Recommendation solution generation method, device, computer equipment and storage medium
By receiving the scheme configuration instructions, the target recommendation dimensions are automatically combined to generate recommendation solutions for the live broadcast client, which solves the problems of high cost and long development of live broadcast recommendation system modification and new solutions in the existing technology, and realizes automatic generation and rapid launch.
Patent Information
- Application Number
- CN202211192101.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-28
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-09-28
AI Technical Summary
When modifying or adding new recommendation plans, a lot of manpower is required to redevelop, resulting in high development costs and long online cycle.
By receiving solution configuration instructions, determine the target recommendation dimensions and combination solutions, and automatically combine them to form the target recommendation solutions for the live broadcast client to reduce the demand for manpower development.
It realizes automatic generation of recommended solutions without a lot of manpower, saves development costs and shortens the online cycle of recommended solutions.
Smart Images

Figure CN115834916B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, computer device and storage medium for generating a recommendation solution. Background Art
[0002] With the development of the live broadcast industry, more and more users have joined the craze of watching live broadcasts. Live broadcast recommendation is the most basic scenario application in live broadcast. The live broadcast recommendation schemes are complex and changeable and change frequently. The existing live broadcast recommendation systems need to spend a lot of manpower to redevelop any modification or addition of recommendation schemes, resulting in a high development cost of the recommendation scheme and a long online cycle for the modified recommendation scheme. Summary of the invention
[0003] In order to solve the above technical problems, the present application provides a method, apparatus, computer device and storage medium for generating a recommendation solution.
[0004] In a first aspect, the present application provides a method for generating a recommendation solution, comprising:
[0005] In the case of receiving the scheme configuration instruction, determining a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to the live broadcast client;
[0006] The target recommendation scheme corresponding to the live broadcast client is determined based on the combination result formed by combining each of the target recommendation dimensions according to the target combination scheme.
[0007] In a second aspect, the present application provides a recommendation scheme generating device, comprising:
[0008] A configuration module, configured to determine, upon receiving a scheme configuration instruction, a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to a live broadcast client;
[0009] A combination module is used to determine the target recommendation scheme corresponding to the live broadcast client by combining the target recommendation dimensions based on the target combination scheme to form a combination result.
[0010] In a third aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the following steps are implemented:
[0011] In the case of receiving the scheme configuration instruction, determining a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to the live broadcast client;
[0012] The target recommendation scheme corresponding to the live broadcast client is determined based on the combination result formed by combining each of the target recommendation dimensions according to the target combination scheme.
[0013] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:
[0014] In the case of receiving the scheme configuration instruction, determining a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to the live broadcast client;
[0015] The target recommendation scheme corresponding to the live broadcast client is determined based on the combination result formed by combining each of the target recommendation dimensions according to the target combination scheme.
[0016] Based on the above-mentioned recommendation scheme generation method, when a scheme configuration instruction is received, multiple target recommendation dimensions and a target combination scheme for combining each of the target recommendation dimensions are determined according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request the generation of a target recommendation scheme corresponding to the live broadcast client; based on the target combination scheme, the target recommendation scheme corresponding to the live broadcast client is determined by combining each of the target recommendation dimensions to form a combination result. Based on the scheme configuration instruction, multiple target recommendation dimensions and a target combination scheme for combining each of the target recommendation dimensions are screened out, and multiple target recommendation dimensions are combined according to the target combination scheme, thereby automatically generating a target recommendation scheme, without spending a lot of manpower on redeveloping the recommendation scheme, saving manpower development costs and shortening the launch cycle of the recommendation scheme. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0019] Figure 1 An application environment diagram of a method for generating a recommendation solution in an embodiment;
[0020] Figure 2 A schematic diagram of a flow chart of a method for generating a recommendation solution in one embodiment;
[0021] Figure 3 A schematic diagram of a flow chart of a method for generating a recommendation solution in one embodiment;
[0022] Figure 4 It is a structural block diagram of a device for generating a recommendation solution in one embodiment;
[0023] Figure 5 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION
[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0025] Figure 1 FIG. 1 is an application environment diagram of a method for generating a recommendation solution in an embodiment. Figure 1 The method for generating a recommendation scheme is applied to a recommendation scheme generating system. The recommendation scheme generating system includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected via a network. The terminal 110 may be a desktop terminal or a mobile terminal, and the mobile terminal may be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The server 120 may be implemented as an independent server or a server cluster consisting of multiple servers.
