CN110012364B

CN110012364B - Anchor recommendation method and device and storage medium

Info

Publication number: CN110012364B
Application number: CN201910301118.0A
Authority: CN
Inventors: 林健勇
Original assignee: Guangzhou Cubesili Information Technology Co Ltd
Current assignee: Guangzhou Jinhong network media Co.,Ltd.
Priority date: 2019-04-15
Filing date: 2019-04-15
Publication date: 2021-06-01
Anticipated expiration: 2039-04-15
Also published as: CN110012364A

Abstract

The application discloses a method and a device for recommending a main broadcast, and belongs to the technical field of live broadcast. The method comprises the following steps: determining an online average value according to the total number of the current online anchor and the current multiple highest simultaneous online user number PCUs, wherein the PCUs are PCUs of a live broadcast room where the multiple online anchors are located; dividing a PCU list according to the online average value to obtain a plurality of PCUs (Power control units) grouped, wherein the PCU list comprises the PCUs which are arranged according to the numerical value; dividing a plurality of anchor identifications included in an anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, wherein the anchor list is obtained by sequencing according to historical live broadcast quality scores of the online anchors; and sending the anchor recommendation information to the terminal according to the plurality of sub-lists. The anchor recommendation method provided by the application can increase the exposure rate of the excellent new anchor.

Description

Anchor recommendation method and device and storage medium

Technical Field

The present application relates to the field of live broadcast technologies, and in particular, to a method and an apparatus for recommending a anchor and a storage medium.

Background

Currently, to attract more users to watch live broadcasts, the server may recommend different anchor broadcasts to the users. Generally, the server may sort, according to a sequence from large to small, the anchor identifiers of all current online anchor according to a PCU (peak current users, the highest number of simultaneous online users) or an ACU (average current users, the average number of simultaneous online users) of a live broadcast in which the current online anchor is located, and then push anchor recommendation information to the terminal according to a sorting result when receiving an anchor recommendation request sent by the terminal.

It can be seen that, in the related art, when recommending the main broadcast to the terminal based on the PCU and the ACU, the higher the PCU or the ACU of the on-line main broadcast is, the easier it is to recommend. That is, the higher the popularity of the online anchor, the greater the probability of being recommended. However, when the popularity of some excellent new anchor is low, since the anchor is relatively backward in the ranking result, the recommended probability is relatively low, and the exposure rate is relatively low, which is not beneficial to the platform mining potential anchor.

Disclosure of Invention

The embodiment of the application provides a method and a device for recommending a main broadcast, which can solve the problem that the exposure rate of an excellent new main broadcast is low in the related art. The technical scheme is as follows:

in a first aspect, a method for anchor recommendation is provided, the method comprising:

determining an online average value according to the total number of the current online anchor and the current multiple highest simultaneous online user number PCUs, wherein the PCUs are PCUs of a live broadcast room where the multiple online anchors are located;

dividing a PCU list according to the online average value to obtain a plurality of PCUs (Power control units) grouped, wherein the PCU list comprises the PCUs which are arranged according to the numerical value;

dividing a plurality of anchor identifications included in an anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, wherein the anchor list is obtained by sequencing according to historical live broadcast quality scores of the online anchors;

and sending the anchor recommendation information to the terminal according to the plurality of sub-lists.

Optionally, the determining an online average value according to the current online anchor total number and the current multiple highest simultaneous online user numbers PCU includes:

acquiring a preset adjustment threshold;

determining the online average value according to the current online anchor total, the PCUs and the adjustment threshold value through the following model;

wherein P is the online average value, N is the current online anchor total number, M is the adjustment threshold, and the PCU_iIs the i-th PCU among the plurality of PCUs.

Optionally, the PCUs included in the PCU list are arranged in order from large to small;

the dividing the PCU list according to the online average value to obtain a plurality of PCU groups includes:

let i be 1, judge whether the ith PCU in the PCU list is smaller than the said online average value;

and if the ith PCU in the PCU list is not smaller than the online average value, grouping the ith PCU as a PCU, enabling the i to be i +1, and returning to the step of judging whether the ith PCU in the PCU list is smaller than the online average value until the i is equal to the current online anchor total number.

Optionally, after determining whether the ith PCU in the PCU list is smaller than the online average value, the method further includes:

if the ith PCU in the PCU list is smaller than the online average value and the i is smaller than the online anchor total number, enabling j to be 1, and judging whether the sum of the ith PCU and the (i + j) th PCU is smaller than the online average value or not, wherein the i + j is not larger than the online anchor total number;

if the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value, the j is made to be j +1, the step of judging whether the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value is returned, the step of judging whether the ith PCU to the (i + j) th PCU is smaller than the online average value is returned until the sum of the ith PCU to the (i + j) th PCU is not smaller than the online average value, the step of judging whether the ith PCU in the PCU list is smaller than the online average value is returned until the sum of the ith PCU to the (i + j) th PCU is equal to the current online anchor total number, or the step of grouping the ith PCU to the (i + j) th PCU.

Optionally, before the dividing the plurality of anchor identifiers included in the anchor list according to the number of PCUs in each PCU packet in the plurality of PCU packets to obtain a plurality of sub-lists, the method further includes:

obtaining parameter values of a plurality of live broadcast parameters of each online anchor in a first time length before the current time, wherein the plurality of live broadcast parameters comprise at least one of video resolution, gift sending value, barrage number, sharing number and user watching time length of a live broadcast room in which the corresponding online anchor is located;

acquiring a historical score corresponding to each live broadcast parameter in a plurality of live broadcast parameters of each online anchor;

and determining the historical live broadcast quality score of each online anchor according to the parameter values of the live broadcast parameters of each online anchor and the historical score corresponding to each live broadcast parameter in the live broadcast parameters.

Optionally, the PCU groups are arranged in a sequence obtained by dividing, and a plurality of anchor identifiers included in the anchor list are arranged in a sequence from high to low according to a historical live broadcast quality score of an online anchor corresponding to each anchor identifier;

the dividing a plurality of anchor identifiers included in an anchor list according to the number of PCUs in each PCU group in the plurality of PCU groups to obtain a plurality of sub-lists comprises:

let k equal to 1, according to the number D of PCUs in the kth PCU group in the plurality of PCU groups_kObtaining a front D from the anchor list_kAn individual anchor identification;

front D to be acquired_kThe individual anchor mark is used as the kth sub-list, and the front D in the anchor list is deleted_kAnd the anchor mark is used for enabling k to be k +1 and returning to the PCU group according to the number D of PCUs in the kth PCU group in the PCU groups_kObtaining a front D from the anchor list_kA step of anchor identification until said k equals the number of said plurality of PCU packets.

Optionally, the sending, according to the plurality of sub-lists, anchor recommendation information to the terminal includes:

receiving anchor recommendation requests sent by a terminal, wherein the anchor recommendation requests carry request quantity, and the request quantity refers to the quantity of anchors requested to be recommended;

according to the anchor recommendation request, acquiring anchor identifications with the number equal to that of the request from the plurality of sub-lists;

and sending the anchor recommendation information corresponding to the obtained anchor identification to the terminal.

