on any items.

That is, for any account, when the probability that the account performs resource transfer based on any item within the target living room satisfies the requirement, the account is determined as the target account. In other words, taking an item as an example, when the account has a probability of purchasing an item within the target living room of 80% (the probability meets the requirements), the account is determined to be the target account.

Illustratively, this process includes the following two steps:

step one, a server obtains first reference information of each account based on matching degree between the article information and article demand information of each account.

Wherein the first reference information indicates a probability of the account transferring resources based on the item within the target live room. For any account, the server calculates an average value of a plurality of matching degrees based on the matching degree between the item information of each item in the target living broadcast room and the item demand information of the account, and obtains first reference information of the account. For example, in the case where any account is taken as an example, and 3 items are included in the target living broadcast room, if the matching degree between the item information of the 3 items and the item demand information of the account is 50%, 60% and 90%, respectively, the first reference information of the account is 67%, which means that the probability of resource transfer of the account based on any item in the target living broadcast room is 67%.

And step two, the server determines an account, of the plurality of accounts, of which the first reference information accords with a first target condition as the target account.

The first target condition is a preset condition, for example, the first target condition is that the first reference information is greater than or equal to 60%. It should be noted that, the specific content of the first target condition may be adjusted according to the requirement in practical application, which is not limited in the embodiments of the present disclosure.

Second, the server determines a target account based on the probability that the account will transfer resources within the target living room based on each item.

That is, for any account, when the probability that the account performs resource transfer based on an item within the target living room satisfies the requirement, the account is determined as the target account. In other words, taking an item as an example, when the account has a probability of purchasing item a of 80% (the probability meets the requirement) within the target living room, the account is determined to be the target account. Illustratively, this process includes the following two steps:

step one, the server obtains second reference information of each account based on the matching degree between the article information and the article demand information of each account.

Wherein the second reference information indicates a probability of the account transferring resources on a per item basis within the target living room. For any account, the server obtains second reference information of the account based on the matching degree between the item information of each item in the target living broadcast room and the item demand information of the account. For example, in the case where the target living room includes 3 items A, B, C, if the matching degrees between the item information of the 3 items and the item demand information of the account are 50%, 60% and 90%, respectively, the second reference information of the account is item a50%, item B60% and item C90%.

And step two, the server determines an account, of the plurality of accounts, of which the second reference information meets a second target condition as a target account.

The second target condition is a preset condition, for example, the second target condition is that the matching degree corresponding to any article in the second reference information is greater than or equal to 60%. It should be noted that, the specific content of the second target condition can be adjusted according to the requirement in practical application, which is not limited in the embodiments of the present disclosure.

In step 306, the server recommends a target live room to at least one target account.

In the embodiment of the disclosure, the server recommends the target live broadcast room to the terminal where the at least one target account is located. Because the at least one target account is determined according to the article information of articles in the target living broadcast room and the article demand information of a plurality of accounts, after the target living broadcast room is recommended by the target accounts, the target accounts are likely to transfer resources based on the articles in the target living broadcast room, so that the accuracy of living broadcast recommendation is improved, and the conversion rate of the accounts in the target living broadcast room is further effectively improved.

In some embodiments, the server recommends the target live room to the at least one target account prior to a live start time of the target live room. By the method, the target live broadcasting room can be timely recommended to the target account before the target live broadcasting room starts to live broadcasting, so that the target account is guided to reserve the target live broadcasting room or timely watch the target live broadcasting room when the target live broadcasting room starts to live broadcasting, and more accounts are attracted to enter the target live broadcasting room. In other embodiments, the server recommends the target live room to the at least one target account after a live start time of the target live room. By the method, more accounts can be purposefully attracted to enter the target live broadcasting room in the live broadcasting process of the target live broadcasting room, the accuracy of live broadcasting recommendation is improved, and the conversion rate of the accounts in the target live broadcasting room is further improved. Of course, the server can also recommend the target live room to at least one target account before and after the live start time of the target live room, so that the account is attracted to enter the live room to the maximum extent, the accuracy of live recommendation is greatly improved, and the conversion rate of the account in the target live room is effectively improved.