[0026] In one embodiment, Figure 2 A flowchart of a method for generating a recommendation scheme in an embodiment is shown in FIG. Figure 2 , provides a method for generating a recommendation solution. This embodiment mainly applies this method to the above Figure 1 Taking the server 120 in the example as an example, the method for generating a recommendation solution specifically includes the following steps:
[0027] Step S210, when a solution configuration instruction is received, a plurality of target recommendation dimensions and a target combination solution for combining the target recommendation dimensions are determined according to the solution configuration instruction.
[0028] The solution configuration instruction is used to request the generation of a target recommendation solution corresponding to the live broadcast client.
[0029] Specifically, the plan configuration instruction is an instruction triggered by the backend personnel through the operation terminal and sent to the server. The plan configuration instruction is used to select the target recommendation dimension and the target combination plan. The target recommendation dimension can be any recommended dimension. The recommended dimension can specifically be live broadcast duration, popularity, hourly ranking, broadcast type, whether it is a public performance, live broadcast status, anchor policy type, anchor rating, etc. The target combination plan includes a target combination relationship between at least any two target recommendation dimensions. The target combination relationship can specifically be an intersection (and), union (or), not, etc.
[0030] Step S220, determining the target recommendation scheme corresponding to the live broadcast client based on the target combination scheme by combining the target recommendation dimensions into a combination result.
[0031] Specifically, multiple target recommendation dimensions and target combination schemes for combining the target recommendation dimensions are screened out based on the scheme configuration instructions, and the multiple target recommendation dimensions are combined according to the target combination scheme, so as to automatically generate a target recommendation scheme. There is no need to spend a lot of manpower to redevelop the recommendation scheme, which saves manpower development costs and shortens the online cycle of the recommendation scheme.
[0032] In one embodiment, the scheme configuration instruction includes a first circle selection instruction and a second circle selection instruction, and the determining of a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction includes:
[0033] Filtering out the multiple target recommended dimensions and target recommendation parameters corresponding to each of the target recommended dimensions in a recommended dimension list according to the first circle selection instruction, wherein the recommended dimension list includes multiple recommended dimensions, and the target recommended dimension is any one of the recommended dimensions;
[0034] The target combination relationship between each of the target recommendation dimensions is screened out in the combination logic list according to the second circle selection instruction, wherein the combination logic list includes a plurality of combination relationships, and the target combination scheme includes the target combination relationship between each of the target recommendation dimensions.
[0035] Specifically, the basic dimension data is calculated by offline analysis using JOB, and the basic dimension data is stored in the mongo basic dimension table through offline scheduling tasks, thus forming the above-mentioned recommended dimension list. The first circle selection instruction is an instruction generated based on the selection operation of different recommended dimensions in the recommended dimension list by the backend personnel, that is, the modification of the recommended dimension by the backend personnel does not need to be implemented by rewriting the code and redeveloping it, but is quickly implemented by modifying the selection status of the abstracted recommended dimension, that is, the selection status of the recommended dimension that needs to participate in the generation of the target recommendation scheme is selected, and the selection status of the recommended dimension that does not need to participate in the generation of the target recommendation scheme is unselected. On this basis, the corresponding target recommendation parameters of each target recommended dimension whose selection status is selected can also be determined according to the first circle selection instruction. For example, the target recommended dimension is the start-up type, and the corresponding target recommendation parameter is 1, indicating the type corresponding to the start-up type of 1.
[0036] The second circle selection instruction is an instruction generated based on the selection operation of different combination relationships in the combination logic list by the backend personnel. That is, the backend personnel do not need to rewrite the code to redevelop the combination relationship between different recommendation dimensions, but can quickly implement it by modifying the selection status of the combination relationship. That is, the selection status of the combination relationship that needs to participate in the generation of the target recommendation plan is selected, and the selection status of the combination relationship that does not need to participate in the generation of the target recommendation plan is unselected.