Optionally, the plurality of sub-lists are arranged according to a sequence obtained by dividing, and the anchor recommendation request also carries a user identifier of the terminal;

the obtaining, according to the anchor recommendation request, anchor identifiers equal in number to the request number from the plurality of sub-lists includes:

if the anchor recommendation request is determined to be the first anchor recommendation request sent by the terminal according to the user identification of the terminal, or if the anchor recommendation request is determined not to be the first anchor recommendation request of the terminal according to the user identification of the terminal and the difference between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is greater than a second duration, sequentially acquiring anchor identifications with the number equal to the number of the requests from the first sub-list of the plurality of sub-lists.

Optionally, the plurality of sub-lists are arranged according to the sequence obtained by dividing;

if it is determined according to the user identifier of the terminal that the anchor recommendation request is not the first anchor recommendation request of the terminal and the difference between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is not greater than a second duration, obtaining historical output data corresponding to the terminal, wherein the historical output data comprises anchor identifiers sent to the terminal according to other anchor recommendation requests sent by the terminal before the anchor recommendation request;

and sequentially acquiring the anchor identifications with the number equal to the number of the requests from the other anchor identifications except the historical output data in the plurality of sub-lists according to the arrangement sequence of the plurality of sub-lists and the arrangement sequence of the anchor identifications included in the plurality of sub-lists.

Optionally, the sending the anchor recommendation information corresponding to the obtained anchor identifier to the terminal includes:

randomly sequencing the anchor identifications belonging to the same sub-list in the sequentially acquired anchor identifications to obtain an anchor recommendation list;

and sending the anchor recommendation information corresponding to each anchor identification to the terminal according to the sequence of the plurality of anchor identifications in the anchor recommendation list.

In a second aspect, there is provided an anchor recommendation apparatus, the apparatus comprising:

the system comprises a first determining module, a second determining module and a third determining module, wherein the first determining module is used for determining an online average value according to the total number of current online anchor and the current multiple highest simultaneous online user number PCUs, and the multiple PCUs are PCUs of a live broadcast room where the multiple online anchors are located;

a first dividing module, configured to divide a PCU list according to the online average value to obtain a plurality of PCU groups, where the PCU list includes the PCUs and the PCUs are arranged according to a numerical value;

a second dividing module, configured to divide a plurality of anchor identifiers included in an anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, where the anchor list is obtained by sorting according to historical live broadcast quality scores of the online anchors;

and the sending module is used for sending the anchor recommendation information to the terminal according to the plurality of sub-lists.

Optionally, the first determining module is configured to:

acquiring a preset adjustment threshold;

the first dividing module is configured to:

Optionally, the first dividing module is further configured to:

Optionally, the apparatus further comprises:

the system comprises a first acquisition module, a second acquisition module and a display module, wherein the first acquisition module is used for acquiring parameter values of a plurality of live broadcast parameters of each online anchor in a first time length before the current time, and the plurality of live broadcast parameters comprise at least one of video resolution, gift sending value, barrage number, sharing number and user watching time length in a live broadcast room where the corresponding online anchor is located;

the second acquisition module is used for acquiring the historical score corresponding to each live broadcast parameter in the plurality of live broadcast parameters of each online anchor;

and the second determining module is used for determining the historical live broadcast quality score of each online anchor according to the parameter values of the live broadcast parameters of each online anchor and the historical score corresponding to each live broadcast parameter in the live broadcast parameters.

the second dividing module is configured to:

Optionally, the sending module includes:

the receiving submodule is used for receiving anchor recommendation requests sent by a terminal, wherein the anchor recommendation requests carry request quantity, and the request quantity refers to the quantity of anchors requested to be recommended;

the obtaining sub-module is used for obtaining anchor identifications with the number equal to that of the requests from the plurality of sub-lists according to the anchor recommendation requests;

and the sending submodule is used for sending the anchor recommendation information corresponding to the obtained anchor identification to the terminal.

the acquisition submodule is configured to:

the obtaining sub-module is further configured to:

Optionally, the sending sub-module is configured to:

In a third aspect, an anchor recommendation apparatus is provided, the apparatus comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to perform the steps of the anchor recommendation method of the first aspect.

In a fourth aspect, a computer-readable storage medium is provided, having instructions stored thereon, which when executed by a processor, implement the steps of the anchor recommendation method of the first aspect described above.

The technical scheme provided by the embodiment of the application can at least bring the following beneficial effects:

according to the method and the device, an online average value can be determined according to the total number of the current online anchor and the PCUs in the live broadcast room where the online anchors are located, then the PCU list is divided according to the online average value to obtain a plurality of PCU groups, then the anchor marks included in the anchor list are divided according to the number of the PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, and finally anchor recommendation information is sent to the terminal according to the sub-lists. The anchor list is obtained by sequencing according to the historical live broadcast quality scores of the online anchors, so that the excellence degree of the anchors can be reflected, that is, the anchor recommendation information can be sent to the terminal according to the excellence degree of the anchors, the anchor with high popularity is prevented from being recommended preferentially all the time, and the exposure rate of the excellent new anchor is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application, 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 structural diagram of an anchor recommendation system provided in an embodiment of the present application;

fig. 2 is a flowchart of an anchor recommendation method provided in an embodiment of the present application;

FIG. 3 is a flowchart of another anchor recommendation method provided by an embodiment of the present application;

fig. 4 is a schematic structural diagram of an anchor recommendation apparatus according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of another anchor recommendation device according to an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

Before explaining the embodiments of the present application in detail, an application scenario and a system architecture of the embodiments of the present application are introduced.

Currently, live broadcasts such as show live, game live, sports live, star show live, social live, and the like are widely known. When a user wants to watch a live broadcast, a live application installed on the terminal may be opened. When detecting a start instruction of the installed live broadcast application program, the terminal can generate a main broadcast recommendation request and send the main broadcast recommendation request to the server. Correspondingly, the server can receive the anchor recommendation request sent by the terminal and push the anchor recommendation information to the terminal according to the received anchor recommendation request. In the related art, when the anchor recommendation information is pushed to the terminal, all anchor identifications of the current online anchor are sorted in descending order according to the PCU or ACU of the live broadcast room where the current online anchor is located, and then the anchor recommendation information is pushed to the terminal according to the sorting. In this way, the anchor recommendation information corresponding to the anchor identifier ranked at the top is always preferentially recommended, and the anchor recommendation information corresponding to the anchor identifier ranked at the bottom is not easily recommended, that is, the probability that the online anchor with high popularity is recommended is higher than the probability that the online anchor with low popularity is recommended. Therefore, the recommended probability of some excellent new anchor with lower popularity is lower, the exposure rate is lower, and the platform excavation potential anchor is not facilitated. Based on this, the embodiment of the application discloses a main broadcast recommendation method, which considers two factors of main broadcast popularity and main broadcast excellence degree, can avoid recommending main broadcasts with higher popularity preferentially all the time, and improves the exposure rate of new excellent main broadcasts.