In some embodiments, the server sends a first open play cue to the terminal where the at least one target account is located, the first open play cue indicating open play information of the target live room. The opening information of the target living room indicates a living title, main broadcasting information, related information of the released objects, and opening time of the target living room, which is not limited in the embodiment of the present disclosure. In some embodiments, the server sends the first multicast prompt to the terminal where the at least one target account is located in the form of a sms, a smart phone, an application message push, a hosting page presentation, and the like. Schematically, taking a short message as an example, after receiving the first play prompt, the terminal of the target account displays the play information of the target live broadcasting room in the form of the short message (for example, the live broadcasting room XX is in the tonight XX play, and a large number of electric appliances are seen by you). It should be noted that, the form of the first play prompt may be adjusted according to actual needs, for example, taking the target platform to provide multiple functions including live broadcasting, short video browsing, article reading, and the like, and the server sends the first play prompt to the terminal where at least one target account is located, so that the terminal where the target account is located displays the play information of the target live broadcasting room in each functional area of the target platform, which is not limited in the embodiments of the present disclosure. By the method, the target account can be timely prompted to watch the target live broadcasting room, so that the purpose of the account in the live broadcasting room after the target live broadcasting room is opened is stronger, the accuracy of live broadcasting recommendation is further improved, and the conversion rate of the account in the live broadcasting room and the time for the account to watch the live broadcasting room are further improved.

In some embodiments, the server sends a second play prompt to the terminal where the at least one target account is located, the second play prompt indicating play information for the target item within the target live room. Wherein, the opening information of the target object in the target living broadcast room indicates the living broadcast title, the main broadcast information, the related information of the released object, the opening time and the like of the target living broadcast room, which is not limited by the embodiment of the disclosure. The target item refers to an item corresponding to the item demand of the target account. Illustratively, the server determines a target item corresponding to the item demand for each target account from at least one item in the target living room based on the item demand information for each target account. For example, taking any one of the target accounts as an example, the item demand information of the target account indicates that the target account purchases an appliance item frequently, indicates that the item demand of the target account is the appliance item, and determines a blower in at least one item of the target living room as the target item. In addition, the second open-cast cue is the same as the first open-cast cue described above, and the embodiment of the present disclosure is not limited thereto. Schematically, taking a short message as an example, after receiving the second play prompt, the terminal of the target account displays the play information of the target object in the target living room in the form of the short message (for example, the living room XX plays in the tonight XX, and you like a blower, etc.). It should be noted that, the number of the target items is not limited in the embodiments of the present disclosure, and when the item requirement of the target account corresponds to a plurality of items in the living room, the plurality of items are all determined as the target items. By the method, the target account can be prompted to view the target live broadcasting room in a targeted manner based on the object demand of the target account, the purpose of the account in the live broadcasting room after the target live broadcasting room is opened is further improved, the accuracy of live broadcasting recommendation is improved, and the conversion rate of the account in the live broadcasting room and the time for viewing the live broadcasting room by the account are further effectively improved.

Through the steps 301 to 306, the server obtains the item information of the items in the target living broadcast room from the living broadcast plan side of the target living broadcast room, so as to sense the release plan of the items in the living broadcast room (such as which items are released in the living broadcast room and which marketing activities are available in the living broadcast room) in advance before the living broadcast starts, obtain the item demand information of a plurality of accounts from the image data side of the accounts, obtain the basic information of the items required by the accounts and the demand information of different resource transfer forms (such as which items the accounts want to purchase and which marketing activities tend to purchase the items through), after obtaining the two types of information, screen out target accounts which are likely to transfer resources based on the items in the target living broadcast room from the plurality of accounts through the matching degree between the two types of information, and recommend the target living broadcast room to the target accounts, thereby improving the accuracy of living broadcast recommendation, and further improving the conversion rate of the accounts in the living broadcast room and the time for watching living broadcast room.

The live broadcast recommendation method is schematically described below with reference to fig. 4 and 5. Fig. 4 is a schematic diagram of a live broadcast program according to an exemplary embodiment, and fig. 5 is a schematic diagram of a live broadcast recommendation method according to an exemplary embodiment. As shown in fig. 4, before the target living room is living, the first account can generate a living program of the target living room by selecting a living program template, creating a living program, adding objects to the program and the like, so as to promote and guide, and in the living process, the objects are released according to the living program by selecting the created living program, after living, living data are analyzed, living program targets are deposited, so that the living program template and the like are further optimized. It should be noted that the illustration in fig. 4 is merely schematic, and in some embodiments, the target live room live schedule can be created by other ways, which is not limited by the embodiments of the present disclosure. Based on the live broadcast plan shown in fig. 4, the live broadcast recommendation method provided by the embodiment of the present disclosure is used to recommend a target live broadcast room, and this process is shown in fig. 5, where a server obtains item information of items in the target live broadcast room (such as what items are distributed in the live broadcast room and what marketing activities are available in the live broadcast room, etc.) from a live broadcast plan side, obtains item demand information of a plurality of accounts (such as what items the accounts want to purchase and what marketing activities tend to purchase, etc.) from an image data side of the accounts, then determines at least one target account from the plurality of accounts according to the matching degree between the two types of information, recommends the target live broadcast room to the at least one target account in various forms (such as main broadcast account homepage recommendation, platform inner/outer push and short video recommendation, etc., which are not limited), and by this live broadcast recommendation method, more accounts can be targeted and targeted to be drained, thereby attracting more accounts into the live broadcast room, improving the accuracy of live broadcast recommendation, and further improving the time length of the live broadcast account viewing room and conversion rate in the live broadcast room.