[0037] The abstraction of the underlying recommendation dimension data facilitates the expansion of rules. For example, if we select the top 5 or top 10 hosts by popularity, such simple rule changes previously required the redevelopment of corresponding rules. Now we only need to abstract the host popularity value, and then use the popularity value to assemble the rules, which greatly simplifies the development of the underlying data.
[0038] In one embodiment, the target recommendation scheme corresponding to the live broadcast client is determined based on the target combination scheme by combining the target recommendation dimensions to form a combination result, including:
[0039] According to the multiple target combination relationships in the target combination scheme, each of the target recommendation dimensions and the corresponding target recommendation parameters are combined to determine at least one live broadcast screening scheme;
[0040] Determine the live broadcast push scheme corresponding to each of the live broadcast screening schemes according to the received third circle selection instruction, wherein the live broadcast push scheme is used to sort and push the live broadcast data screened based on the corresponding live broadcast screening scheme;
[0041] According to the live broadcast screening scheme and the live broadcast pushing scheme, a target recommendation scheme corresponding to the live broadcast client is determined.
[0042] Specifically, multiple target recommendation dimensions and multiple target combination relationships are combined to determine at least one live broadcast screening scheme. For example, the target recommendation dimensions include anchor score, anchor policy type, broadcast type, and popularity, and the target recommendation parameter corresponding to the anchor score is greater than or equal to 8 points, the target recommendation parameter corresponding to the anchor policy type is 3, the target recommendation parameter corresponding to the broadcast type is 1, and the target recommendation parameter corresponding to the popularity indicates descending order. The target combination scheme includes two combination relationships: intersection and union. Therefore, it can be combined into a live broadcast screening scheme of "anchors with anchor score >= 8 points or anchor policy type 3, and broadcast type 1, in descending order of popularity", that is, different live broadcast screening schemes are used to indicate different anchor screening rules, and multiple push data can be screened out based on the live broadcast screening scheme.
[0043] The third circle selection instruction is used to select the live push scheme corresponding to the live screening scheme, and the live push scheme is used to indicate the sorting and pushing rules for sorting and pushing multiple push data selected based on the live screening scheme. The live push scheme can be specifically a random sorting scheme, a weighted random sorting scheme, a rule hit probability sorting scheme, a fixed position sorting scheme, etc. The random sorting scheme is to randomly sort multiple push data and push them in order. The weighted random sorting scheme is to perform weighted processing on each push data, randomly sort the weighted push data and push them in order. The push data with a larger weight value has a higher probability of being sorted in the front, or the push data can be directly sorted in descending order according to the weight value. The rule hit probability sorting scheme is to allocate the hit probability of each weighted push data, obtain the corresponding probability distribution data, randomly sort each probability distribution data and push them in order. The probability distribution data with a larger hit probability has a higher probability of being sorted in the front, or the probability distribution data can be directly sorted in descending order according to the hit probability and pushed in order. The fixed position sorting scheme pushes the push data filtered by the live broadcast filtering scheme associated with each fixed position, and the fixed position is used to indicate the display area in the live broadcast client display interface.
[0044] Based on the live broadcast screening plan and the live broadcast push plan, that is, according to the anchor screening rules and the sorting and push rules after screening, the target recommendation plan corresponding to the live broadcast client is determined.
[0045] In one embodiment, determining the target recommendation scheme corresponding to the live broadcast client according to the live broadcast screening scheme and the live broadcast push scheme includes:
[0046] Generate corresponding live broadcast candidate solutions according to each of the live broadcast screening solutions and the corresponding live broadcast push solutions;
[0047] Determining, according to the received scheme allocation instruction, the corresponding relationship between each page in the live broadcast client and each of the live broadcast candidate schemes;
[0048] According to the candidate live broadcast solutions that have a corresponding relationship with each page in the live broadcast client, a target recommendation solution corresponding to the live broadcast client is generated.