Fig. 1 is a schematic structural diagram of an anchor recommendation system 100 according to an embodiment of the present application. Referring to fig. 1, the anchor recommendation system includes: a terminal 101 and a server 102. The number of the terminals 101 may be multiple, and the server 102 establishes a communication connection with each terminal 101.

The terminal 101 may send an anchor recommendation request to the server, and accordingly, the server 102 may receive the anchor recommendation request sent by the terminal 101. When the server 102 receives the anchor recommendation request, the anchor recommendation information may be sent to the terminal by using the anchor recommendation method provided in the embodiment of the present application. When the terminal 101 receives the anchor recommendation information transmitted by the server 102, the anchor recommendation information may be displayed on a display interface of the live application.

It should be noted that the terminal 101 may be an intelligent terminal such as a smart phone or a tablet computer with a live application installed therein, the server 102 may be an application server of the live application installed on the terminal 101, and the server 102 may be an independent server or a server cluster, which is not limited in this embodiment of the present application.

Fig. 2 is a flowchart of an anchor recommendation method provided in an embodiment of the present application, where the method may be applied to a server. Referring to fig. 2, the method includes:

step 201: and determining an online average value according to the total number of the current online anchor and a plurality of current PCUs (Power control units), wherein the PCUs refer to PCUs of the live broadcasting rooms where the online anchors are located.

Step 202: and dividing a PCU list according to the online average value to obtain a plurality of PCUs, wherein the PCU list comprises the PCUs which are arranged according to the numerical value.

Step 203: and dividing a plurality of anchor identifications included in an anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, wherein the anchor list is obtained by sequencing according to historical live broadcast quality scores of the online anchors.

Step 204: and sending the anchor recommendation information to the terminal according to the plurality of sub-lists.

In the embodiment of the application, an online average value is determined according to the total number of the current online anchor and a plurality of PCUs in a live broadcast room where a plurality of online anchors are located, then a PCU list is divided according to the online average value to obtain a plurality of PCU groups, a plurality of anchor marks included in the anchor list are divided according to the number of PCUs in each PCU group in the plurality of PCU groups to obtain a plurality of sub-lists, and finally anchor recommendation information is sent to a terminal according to the plurality of sub-lists. The anchor list is obtained by sequencing according to the historical live broadcast quality scores of the online anchors, so that the excellence degree of the anchors can be reflected, that is, the anchor recommendation information can be sent to the terminal according to the excellence degree of the anchors, the anchor with high popularity is prevented from being recommended preferentially all the time, and the exposure rate of the excellent new anchor is improved.

Optionally, determining an online average value according to the current online anchor total number and the current multiple highest simultaneous online user numbers PCU, includes:

acquiring a preset adjustment threshold;

determining an online average value through the following model according to the current online anchor total number, the PCUs and an adjustment threshold;

wherein P is the on-line average value, N is the current on-line anchor total number, M is the adjustment threshold, PCU_iIs the i-th PCU among the plurality of PCUs.

dividing the PCU list according to the online average value to obtain a plurality of PCU groups, wherein the PCU groups comprise:

let i be 1, judge whether the ith PCU in the PCU list is smaller than the online average value;

and if the ith PCU in the PCU list is not smaller than the online average value, grouping the ith PCU as one PCU, enabling i to be i +1, returning to the step of judging whether the ith PCU in the PCU list is smaller than the online average value or not, and till i is equal to the current online anchor total number.

if the ith PCU in the PCU list is smaller than the online average value and i is smaller than the online anchor total number, making j equal to 1, and judging whether the sum of the ith PCU and the (i + j) th PCU is smaller than the online average value or not, wherein i + j is not larger than the online anchor total number;

if the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value, j is made to be j +1, the step of judging whether the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value is returned, until the sum of the ith PCU to the (i + j) th PCU is not smaller than the online average value, the ith PCU to the (i + j) th PCU are taken as a PCU group, i is made to be i + j +1, the step of judging whether the ith PCU in a PCU list is smaller than the online average value is returned, or until i + j is equal to the current online anchor total number, the ith PCU to the (i + j) th PCU are taken as a PCU group.

Optionally, before the dividing the multicast identifier included in the multicast list according to the number of PCUs in each PCU packet in the plurality of PCU packets to obtain a plurality of sub-lists, the method further includes:

dividing a plurality of anchor identifiers included in an anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists, wherein the sub-lists comprise:

let k equal to 1, and according to the number D of PCUs in the kth PCU group in the plurality of PCU groups_kObtaining front D from the Anchor list_kAn individual anchor identification;

front D to be acquired_kThe individual anchor mark is used as the kth sub-list, and the front D in the anchor list is deleted_kThe number of PCUs in the kth PCU group in the plurality of PCU groups is returned according to the anchor mark, k is k +1_kObtaining front D from the Anchor list_kA step of anchor identification until k equals the number of PCU packets.

Optionally, sending anchor recommendation information to the terminal according to the plurality of sub-lists, where the sending includes:

receiving anchor recommendation requests sent by a terminal, wherein the anchor recommendation requests carry request quantity, and the request quantity refers to the quantity of anchors requesting recommendation;

acquiring anchor identifications with the quantity equal to the request quantity from the plurality of sub-lists according to the anchor recommendation request;

Optionally, the plurality of sub-lists are arranged according to the sequence obtained by dividing, and the anchor recommendation request also carries a user identifier of the terminal;

according to the anchor recommendation request, acquiring anchor identifications with the number equal to the number of the requests from the plurality of sub-lists, and the method comprises the following steps:

if the anchor recommendation request is determined to be the first anchor recommendation request sent by the terminal according to the user identification of the terminal, or if the anchor recommendation request is determined not to be the first anchor recommendation request of the terminal according to the user identification of the terminal and the difference between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is greater than the second time length, the anchor identifications with the number equal to the number of the requests are sequentially obtained from the first sub-list of the plurality of sub-lists.

if the fact that the anchor recommendation request is not the first anchor recommendation request of the terminal and the difference value between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is not larger than the second duration is determined according to the user identification of the terminal, historical output data corresponding to the terminal are obtained, wherein the historical output data comprise anchor identifications sent to the terminal according to other anchor recommendation requests sent by the terminal before the anchor recommendation request;

and sequentially acquiring the anchor identifications with the number equal to the number of the requests from the other anchor identifications except the historical output data in the plurality of sub-lists according to the arrangement order of the plurality of sub-lists and the arrangement order of the anchor identifications included in the plurality of sub-lists.

Optionally, sending anchor recommendation information corresponding to the obtained anchor identifier to the terminal, including:

All the above optional technical solutions can be combined arbitrarily to form an optional embodiment of the present application, and the present application embodiment is not described in detail again.