In summary, in the live broadcast recommendation method provided by the embodiment of the disclosure, after item information of an item in a live broadcast room and item demand information of a plurality of accounts are obtained according to a live broadcast plan of the live broadcast room and image data of the plurality of accounts, at least one target account is determined from the plurality of accounts based on matching degree between the two types of information, and the live broadcast room is recommended to the at least one target account. In the process, the release plan of the articles in the living broadcast room is obtained in advance according to the living broadcast plan, so that the article demand information of a plurality of accounts can be combined, target accounts which are likely to transfer resources based on the articles in the living broadcast room can be screened out from the plurality of accounts, the living broadcast room is recommended to the target accounts in a targeted and targeted manner, the accuracy of living broadcast recommendation is improved, and the conversion rate of the accounts in the living broadcast room is further effectively improved.

Fig. 6 is a block diagram illustrating a live recommender in accordance with an exemplary embodiment. Referring to fig. 6, the apparatus includes a first determining unit 601, a second determining unit 602, and a recommending unit 603.

A first determining unit 601 configured to perform a live broadcast program based on a target live broadcast room, determining item information of an item in the target live broadcast room, the live broadcast program indicating a distribution program of the item in the target live broadcast room;

A second determining unit 602 configured to perform determining item demand information for a plurality of accounts based on the portrait data of the plurality of accounts, the item demand information indicating demand information for items by the accounts;

and a recommending unit 603 configured to perform a recommendation of the target live room to at least one target account from among the plurality of accounts based on the matching degree between the item information and the item demand information of the plurality of accounts.

In some embodiments, the first determining unit 601 is configured to perform:

In some embodiments, the second determining unit 602 is configured to perform:

In some embodiments, the recommending unit 603 is configured to perform:

In some embodiments, the apparatus further comprises:

the recommending unit 603 is configured to perform:

It should be noted that: in the live broadcast recommendation apparatus provided in the foregoing embodiment, only the division of the functional units is used for illustration, and in practical application, the allocation of the functions may be performed by different functional units according to needs, that is, the internal structure of the device is divided into different functional units, so as to complete all or part of the functions described above. In addition, the live broadcast recommending device and the live broadcast recommending method provided in the foregoing embodiments belong to the same concept, and specific implementation processes of the live broadcast recommending device and the live broadcast recommending method are detailed in the method embodiments and are not repeated here.

The embodiment of the disclosure also provides an electronic device, which comprises:

one or more processors;

a memory for storing the processor-executable program code;

wherein the processor is configured to execute the program code to implement a procedure executed by the server in the live recommendation method provided in the foregoing method embodiments.

In some embodiments, the program code according to the embodiments of the present application may be deployed to be executed on one electronic device or on a plurality of electronic devices located at one site, and a plurality of electronic devices distributed across a plurality of sites and interconnected by a wired network or a wireless network may constitute a blockchain system.

Taking an electronic device as a server as an example, fig. 7 is a block diagram of a server according to an exemplary embodiment, where the server 700 may have a relatively large difference due to configuration or performance, and may include one or more processors (Cent ral Processing Units, CPU) 701 and one or more memories 702, where the one or more memories 902 store at least one program code, and the at least one program code is loaded and executed by the one or more processors 701 to implement a procedure performed by the server in the live broadcast recommendation method provided in the foregoing method embodiments. Of course, the server 700 may also have a wired or wireless network interface, a keyboard, an input/output interface, and other components for implementing the functions of the device, which are not described herein.

In an exemplary embodiment, a computer readable storage medium is also provided, e.g. a memory 702 comprising program code, which is executable by the processor 701 of the server 700 to perform the live recommendation method described above. Alternatively, the computer readable storage medium may be a read-only memory (ROM), a random access memory (random access memory, RAM), a compact-disk read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

In an exemplary embodiment, a computer program product is also provided, comprising one or more instructions that are executable by one or more processors of an electronic device, to enable the electronic device to perform the live recommendation method described above.