[0049] Specifically, the indicator screening scheme and the corresponding live broadcast push scheme are combined to generate a live broadcast candidate scheme. Since the live broadcast client contains multiple pages, each page corresponds to different anchor screening rules and different live broadcast push schemes. The scheme allocation instruction is used to allocate corresponding live broadcast candidate schemes to each page, that is, to determine the correspondence between each page and each live broadcast candidate scheme. Each page has a corresponding relationship with at least one live broadcast candidate scheme. Based on the live broadcast candidate scheme corresponding to each page, the target recommendation scheme corresponding to the live broadcast client is generated, that is, the anchor screening rules and push sorting rules are assigned to each page to form the target recommendation scheme corresponding to the live broadcast client. There is still no need to rewrite the code of the page and redevelop it, and the configuration of the anchor screening rules and sorting push rules corresponding to the page can be realized.
[0050] In one embodiment, determining the correspondence between each page in the live broadcast client and each of the live broadcast candidate solutions according to the received solution allocation instruction includes:
[0051] Determine, according to the received scheme allocation instruction, the correspondence between different blocks in the target page in the live client and each of the live candidate schemes, wherein the target page is any page in the live client;
[0052] The live broadcast candidate solutions that have a corresponding relationship with each block in the target page form a first solution set, and the corresponding relationship between the target page and the first solution set is determined.
[0053] Specifically, each page contains multiple blocks. For example, the target page is a popular page, and the popular page contains a voice block, a new user block, a video anchor block, etc. The live broadcast candidate schemes corresponding to each block in the same page are determined based on the scheme allocation instruction, and the live broadcast candidate schemes corresponding to each block are cached in the first scheme set (mongo set), that is, the live broadcast candidate schemes corresponding to all blocks in the same page are cached in a first scheme set, and different pages correspond to different first scheme sets. The corresponding update refresh rate is configured for each live broadcast candidate scheme in the corresponding first scheme set according to the offline frequency of different pages. For example, the popular page has a relatively high real-time requirement, so the offline frequency of the popular page is relatively short, and the update refresh rate corresponding to each live broadcast candidate scheme in the first scheme set corresponding to the popular page is 1 second, that is, the filtered push data is pushed out every 1 second according to the live broadcast candidate scheme.
[0054] In one embodiment, determining the correspondence between different blocks in the target page in the live broadcast client and each of the live broadcast candidate solutions according to the received solution allocation instruction includes:
[0055] According to the received scheme allocation instruction, determine the correspondence between different functional points in the target block and each of the live broadcast candidate schemes, wherein the target block is any block in the target page, and the functional point is a fixed point or a waterfall flow point;
[0056] The live broadcast candidate solutions that have a corresponding relationship with each functional point in the target block form a second solution set, and the corresponding relationship between the target block and the second solution set is determined.
[0057] Specifically, since the block contains multiple functional points, different functional points may be used to display different push data. According to the plan allocation instruction, the corresponding live broadcast candidate plan is allocated to each functional point in the target block. The functional points are divided into fixed points or waterfall points. Fixed positioning and waterfall points are used to indicate different content layout types in the block. At fixed points, the filtered push data is fixedly displayed according to the corresponding live broadcast candidate plan. At waterfall points, the filtered push data can be slidably displayed according to the corresponding live broadcast candidate plan.
[0058] The live broadcast candidate solution sets corresponding to different functional points in the same block are formed into a second solution set, and the corresponding relationship between each block and each second solution set is determined, that is, each block corresponds to a second solution set.
[0059] In one embodiment, generating a target recommendation scheme corresponding to the live broadcast client according to the live broadcast candidate schemes corresponding to each page in the live broadcast client includes:
[0060] Generate a candidate recommendation scheme corresponding to the live broadcast client according to the live broadcast candidate schemes that have a corresponding relationship with each page in the live broadcast client
[0061] The obtained personalized recommendation scheme is combined with the candidate recommendation scheme to generate a target recommendation scheme corresponding to the live broadcast client.