For ease of understanding, the anchor recommendation method provided in the embodiment of fig. 2 is described below with reference to fig. 3. Fig. 3 is a flowchart of an anchor recommendation method provided in an embodiment of the present application, where the method may be applied to a server. Referring to fig. 3, the method includes:

step 301: and determining an online average value according to the total number of the current online anchor and a plurality of current PCUs (Power control units), wherein the PCUs refer to PCUs of the live broadcasting rooms where the online anchors are located.

It should be noted that, a higher PCU of the live broadcast room where the online anchor is located indicates a higher popularity of the online anchor, and conversely, a lower PCU of the live broadcast room where the online anchor is located indicates a lower popularity of the online anchor. In the embodiment of the application, the server can count the total number of the online anchor and the PCU of the live broadcast room where each online anchor is located in real time.

After the current online anchor total number and the current PCUs are obtained through statistics, the server can obtain a preset adjustment threshold value; determining an online average value through the following model according to the current online anchor total number, the PCUs and an adjustment threshold;

wherein P is the on-line average value, N is the current on-line anchor total number, M is the adjustment threshold, PCU_iThe i-th PCU among the plurality of PCUs.

It should be noted that, since the online average is subsequently used to divide the PCU list in step 302, in order to finely divide the PCU list and obtain as many PCU groups as possible, and not to make the number of PCUs in each PCU group too large, an adjustment threshold may be set for the PCUs. Wherein, the smaller the adjustment threshold, the more PCU groups will be divided. In addition, the adjustment threshold value can be preset and can be set according to the total number of the current online anchor, as long as the online average value determined by the model meets the requirement.

Step 302: and dividing a PCU list according to the online average value to obtain a plurality of PCUs, wherein the PCU list comprises the PCUs which are arranged according to the numerical value.

It should be noted that, in order to facilitate the partition of the PCU list, the PCUs included in the PCU list may be arranged in descending order.

Illustratively, the implementation process of this step may be: let i be 1, judge whether the ith PCU in the PCU list is smaller than the online average value; and if the ith PCU in the PCU list is not smaller than the online average value, grouping the ith PCU as one PCU, enabling i to be i +1, returning to the step of judging whether the ith PCU in the PCU list is smaller than the online average value or not, and till i is equal to the current online anchor total number. If the ith PCU in the PCU list is smaller than the online average value, making j equal to 1, and judging whether the sum of the ith PCU and the (i + j) th PCU is smaller than the online average value; if the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value, j is made to be j +1, the step of judging whether the sum of the ith PCU to the (i + j) th PCU is smaller than the online average value is returned, until the sum of the ith PCU to the (i + j) th PCU is not smaller than the online average value, the ith PCU to the (i + j) th PCU are taken as a PCU group, i is made to be i + j +1, the step of judging whether the ith PCU in a PCU list is smaller than the online average value is returned, or until i + j is equal to the current online anchor total number, the ith PCU to the (i + j) th PCU are taken as a PCU group.

The PCU group may be configured to obtain a first PCU group by grouping the first PCU as one PCU if the first PCU in the PCU list is smaller than the online average value. The server may then continue to determine whether the second PCU is less than the online average. If the second PCU is less than the online average, the server may determine the sum of the second PCU and the third PCU. If the sum of the second PCU and the third PCU is not less than the on-line average value, the server may group the second PCU and the third PCU as one PCU, thereby obtaining a second PCU group. Then, starting from the fourth PCU, it is newly determined whether the fourth PCU is smaller than the on-line average value, and the fourth PCU is processed in accordance with the determination result with reference to the aforementioned processing method for the second PCU.

If the sum of the second PCU and the third PCU is still less than the on-line average value, the server may continue to calculate the sum of the second PCU, the third PCU, and the fourth PCU, and so on, until the sum of the second PCU to the (2+ j) th PCU is not less than the on-line average value, and may group the second PCU to the (2+ j) th PCU as one PCU. Thereafter, the server may determine whether the (2+ j +1) th PCU is less than the online average value, starting with the (2+ j +1) th PCU. If the 2+ j +1 th PCU is the last PCU, the (2+ j +1) th PCU can be regarded as the last packet regardless of whether the (2+ j +1) th PCU is smaller than the online average value. Of course, if the sum of the 2 nd PCU to the 2+ j PCU is still less than the online average value until 2+ j equals the online anchor total number, i.e., the sum from the second PCU to the last PCU is still less than the online average value, the server may group the second PCU to the last PCU as one PCU.

Alternatively, if the second PCU is not less than the online average, the server may group the second PCU alone as one PCU, thereby obtaining a second PCU group. The server may then continue to determine whether the third PCU is less than the on-line average and process the third PCU based on the determination in reference to the aforementioned processing method for the second PCU, and so on.

For ease of understanding, the above implementation is described below by way of an example.

Assuming that the online average determined by step 301 is 5 and the current online anchor total is 5, the PCUs of the 5 online anchors are shown in table 1 below. The server may first determine whether the first PCU is less than 5. As can be seen from the data shown in table 1, the first PCU is not less than 5, and at this time, the server may group the first PCU alone as one PCU, thereby obtaining a first PCU group. The server may then continue to determine whether the second PCU is less than 5. Since the second PCU is also not smaller than 5, the second PCU alone can be grouped as one PCU, thereby obtaining a second PCU group. Next, the server may determine whether the third PCU is less than 5. Since the third PCU is less than 5, the server may calculate the sum of the third PCU and the fourth PCU and determine whether the sum of the third PCU and the fourth PCU is less than 5. Since the sum of the third PCU and the fourth PCU is not less than 5, the third PCU and the fourth PCU can be grouped as one PCU, resulting in a third PCU group. Thereafter, the server may determine whether the 5 th PCU is smaller than 5, and although the 5 th PCU is smaller than 5, since the 5 th PCU is the last PCU, the 5 th PCU may be grouped as the last PCU.

TABLE 1

Online anchor ordering	PCU
		1	10
2	5
		3	3
4	2
		5	1

Note that the present embodiment is described only by taking the data shown in table 1 above as an example, and table 1 above does not limit the present embodiment.

Step 303: the method comprises the steps of obtaining parameter values of a plurality of live broadcast parameters of each online anchor in a first time length before the current time, wherein the live broadcast parameters comprise at least one of video resolution, gift sending value, barrage number, sharing number and user watching time length in a live broadcast room where the corresponding online anchor is located.

It should be noted that the first duration may be preset, and the length of the first duration is not specifically limited in this application embodiment. When the anchor is played, the higher the video resolution ratio of the live broadcast room is, the easier the attraction is for the user to watch, the higher the gift sending value of the live broadcast room is, the more the barrage quantity is, the more the sharing quantity is, and the longer the watching time length of the user is, the higher the live broadcast quality of the live broadcast room is shown to be favored by the user, conversely, the lower the video resolution ratio of the live broadcast room is, the less the attraction is for the user to watch, the lower the gift sending value of the live broadcast room is, the less the barrage quantity is, the less the sharing quantity is, and the shorter the watching time length of the user is, the lower the live broadcast quality of the live broadcast room is shown. Therefore, the live broadcast parameters can reflect the quality of live broadcast.