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any adaptations, uses, or adaptations of the disclosure following the general principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It is to be understood that the present disclosure is not limited to the precise arrangements and instrumentalities shown in the drawings, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A live recommendation method, applied to a server, comprising:

when a live broadcast plan issuing request of a target live broadcast room is received, acquiring a live broadcast plan of the target live broadcast room, determining article information of articles in the target live broadcast room based on the live broadcast plan, wherein the article information indicates basic information of the articles and a resource transfer form for resource transfer based on the articles, the live broadcast plan indicates an issuing plan of the articles in the target live broadcast room, and the resource transfer form indicates a transaction form and a marketing activity form of the articles;

determining article demand information of a plurality of accounts based on portrait data of the accounts, wherein the article demand information indicates basic information of articles required by the accounts and demand information of the accounts for different resource transfer forms, and the accounts are all accounts logged in on a target platform or are accounts for starting a live broadcast recommendation function;

Calculating an average value of a plurality of matching degrees based on the matching degree between the item information and the item demand information of each account to obtain first reference information of the account, wherein the first reference information indicates the probability that the account performs resource transfer based on the item in the target living broadcast room, the account, of which the first reference information meets a first target condition, is determined to be at least one target account, the target living broadcast room is recommended to the at least one target account before the living broadcast starting time of the target living broadcast room, and/or the target living broadcast room is recommended to the at least one target account after the living broadcast starting time of the target living broadcast room.

2. The live recommendation method according to claim 1, wherein the determining item information of items in the target live room based on the live program of the target live room comprises:

3. The live recommendation method according to claim 1, wherein the determining the item demand information of the plurality of accounts based on the portrait data of the plurality of accounts includes:

4. The live recommendation method according to claim 1, further comprising:

and determining an account, of the plurality of accounts, in which the second reference information meets a second target condition as the target account.

5. The live recommendation method of claim 1, wherein the recommending the target live room to the at least one target account comprises:

And sending a first broadcasting prompt to the terminal where the at least one target account is located, wherein the first broadcasting prompt indicates the broadcasting information of the target live broadcasting room.

6. The live recommendation method according to claim 1, further comprising:

determining a target item corresponding to the item demand of each target account from at least one item in the target living room based on the item demand information of each target account;

the recommending the target live room to the at least one target account comprises:

and sending a second broadcasting prompt to the terminal where the at least one target account is located, wherein the second broadcasting prompt indicates the broadcasting information of the target object in the target live broadcasting room.

7. A live recommendation device, configured in a server, the device comprising:

a first determining unit configured to execute, when a live broadcast plan issue request of a target live broadcast room is received, acquiring a live broadcast plan of the target live broadcast room, determining item information of an item in the target live broadcast room based on the live broadcast plan, the item information indicating basic information of the item and a resource transfer form for resource transfer based on the item, the live broadcast plan indicating an issue plan of the item in the target live broadcast room, the resource transfer form indicating a transaction form and a marketing activity form of the item;

The second determining unit is configured to determine article demand information of a plurality of accounts based on portrait data of the accounts, wherein the article demand information indicates basic information of articles required by the accounts and demand information of the accounts for different resource transfer forms, the accounts are all accounts logged in on a target platform, or the accounts are accounts for starting a live broadcast recommending function;

and the recommending unit is configured to execute matching degree between the article information and the article demand information of each account, calculate an average value of a plurality of matching degrees, obtain first reference information of the accounts, the first reference information indicates the probability of the accounts performing resource transfer based on articles in the target living broadcast room, determine the accounts, of the plurality of accounts, that the first reference information meets a first target condition as at least one target account, recommend the target living broadcast room to the at least one target account before the living broadcast starting time of the target living broadcast room, and/or recommend the target living broadcast room to the at least one target account after the living broadcast starting time of the target living broadcast room.

8. The live recommendation device of claim 7, wherein the first determination unit is configured to perform:

9. The live recommendation device according to claim 7, wherein the second determination unit is configured to perform:

10. The live recommendation device of claim 7, wherein the recommendation unit is configured to perform:

11. The live recommendation device of claim 7, wherein the recommendation unit is configured to perform:

12. The live recommendation device of claim 7, wherein the device further comprises:

a third determining unit configured to perform determining, based on the item demand information of each of the target accounts, a target item corresponding to an item demand of each of the target accounts from at least one item of the target living room;

the recommendation unit is configured to perform:

13. An electronic device, the electronic device comprising:

one or more processors;

a memory for storing the processor-executable program code;

Wherein the processor is configured to execute the program code to implement the live recommendation method of any one of claims 1 to 6.

14. A computer readable storage medium, characterized in that program code in the computer readable storage medium, when executed by a processor of an electronic device, enables the electronic device to perform the live recommendation method of any one of claims 1 to 6.