[0062] Specifically, candidate recommendation plans are generated according to the live broadcast candidate plans corresponding to each page in the live broadcast client, that is, the first plan set corresponding to each page is integrated to form a candidate recommendation plan. The candidate recommendation plan is a recommendation plan generated based on the operation of the backend personnel. The backend personnel operate according to business needs, that is, the candidate recommendation plan can be dynamically customized based on actual business needs, so as to simply and quickly implement the change of recommendation rules without manual code redevelopment, such as Figure 3 As shown, only target recommendation dimensions and target combination solutions are selected based on business needs to allocate rules for functional points in blocks layer by layer, and then rule sets are made for multiple blocks in the page, and then rule sets are made for all pages, so as to integrate and generate candidate recommendation solutions. Figure 3 The circle selection rule in is used to indicate the live broadcast screening plan, the push rule is used to indicate the live broadcast push plan, area is used to indicate the functional point, and the cached page1 information is used to indicate the first plan set corresponding to a page.
[0063] On this basis, we combine personalized recommendation solutions to generate target recommendation solutions for live broadcast clients. The personalized recommendation solutions are based on the user's historical behavior log as a screening rule to push data that the user is interested in. All pages corresponding to the live broadcast client call the corresponding recommendation solutions through a unified API interface. The unified API layer facilitates the maintenance of the interface, and the page will be scheduled in real time according to the real-time nature of the configuration. For example, the homepage is composed of fixed position recommendations and personalized recommendations from the operator. The interface page supports page data calls and personalized data real-time call encapsulation.
[0064] In summary, it used to take a long development cycle to configure a recommendation page. Now, there is no need to redevelop it, and you only need to dynamically configure the page. Therefore, after going online, the development cycle has been greatly shortened, saving a lot of manpower. And based on a unified solution set, it is convenient to quickly locate the cause of page data anomalies.
[0065] Figure 2 FIG. 1 is a flow chart of a method for generating a recommendation solution in one embodiment. It should be understood that although Figure 2The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 2 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.
[0066] In one embodiment, Figure 4 As shown, a device for generating a recommendation scheme is provided, comprising:
[0067] The configuration module 310 is used to determine a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction when receiving the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to the live broadcast client;
[0068] The combination module 320 is used to determine the target recommendation scheme corresponding to the live broadcast client by combining the target recommendation dimensions based on the target combination scheme to form a combination result.
[0069] In one embodiment, the configuration module 310 is specifically used to:
[0070] Filtering out the multiple target recommended dimensions and target recommendation parameters corresponding to each of the target recommended dimensions in a recommended dimension list according to the first circle selection instruction, wherein the recommended dimension list includes multiple recommended dimensions, and the target recommended dimension is any one of the recommended dimensions;
[0071] The target combination relationship between each of the target recommendation dimensions is screened out in the combination logic list according to the second circle selection instruction, wherein the combination logic list includes a plurality of combination relationships, and the target combination scheme includes the target combination relationship between each of the target recommendation dimensions.
[0072] In one embodiment, the combining module 320 is specifically used for:
[0073] According to the multiple target combination relationships in the target combination scheme, each of the target recommendation dimensions and the corresponding target recommendation parameters are combined to determine at least one live broadcast screening scheme;
[0074] Determine the live broadcast push scheme corresponding to each of the live broadcast screening schemes according to the received third circle selection instruction, wherein the live broadcast push scheme is used to sort and push the live broadcast data screened based on the corresponding live broadcast screening scheme;
[0075] According to the live broadcast screening scheme and the live broadcast pushing scheme, a target recommendation scheme corresponding to the live broadcast client is determined.
[0076] In one embodiment, the combining module 320 is specifically used for:
[0077] Generate corresponding live broadcast candidate solutions according to each of the live broadcast screening solutions and the corresponding live broadcast push solutions;
[0078] Determining, according to the received scheme allocation instruction, the corresponding relationship between each page in the live broadcast client and each of the live broadcast candidate schemes;
[0079] According to the candidate live broadcast solutions that have a corresponding relationship with each page in the live broadcast client, a target recommendation solution corresponding to the live broadcast client is generated.