In addition, when the anchor is played, the server can obtain the parameter values of the live broadcast parameters such as video resolution, gift sending value, barrage number, sharing number, user watching duration and the like of the live broadcast room where each online anchor is located in real time, and store the parameter values of the live broadcast parameters. Based on this, at the current moment, the server can count a plurality of parameter values of each live broadcast parameter within a first time length before the current moment from the current moment, so as to obtain parameter values of a plurality of live broadcast parameters of a live broadcast room in which each online anchor is located. Wherein, for different live broadcast parameters, the server can adopt different statistical methods. For example, for the video resolution, the server may obtain a plurality of video resolutions within a first duration, and calculate an average value of the plurality of video resolutions, so as to use the average value as a parameter value of the video resolution of the live broadcast room where the online anchor is located. For example, for the gift sending value, the server may count the total value of all gifts received by the online anchor in the live broadcast room within the first duration, and then use the total value as a parameter value of the gift sending value of the corresponding live broadcast room. For other live broadcast parameters, the server may select a suitable statistical method for performing statistics, which is not limited in the embodiment of the present application.

Step 304: and acquiring a history score corresponding to each live broadcast parameter in the plurality of live broadcast parameters of each online anchor.

It should be noted that, the server may store in advance a correspondence between a parameter value of the live broadcast parameter and the historical score. For any one live broadcast parameter in the live broadcast parameters of any one online anchor, after the server acquires the parameter value of the live broadcast parameter of the online anchor within a first time before the current time, the server can acquire a history score corresponding to the parameter value of the live broadcast parameter from the corresponding relation between the parameter value of the live broadcast parameter and the history score as the history score corresponding to the live broadcast parameter of the online anchor based on the parameter value of the live broadcast parameter.

Step 305: and determining the historical live broadcast quality score of each online anchor according to the parameter values of the live broadcast parameters of each online anchor and the historical score corresponding to each live broadcast parameter in the live broadcast parameters.

It should be noted that, since the plurality of live broadcast parameters may reflect the live broadcast quality, the historical live broadcast quality score of each online anchor may be determined according to the parameter values of the plurality of live broadcast parameters of each online anchor and the historical score corresponding to each live broadcast parameter of the plurality of live broadcast parameters. The historical live broadcast quality score can also reflect the live broadcast quality, and the higher the historical live broadcast quality score is, the higher the live broadcast quality is, the lower the historical live broadcast quality score is, and the lower the live broadcast quality is.

It should be noted that, for any one online anchor, the server may determine a product of a parameter value of each live broadcast parameter of the live broadcast room where the online anchor is located and the corresponding historical score, and then take a sum of the determined products as the historical live broadcast quality score of the online anchor.

After the historical live broadcast quality scores of each online anchor are determined through steps 303 to 305, anchor identifiers of a plurality of online anchors may be sorted according to the historical live broadcast quality scores of the plurality of online anchors to obtain an anchor list, and then the anchor identifiers included in the anchor list may be divided through step 306 to obtain a plurality of sub-lists.

Step 306: and dividing a plurality of anchor identifiers included in the anchor list according to the number of PCUs in each PCU group in the PCU groups to obtain a plurality of sub-lists.

The anchor list includes a plurality of anchor identifications, the anchor identifications refer to anchor identifications of all current online anchors, and each anchor identification can be used for uniquely identifying one anchor. And the plurality of anchor identifications are arranged in the sequence from high to low according to the historical live broadcast quality scores of the online anchors corresponding to each anchor identification. As can be seen from step 302, when the PCU list is divided, a plurality of PCU groups may be obtained in sequence, and in this case, the plurality of PCU groups may be arranged in the order of the division.

Illustratively, the implementation process of this step may be: let k equal to 1, and according to the number D of PCUs in the kth PCU group in the plurality of PCU groups_kObtaining front D from the Anchor list_kAn individual anchor identification; front D to be acquired_kThe individual anchor mark is used as the kth sub-list, and the front D in the anchor list is deleted_kThe number of PCUs in the kth PCU group in the plurality of PCU groups is returned according to the anchor mark, k is k +1_kObtaining front D from the Anchor list_kA step of anchor identification until k equals the number of PCU packets.

The PCU groups are arranged in the order of the divided PCU groups, so the server may first determine the number D of PCUs in the 1 st PCU group among the PCU groups₁Obtaining front D from the Anchor list₁Individual anchor identification, front D to be acquired₁The individual anchor mark is used as the 1 st sub-list, and the front D in the anchor list is deleted₁An individual anchor identification. The server may then determine a number D of PCUs in a 2 nd PCU group of the plurality of PCU groups₂From before deletion D₁Front D is obtained from anchor list after individual anchor identification₂Individual anchor identification, front D to be acquired₂The individual anchor mark is used as the 2 nd sub-list, and the top D in the anchor list is deleted₂An individual anchor identification. Next, the server may continue to determine a number D of PCUs in a 3 rd PCU group of the plurality of PCU groups₃And before deletion D₁Sum of before D₂Front D is obtained from anchor list after individual anchor identification₃Individual anchor identification, front D to be acquired₃The individual anchor mark is used as the 3 rd sub-list, and the top D in the anchor list is deleted₃An individual anchor identification. And so on until the server determines from the last PCU group that the last sublist is obtained.

Illustratively, still taking the example shown in table 1 in step 302 as an example, after the server sequentially groups the 1 st PCU as a first PCU, the 2 nd PCU as a second PCU, the 3 rd PCU and the 4 th PCU as a third PCU, and the 5 th PCU as a fourth PCU, the server may first obtain the first 1 anchor identifiers from the anchor list according to the number 1 of PCUs in the 1 st PCU group, use the obtained 1 anchor identifier as the 1 st sub-list, and delete the 1 st anchor identifier in the anchor list. Then, the server may obtain the first 1 anchor identifier from the anchor list after deleting the 1 st anchor identifier according to the number 1 of PCUs in the 2 nd PCU group, use the obtained anchor identifier as the 2 nd sub-list, and delete the 2 nd anchor identifier in the anchor list. Then, the server may obtain the first 2 anchor identifiers from the anchor list from which the 1 st and 2 nd anchor identifiers are deleted, according to the number 2 of PCUs in the 3 rd PCU group, use the obtained 2 anchor identifiers as a 3 rd sub-list, and delete the 3 rd and 4 th anchor identifiers in the anchor list. Finally, the server may identify the last 1 anchor identifier in the anchor list as the 4 th sub-list according to the number 1 of PCUs in the 4 th PCU group.