[0080] In one embodiment, the combining module 320 is specifically used for:
[0081] Determine, according to the received scheme allocation instruction, the correspondence between different blocks in the target page in the live client and each of the live candidate schemes, wherein the target page is any page in the live client;
[0082] The live broadcast candidate solutions that have a corresponding relationship with each block in the target page form a first solution set, and the corresponding relationship between the target page and the first solution set is determined.
[0083] In one embodiment, the combining module 320 is specifically used for:
[0084] According to the received scheme allocation instruction, determine the correspondence between different functional points in the target block and each of the live broadcast candidate schemes, wherein the target block is any block in the target page, and the functional point is a fixed point or a waterfall flow point;
[0085] The live broadcast candidate solutions that have a corresponding relationship with each functional point in the target block form a second solution set, and the corresponding relationship between the target block and the second solution set is determined.
[0086] In one embodiment, the combining module 320 is specifically used for:
[0087] Generate a candidate recommendation scheme corresponding to the live broadcast client according to the live broadcast candidate schemes that have a corresponding relationship with each page in the live broadcast client
[0088] The obtained personalized recommendation scheme is combined with the candidate recommendation scheme to generate a target recommendation scheme corresponding to the live broadcast client.
[0089] Figure 5 The internal structure diagram of a computer device in one embodiment is shown. The computer device may specifically be Figure 1 The server 120 in FIG. Figure 5 As shown, the computer device includes a processor, a memory, a network interface, an input device and a display screen connected through a system bus. Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the recommended solution generation method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute the recommended solution generation method. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse, etc.
[0090] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0091] In one embodiment, the recommendation scheme generating device provided by the present application can be implemented in the form of a computer program. The computer program can be Figure 5 The memory of the computer device may store various program modules constituting the recommendation scheme generating device, for example, Figure 4 The configuration module 310 and the combination module 320 are shown. The computer program composed of various program modules enables the processor to execute the steps of the recommendation solution generation method of each embodiment of the present application described in this specification.
[0092] Figure 5 The computer device shown can be Figure 4The configuration module 310 in the recommendation scheme generating device shown in the figure performs, upon receiving the scheme configuration instruction, determining a plurality of target recommendation dimensions and a target combination scheme for combining each of the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request the generation of a target recommendation scheme corresponding to the live broadcast client. The computer device can perform the combination result formed by combining each of the target recommendation dimensions based on the target combination scheme through the combination module 320 to determine the target recommendation scheme corresponding to the live broadcast client.
[0093] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method described in any one of the above embodiments is implemented when the processor executes the computer program.
[0094] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the above embodiments is implemented.
[0095] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0096] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0097] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A method for generating a recommendation scheme, characterized in that: The method comprises: In the case of receiving the scheme configuration instruction, determining a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to the live broadcast client; Determine the target recommendation scheme corresponding to the live broadcast client based on the target combination scheme by combining the target recommendation dimensions to form a combination result; The target recommendation scheme corresponding to the live broadcast client is determined by combining the target recommendation dimensions based on the target combination scheme to form a combination result, including: According to the multiple target combination relationships in the target combination scheme, each of the target recommendation dimensions and the corresponding target recommendation parameters are combined to determine at least one live broadcast screening scheme; Determine the live broadcast push scheme corresponding to each of the live broadcast screening schemes according to the received third circle selection instruction, wherein the live broadcast push scheme is used to sort and push the live broadcast data screened based on the corresponding live broadcast screening scheme; Determine a target recommendation scheme corresponding to the live broadcast client according to the live broadcast screening scheme and the live broadcast push scheme; The determining, according to the live broadcast screening scheme and the live broadcast push scheme, a target recommendation scheme corresponding to the live broadcast client comprises: Generate corresponding live broadcast candidate solutions according to each of the live broadcast screening solutions and the corresponding live broadcast push solutions; Determining, according to the received scheme allocation instruction, the corresponding relationship between each page in the live broadcast client and each of the live broadcast candidate schemes; Generate a target recommendation solution corresponding to the live broadcast client according to the live broadcast candidate solutions that have a corresponding relationship with each page in the live broadcast client; The determining, according to the received scheme allocation instruction, the corresponding relationship between each page in the live broadcast client and each of the live broadcast candidate schemes includes: Determine, according to the received scheme allocation instruction, the correspondence between different blocks in the target page in the live client and each of the live candidate schemes, wherein the target page is any page in the live client; The live broadcast candidate solutions that have a corresponding relationship with each block in the target page form a first solution set, and the corresponding relationship between the target page and the first solution set is determined.
2. The method according to claim 1, characterized in that: The scheme configuration instruction includes a first circle selection instruction and a second circle selection instruction, and the determining of a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction includes: Filtering out the multiple target recommended dimensions and target recommendation parameters corresponding to each of the target recommended dimensions in a recommended dimension list according to the first circle selection instruction, wherein the recommended dimension list includes multiple recommended dimensions, and the target recommended dimension is any one of the recommended dimensions; The target combination relationship between each of the target recommendation dimensions is screened out in the combination logic list according to the second circle selection instruction, wherein the combination logic list includes a plurality of combination relationships, and the target combination scheme includes the target combination relationship between each of the target recommendation dimensions.
3. The method according to claim 1, characterized in that The determining, according to the received scheme allocation instruction, the corresponding relationship between different blocks in the target page in the live broadcast client and each of the live broadcast candidate schemes includes: According to the received scheme allocation instruction, determine the correspondence between different functional points in the target block and each of the live broadcast candidate schemes, wherein the target block is any block in the target page, and the functional point is a fixed point or a waterfall flow point; The live broadcast candidate solutions that have a corresponding relationship with each functional point in the target block form a second solution set, and the corresponding relationship between the target block and the second solution set is determined.
4. The method according to claim 1, characterized in that: The generating a target recommendation scheme corresponding to the live broadcast client according to the live broadcast candidate schemes corresponding to each page in the live broadcast client includes: Generate a candidate recommendation scheme corresponding to the live broadcast client according to the live broadcast candidate schemes that have a corresponding relationship with each page in the live broadcast client The obtained personalized recommendation scheme is combined with the candidate recommendation scheme to generate a target recommendation scheme corresponding to the live broadcast client.
5. A device for generating a recommendation scheme, characterized in that: The device comprises: A configuration module, configured to determine, upon receiving a scheme configuration instruction, a plurality of target recommendation dimensions and a target combination scheme for combining the target recommendation dimensions according to the scheme configuration instruction, wherein the scheme configuration instruction is used to request generation of a target recommendation scheme corresponding to a live broadcast client; A combination module, configured to determine a target recommendation scheme corresponding to the live broadcast client based on a combination result formed by combining each of the target recommendation dimensions according to the target combination scheme; The combined module is specifically used for: According to the multiple target combination relationships in the target combination scheme, each of the target recommendation dimensions and the corresponding target recommendation parameters are combined to determine at least one live broadcast screening scheme; Determine the live broadcast push scheme corresponding to each of the live broadcast screening schemes according to the received third circle selection instruction, wherein the live broadcast push scheme is used to sort and push the live broadcast data screened based on the corresponding live broadcast screening scheme; Determine a target recommendation scheme corresponding to the live broadcast client according to the live broadcast screening scheme and the live broadcast push scheme; The combined module is specifically used for: Generate corresponding live broadcast candidate solutions according to each of the live broadcast screening solutions and the corresponding live broadcast push solutions; Determining, according to the received scheme allocation instruction, the corresponding relationship between each page in the live broadcast client and each of the live broadcast candidate schemes; Generate a target recommendation solution corresponding to the live broadcast client according to the live broadcast candidate solutions that have a corresponding relationship with each page in the live broadcast client; The combined module is specifically used for: Determine, according to the received scheme allocation instruction, the correspondence between different blocks in the target page in the live client and each of the live candidate schemes, wherein the target page is any page in the live client; The live broadcast candidate solutions that have a corresponding relationship with each block in the target page form a first solution set, and the corresponding relationship between the target page and the first solution set is determined.
6. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.