It should be noted that, after the server obtains the plurality of sub-lists, the server may set a corresponding key value for each sub-list in the plurality of sub-lists, and store the plurality of sub-lists and the corresponding key values in a distributed key-value storage database redis, so that when the server includes a plurality of distributed service nodes, and when the plurality of distributed service nodes receive an anchor recommendation request sent by the terminal, the plurality of distributed service nodes may call the database to obtain the corresponding sub-list, and send anchor recommendation information to the terminal according to the obtained sub-list.

Step 307: and sending the anchor recommendation information to the terminal according to the plurality of sub-lists.

After the plurality of sub-lists are obtained through division, the server can send the anchor recommendation information to the terminal according to the anchor identification contained in each sub-list. It should be noted that the anchor recommendation information may include information related to the anchor, such as a video of a live broadcast room where the anchor is located, a total number of fans of the anchor, a level of the anchor, and the like, which is not limited in this embodiment of the application.

Specifically, this step can be implemented by steps 3071 to 3073 as follows:

step 3071: and receiving the anchor recommendation requests sent by the terminal, wherein the anchor recommendation requests carry the request quantity, and the request quantity refers to the quantity of the anchors requesting recommendation.

When the terminal detects a start instruction of the installed live broadcast application program, a main broadcast recommendation request can be generated and sent to the server. Accordingly, the server may receive an anchor recommendation request sent by the terminal.

In order to avoid that the terminal frequently requests the anchor recommendation information from the server, the number of requests carried in the anchor recommendation request may be greater than the number of anchor recommendation information allowed to be displayed each time in the display interface of the live application program. For example, assuming that the display number of anchor recommendation information allowed to be displayed each time in the display interface of the live application is 8, the request number carried in the anchor recommendation request may be greater than 8, for example, may be 10, 30, or 60.

It should be noted that, in order to identify which terminal the anchor recommendation request is sent by, the anchor recommendation request may also carry a user identifier of the terminal. The user identifier may be information such as a user account for logging in the live application program.

Step 3072: and acquiring the anchor identifications with the quantity equal to the quantity of the requests from the plurality of sub-lists according to the anchor recommendation request, wherein the plurality of sub-lists are arranged according to the sequence obtained by division.

As shown in step 306, the sub-lists are obtained by sequentially dividing the anchor list according to the PCU groups, in which case, the sub-lists may be arranged according to the order of the division. And the popularity and the live broadcast quality of the anchor corresponding to the anchor identification included in the sub-list at the higher ranking are higher. Based on this, the server may push the anchor recommendation information to the terminal in the order of the plurality of sub-lists after receiving the anchor recommendation request.

It should be noted that the anchor recommendation request received by the server may be the first anchor recommendation request sent by the terminal, or may be a non-first anchor recommendation request in a plurality of anchor recommendation requests continuously sent by the terminal. Based on this, the server can push different anchor recommendation information to the terminal for different anchor recommendation requests.

Specifically, when it is determined that the anchor recommendation request is the first anchor recommendation request sent by the terminal according to the user identifier of the terminal, or when it is determined that the anchor recommendation request is not the first anchor recommendation request of the terminal according to the user identifier of the terminal and a difference between a receiving time of the anchor recommendation request and a receiving time of a previous anchor recommendation request is greater than a second duration, the anchor identifications equal to the requested number are sequentially obtained from the first sub-list of the plurality of sub-lists.

It should be noted that, when the server receives the anchor recommendation request sent by the terminal, it may search, according to the user identifier of the terminal carried in the anchor recommendation request, whether there is a record of the user identifier of the terminal from the correspondence between the stored user identifiers and the reception time of the anchor recommendation request, where the reception time corresponding to each user identifier in the stored correspondence is the time of the anchor recommendation request sent by the terminal of the corresponding user identifier that the server has received last time. Thus, if the user identifier of the terminal is not found, it may be determined that the currently received anchor recommendation request is the first anchor recommendation request sent by the terminal, and then the anchor identifiers with the number equal to the number of requests may be sequentially obtained from the first sub-list of the plurality of sub-lists; if the user identifier of the terminal is found, the server may obtain the receiving time corresponding to the user identifier of the terminal, and calculate a difference between the receiving time of the currently received anchor recommendation request and the obtained receiving time, and if the difference is greater than the second time length, it may be considered that a time interval between the currently received anchor recommendation request and the last received anchor recommendation request is longer, and at this time, the server may sequentially obtain the anchor identifiers equal to the requested number from a first sub-list of the plurality of sub-lists.

When the number of the anchor identifiers equal to the number of the requests is sequentially acquired from the first sub-list in the plurality of sub-lists, m sub-lists can be selected according to the number of the anchor identifiers and the number of the requests included in each sub-list, wherein the total number of the anchor identifiers included in the first m-1 sub-lists in the m sub-lists is smaller than the number of the requests, and the total number of the anchor identifiers included in the m sub-lists is greater than or equal to the number of the requests. When the total number of the anchor identifications included in the m sub-lists is equal to the request number, acquiring the anchor identifications included in the m sub-lists; when the total number of the anchor identifications included in the m sub-lists is greater than the request number, the anchor identifications included in the first m-1 sub-lists are obtained, and the anchor identifications with the residual number, which is the difference between the request number and the total number of the anchor identifications included in the first m-1 sub-lists, are obtained from the anchor identifications included in the last sub-list in the m sub-lists according to the sorting of the anchor identifications.

Further, if it is determined according to the user identifier of the terminal that the anchor recommendation request is not the first anchor recommendation request of the terminal and the difference between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is not greater than a second duration, obtaining historical output data corresponding to the terminal, wherein the historical output data comprises anchor identifiers sent to the terminal according to other anchor recommendation requests sent by the terminal before the anchor recommendation request; and sequentially acquiring the anchor identifications with the number equal to the number of the requests from the other anchor identifications except the historical output data in the plurality of sub-lists according to the arrangement order of the plurality of sub-lists and the arrangement order of the anchor identifications included in the plurality of sub-lists.

The server may store a corresponding relationship between the user identifier and the historical output data, where the historical output data corresponding to each user identifier includes an anchor identifier corresponding to anchor recommendation information sent to a corresponding terminal according to an anchor recommendation request carrying the user identifier before the current time. In this way, when the server receives the anchor recommendation request sent by the terminal at the current time, if the anchor recommendation request is not the first anchor recommendation request of the terminal and the difference between the receiving time of the anchor recommendation request and the receiving time of the last anchor recommendation request is not greater than the second time length, the server may obtain the historical output data corresponding to the user identifier of the terminal from the corresponding relationship according to the user identifier of the terminal carried in the currently received anchor recommendation request. And then, sequentially acquiring the same number of anchor identifications as the number of requests from other anchor identifications except the acquired historical output data in the plurality of sub-lists. In this way, it is possible to avoid repeatedly sending the same anchor identification to the terminal in a short time.

The manner of sequentially acquiring the anchor identifiers equal to the number of requests from the anchor identifiers in the plurality of sub-lists except the historical output data may refer to the manner of sequentially acquiring the anchor identifiers equal to the number of requests from the first sub-list in the plurality of sub-lists, that is, when the server acquires the anchor identifiers from the plurality of sub-lists, the anchor identifiers equal to the number of requests are sequentially acquired from the remaining anchor identifiers according to the sequence of the plurality of sub-lists except the anchor identifiers that have been output before, which is not described herein again.

It should be noted that, in order to play a role of promotion, the server may control individual anchor recommendation information to be top-displayed, in this case, the server may add an identifier to the anchor recommendation information (top-set data) when storing the anchor recommendation information, so that, when receiving an anchor recommendation request sent by the terminal, the server may preferentially obtain the top-set data, and then obtain the rest number of anchor identifiers except the top-set data from the plurality of sub-lists according to the number of requests in the anchor recommendation request.

Step 3073: and sending the anchor recommendation information corresponding to the obtained anchor identification to the terminal.

After the server acquires the anchor identifications with the quantity equal to the requested quantity from the plurality of sub-lists according to the anchor recommendation request sent by the terminal, the server can send the anchor recommendation information corresponding to the acquired anchor identifications to the terminal corresponding to the user identification of the terminal. When the server sends the anchor recommendation information to the terminal, the anchor identification corresponding to the anchor recommendation information and the serial number of the sub-list where the anchor identification is located can be sent to the terminal. Correspondingly, the terminal can receive the anchor recommendation information sent by the server, the anchor identification corresponding to the anchor recommendation information and the serial number of the sub-list where the anchor identification is located, and display the received anchor recommendation information on the display interface of the live broadcast application program according to the display number of the display interface of the live broadcast application program, the anchor identification corresponding to the anchor recommendation information and the serial number of the sub-list where the anchor identification is located.

Specifically, the implementation process of this step may be: randomly sequencing the anchor identifications belonging to the same sub-list in the sequentially acquired anchor identifications to obtain an anchor recommendation list; and sending the anchor recommendation information corresponding to each anchor identification to the terminal according to the sequence of the plurality of anchor identifications in the anchor recommendation list.

As can be seen from the description in step 3072, when the server obtains the same number of anchor identifiers as the requested number from the plurality of sub-lists according to the anchor recommendation request, the anchor identifiers are sequentially obtained according to the sorting order of the sub-lists, and the anchor identifiers in the sub-lists are sequentially sorted according to the historical quality scores, so that the obtained anchor identifiers are all arranged according to the order of the sub-lists and the sorting order of the anchor identifiers in the sub-lists. On this basis, in order to improve the recommendation rate of the anchor recommendation information corresponding to the anchor identifications ranked later in each sub-list, after acquiring the anchor identifications with the number equal to the requested number in sequence, the server may randomly rank the anchor identifications belonging to the same sub-list, so as to obtain an anchor recommendation list, wherein the ranking of the anchor identifications located more forward in the corresponding sub-list in the anchor recommendation list is more forward, and the ranking of the anchor identifications belonging to the same sub-list is random. Then, the server may send anchor recommendation information corresponding to each anchor identifier to the terminal according to the sequence of the plurality of anchor identifiers in the anchor recommendation list.

Exemplarily, it is assumed that the server sequentially obtains 6 anchor identifications, and the 6 anchor identifications are named as a 1 st anchor identification, a 2 nd anchor identification, a 3 rd anchor identification, a 4 th anchor identification, a 5 th anchor identification, and a 6 th anchor identification in sequence. Wherein, the 1 st anchor mark belongs to the 1 st sub-list, the 2 nd and 3 rd anchor marks belong to the 2 nd sub-list, the 4 th, 5 th and 6 th anchor marks belong to the 3 rd sub-list, according to the sequence of the sub-lists to which each anchor mark belongs, the 1 st anchor mark is arranged before the 2 nd anchor mark and the 3 rd anchor mark, and the 2 nd anchor mark and the 3 rd anchor mark are arranged before the 4 th anchor mark, the 5 th anchor mark and the 6 th anchor mark. On this basis, the server may randomly sort the 2 nd anchor identifier and the 3 rd anchor identifier that belong to the 2 nd sub-list, and randomly sort the 4 th anchor identifier, the 5 th anchor identifier and the 6 th anchor identifier that belong to the 3 rd sub-list, thereby obtaining an anchor recommendation list. Then, the server may send anchor recommendation information corresponding to each anchor identifier to the terminal based on the sequence of the plurality of anchor identifiers. In this way, the anchor recommendation information corresponding to the anchor identification ranked at the back in the same sub-list may be displayed in advance, increasing the exposure of the excellent new anchor.

When the terminal displays the received anchor recommendation information on the display interface of the live broadcast application program according to the display number of the display interface of the live broadcast application program, the anchor identification corresponding to the recommendation information and the serial number of the sub-list where the anchor identification is located, the sub-lists can be displayed according to the serial number sequence of the sub-list where the anchor identification is located. That is, the earlier the sequence number of the sub-list corresponding to the anchor identifier is, the earlier the anchor recommendation information corresponding to the anchor identifier in the sub-list can be displayed.

In addition, when the server sends the obtained anchor recommendation information corresponding to the anchor identification to the terminal, the anchor identifications belonging to the same sub-list in the successively obtained anchor identifications can be randomly sequenced to obtain an anchor recommendation list, and then the anchor recommendation information corresponding to each anchor identification is sent to the terminal according to the sequence of the plurality of anchor identifications in the anchor recommendation list.

Fig. 4 is a schematic structural diagram of an anchor recommendation apparatus according to an embodiment of the present application. Referring to fig. 4, the apparatus includes: a first determining module 401, a first dividing module 402, a second dividing module 403 and a sending module 404.

A first determining module 401, configured to determine an online average value according to a total number of current online anchor and a current number of highest simultaneous online users PCU, where the PCUs are PCUs in a live broadcast room where multiple online anchors are located;

a first dividing module 402, configured to divide a PCU list according to the online average value to obtain a plurality of PCU groups, where the PCU list includes the PCUs, and the PCUs are arranged according to a numerical value;

a second dividing module 403, configured to divide the multiple anchor identifiers included in the anchor list according to the number of PCUs in each PCU group in the multiple PCU groups to obtain multiple sub-lists, where the anchor list is obtained by sorting according to historical live broadcast quality scores of the multiple online anchors;

a sending module 404, configured to send anchor recommendation information to the terminal according to the plurality of sub-lists.

Optionally, the first determining module 401 is configured to:

acquiring a preset adjustment threshold;

a first partitioning module 402 for:

Optionally, the first dividing module 402 is further configured to:

Optionally, the apparatus further comprises:

a second dividing module 403, configured to:

Optionally, the sending module 404 includes:

the receiving submodule is used for receiving anchor recommendation requests sent by the terminal, the anchor recommendation requests carry request quantity, and the request quantity refers to the quantity of the anchors requesting recommendation;

the acquisition sub-module is used for acquiring anchor identifications with the quantity equal to the quantity of the requests from the plurality of sub-lists according to the anchor recommendation requests;

an acquisition submodule for:

an acquisition submodule, further configured to:

Optionally, the sending submodule is configured to:

It should be noted that: in the foregoing embodiment, when recommending a main broadcast, the main broadcast recommendation apparatus is described by way of example only by dividing the functional modules, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure of the apparatus is divided into different functional modules to complete all or part of the functions described above. In addition, the anchor recommendation device provided by the above embodiment and the anchor recommendation method embodiment belong to the same concept, and specific implementation processes thereof are detailed in the method embodiment and are not described herein again.

Fig. 5 is a schematic structural diagram of an anchor recommendation apparatus provided in an embodiment of the present application, where the apparatus may be a server 500, and the server 500 may be a server in a background server cluster. Specifically, the method comprises the following steps:

the server 500 includes a Central Processing Unit (CPU)501, a system memory 504 including a Random Access Memory (RAM)502 and a Read Only Memory (ROM)503, and a system bus 505 connecting the system memory 504 and the central processing unit 501. The server 500 also includes a basic input/output system (I/O system) 506, which facilitates transfer of information between devices within the computer, and a mass storage device 507, which stores an operating system 513, application programs 514, and other program modules 515.

The basic input/output system 506 comprises a display 508 for displaying information and an input device 509, such as a mouse, keyboard, etc., for user input of information. Wherein a display 508 and an input device 509 are connected to the central processing unit 501 through an input/output controller 510 connected to the system bus 505. The basic input/output system 506 may also include an input/output controller 510 for receiving and processing input from a number of other devices, such as a keyboard, mouse, or electronic stylus. Similarly, an input/output controller 510 also provides output to a display screen, a printer, or other type of output device.

The mass storage device 507 is connected to the central processing unit 501 through a mass storage controller (not shown) connected to the system bus 505. The mass storage device 507 and its associated computer-readable media provide non-volatile storage for the server 500. That is, the mass storage device 507 may include a computer readable medium (not shown) such as a hard disk or CD-ROM drive.

Without loss of generality, computer readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid state storage technology, and includes CD-ROM, DVD, or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media is not limited to the foregoing. The system memory 504 and mass storage device 507 may be collectively referred to as memory.

According to various embodiments of the invention, server 500 may also operate as a remote computer connected to a network through a network, such as the Internet. That is, the server 500 may be connected to the network 512 through the network interface unit 511 connected to the system bus 505, or may be connected to other types of networks or remote computer systems (not shown) using the network interface unit 511.

The memory further includes one or more programs, and the one or more programs are stored in the memory and configured to be executed by the CPU. The one or more programs contain instructions for performing the anchor recommendation method provided by embodiments of the present application.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only exemplary of the present application and should not be taken as limiting, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An anchor recommendation method, the method comprising:

determining an online average value according to the total number of the current online anchor and the current multiple highest simultaneous online user number PCUs, wherein the PCUs are PCUs of a live broadcast room where the multiple online anchors are located, and the PCUs are arranged in a PCU list according to numerical values in a descending order;

dividing the PCU list into a plurality of PCU groups according to the arrangement sequence of the PCUs from large to small, wherein the sum of PCUs in each PCU group in the PCU groups is larger than or equal to the online average value, and the PCU in each PCU group is an adjacent PCU in the PCU list, or the sum of PCUs in each PCU group in T PCU groups in the PCU groups is larger than or equal to the online average value, the sum of PCUs in one PCU group is smaller than the online average value, and T +1 is the number of the PCU groups;

according to the sequence of the PCU groups, acquiring anchor identifiers corresponding to the number of PCUs in each PCU group from a plurality of anchor identifiers included in an anchor list to obtain a plurality of sub-lists, wherein the anchor list is obtained by sequencing from high to low according to historical live broadcast quality scores of online anchors corresponding to the anchor identifiers;

2. The method of claim 1, wherein determining an online average value based on the current online anchor total and a current plurality of highest simultaneous online user numbers PCU comprises:

acquiring a preset adjustment threshold;

3. The method of claim 1, wherein the dividing the list of PCUs into a plurality of PCU groups in an order of the PCUs from large to small comprises:

4. The method of claim 3, wherein after determining whether an ith PCU in the PCU list is less than the online average, further comprising:

5. The method of claim 1, wherein before the dividing the plurality of anchor identifiers included in the anchor list according to the number of PCUs in each PCU group of the plurality of PCU groups to obtain a plurality of sub-lists, the method further comprises:

6. The method according to any of claims 1 to 5, wherein said obtaining, according to the precedence division order of the PCU groups, a anchor identifier corresponding to the number of PCUs in each PCU group from a plurality of anchor identifiers included in an anchor list to obtain a plurality of sub-lists, includes:

7. The method of claim 1, wherein the sending the anchor recommendation information to the terminal according to the plurality of sub-lists comprises:

8. The method according to claim 7, wherein the plurality of sub-lists are arranged according to a sequence obtained by dividing, and the anchor recommendation request further carries a user identifier of the terminal;

9. The method of claim 7, wherein the plurality of sub-lists are arranged in a sequential order obtained by the dividing;

10. The method according to claim 8 or 9, wherein the sending the anchor recommendation information corresponding to the obtained anchor identifier to the terminal includes:

11. An anchor recommendation apparatus, the apparatus comprising:

the system comprises a first determining module, a second determining module and a third determining module, wherein the first determining module is used for determining an online average value according to the total number of current online anchor and a plurality of current highest simultaneous online user number PCUs, the PCUs are PCUs of a live broadcast room where the online anchor is located, and the PCUs are arranged in a PCU list according to numerical values in a descending order;

a first dividing module, configured to divide the PCU list into a plurality of PCU groups according to an arrangement order of the PCUs from large to small, where a sum of PCUs in each PCU group in the PCU groups is greater than or equal to the online average value, and the PCU in each PCU group is an adjacent PCU in the PCU list, or a sum of PCUs in T PCU groups in the PCU groups is greater than or equal to the online average value, a sum of PCUs in one group is smaller than the online average value, and T +1 is a number of the PCU groups;

a second dividing module, configured to obtain, according to a sequential dividing order of the plurality of PCU groups, a anchor identifier corresponding to the number of PCUs in each PCU group from a plurality of anchor identifiers included in an anchor list, and obtain a plurality of sub-lists, where the anchor list is obtained by sorting, in order from high to low, historical live broadcast quality scores of online anchors corresponding to the plurality of anchor identifiers;

12. An anchor recommendation apparatus, the apparatus comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to perform the steps of any of the methods of claims 1-10.

13. A computer-readable storage medium having instructions stored thereon, wherein the instructions, when executed by a processor, implement the steps of any of the methods of claims 1-10.