Content recommendation method and device, storage medium and computing equipment

Abstract

The embodiment of the invention provides a content recommendation method and device, a storage medium and computing equipment. Comprising the following steps: in response to an access request of a target service, obtaining service logics of a plurality of execution stages abstracted from general recommendation logics; configuring a configuration item corresponding to the service logic of each execution stage, and generating a content recommendation logic for the target service according to the configured service logic of each execution stage; and based on the generated content recommendation logic, performing content recommendation on a recommendation request generated in the target service.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and more specifically, to a content recommendation method, apparatus, storage medium, and computing device. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the present disclosure. No description herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] As the content on content platforms increases, users are faced with a huge amount of information. How to quickly and accurately obtain the content they are interested in has become an important issue. Content recommendation is a solution. Content platforms can actively recommend content of interest to users, thereby helping users quickly obtain the content they need.

[0004] In related technologies, content recommendation can usually provide users with personalized content recommendation services by analyzing users' historical behaviors.

[0005] However, due to the influence of market segments, content platforms continue to launch various segmented business scenarios in the content platform according to the characteristics of the newly emerging market segments to meet the content consumption needs of various segmented users. Taking the music platform as an example, there are usually entrances to various different segmented business scenarios on the homepage, such as business scenario entrances for different content types (such as radio, listening to books, listening to songs, MVs, etc.), business scenario entrances for different music styles (such as Chinese style, European and American style, etc.), other business scenario entrances (such as daily promotion, roaming, sports, etc.), etc.

[0006] Since the business of each business scenario is different, the recommendation logic for content recommendation is also different. Therefore, whenever a new business scenario is introduced, it is necessary to develop a separate set of content recommendation logic that adapts to the business scenario to meet the content recommendation needs in the new business scenario. Summary of the invention

[0007] In a first aspect of the embodiments of the present disclosure, a content recommendation method is provided. The method comprises:

[0008] In response to an access request of a target business, obtain business logic of several execution stages abstracted from the general recommendation logic;

[0009] Configure the configuration items corresponding to the business logic of each execution stage, and generate content recommendation logic for the target business according to the configured business logic of each execution stage;

[0010] Based on the generated content recommendation logic, content recommendation is performed for the recommendation request generated in the target business.

[0011] Optionally, the business logic of the several execution stages includes at least one of preprocessing logic, recall logic, and sorting logic.

[0012] Optionally, the pre-processing logic includes execution logic related to parsing the recommendation request, filtering the recommendation request based on the filtering policy configured in the configuration item, and pulling user behavior data;

[0013] The configuration items corresponding to the business logic of each execution stage include:

[0014] In response to a target general filtering policy and / or a newly added custom filtering policy selected from the general filtering policies, the target general filtering policy and / or the custom filtering policy configuration is configured in a configuration item corresponding to the preprocessing logic.

[0015] Optionally, the recall logic includes an execution logic for determining candidate recommended content based on a recall strategy configured in the configuration item;

[0016] The configuration items corresponding to the business logic of each execution stage include:

[0017] In response to a target recall strategy selected from the general recall strategies, the target recall strategy is configured in a configuration item corresponding to the recall logic.

[0018] Optionally, the general recall strategy includes at least one of the following:

[0019] Recall strategies for content recall from premium content,

[0020] Recall strategies for content recall based on content type,

[0021] Recall strategy for content recall based on users’ long-term and short-term behavior data;

[0022] A recall strategy for content recall based on a preset recall model.

[0023] Optionally, the sorting logic includes execution logic related to a coarse sorting strategy and a fine sorting strategy configured in the configuration item; wherein the coarse sorting strategy is used to perform coarse-grained sorting on the recall results, and the fine sorting strategy is used to perform fine-grained sorting on a preset number of recall results in the coarse sorting results;

[0024] The configuration items corresponding to the business logic of each execution stage include:

[0025] In response to a first sorting model for a coarse sorting strategy and a second sorting model for a fine sorting strategy selected from a general sorting model, the first sorting model and the second sorting model are configured in a configuration item corresponding to the sorting logic.

[0026] Optionally, the method is applied to a business logic layer of a universal recommendation system, and the universal recommendation system further includes a business access layer and a data docking layer; wherein the business access layer is used to access a number of target businesses, and the data docking layer is used to dock a number of data storage objects storing content;

[0027] The responding to the access request of the target service includes:

[0028] Responding to an access request initiated by a target service accessed in the service access layer;

[0029] The step of performing content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic includes:

[0030] Based on the generated content recommendation logic, the data storage object in the data docking layer performs content recommendation for the recommendation request generated in the target business.

[0031] Optionally, access protocols of various target services in the service access layer are consistent, and docking protocols of various data storage objects in the data docking layer are consistent.

[0032] In a second aspect of the embodiments of the present disclosure, a content recommendation device is provided, the device comprising:

[0033] An acquisition unit, in response to an access request of a target service, acquires service logic of several execution stages abstracted from the general recommendation logic;

[0034] A configuration unit, which configures configuration items corresponding to the business logic of each execution stage, and generates content recommendation logic for the target business according to the configured business logic of each execution stage;

[0035] The recommendation unit performs content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic.

[0036] Optionally, the business logic of the several execution stages includes at least one of preprocessing logic, recall logic, and sorting logic.

[0037] Optionally, the pre-processing logic includes execution logic related to parsing the recommendation request, filtering the recommendation request based on the filtering policy configured in the configuration item, and pulling user behavior data;

[0038] The configuration unit is further used to configure the target general filtering policy and / or the newly added custom filtering policy in the configuration item corresponding to the preprocessing logic in response to the target general filtering policy and / or the newly added custom filtering policy selected from the general filtering policies.

[0039] Optionally, the recall logic includes an execution logic for determining candidate recommended content based on a recall strategy configured in the configuration item;

[0040] The configuration unit is further configured to configure the target recall strategy in the configuration item corresponding to the recall logic in response to the target recall strategy selected from the general recall strategy.

[0041] Optionally, the general recall strategy includes at least one of the following:

[0042] Recall strategies for content recall from premium content,

[0043] Recall strategies for content recall based on content type,

[0044] Recall strategy for content recall based on users’ long-term and short-term behavior data;

[0045] A recall strategy for content recall based on a preset recall model.

[0046] Optionally, the sorting logic includes execution logic related to a coarse sorting strategy and a fine sorting strategy configured in the configuration item; wherein the coarse sorting strategy is used to perform coarse-grained sorting on the recall results, and the fine sorting strategy is used to perform fine-grained sorting on a preset number of recall results in the coarse sorting results;

[0047] The configuration unit is further used to configure the first sorting model and the second sorting model in the configuration items corresponding to the sorting logic in response to a first sorting model for a coarse sorting strategy and a second sorting model for a fine sorting strategy selected from the general sorting model.

[0048] Optionally, the device is applied to a business logic layer of a general recommendation system, and the general recommendation system further includes a business access layer and a data docking layer; wherein the business access layer is used to access a number of target businesses, and the data docking layer is used to dock with a number of data storage objects storing content;

[0049] The acquisition unit is further configured to respond to an access request initiated by a target service accessed in the service access layer;

[0050] The recommendation unit is further configured to perform content recommendation for the recommendation request generated in the target business by the data storage object in the data docking layer based on the generated content recommendation logic.

[0051] Optionally, access protocols of various target services in the service access layer are consistent, and docking protocols of various data storage objects in the data docking layer are consistent.

[0052] In a third aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, comprising:

[0053] When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the content recommendation method as described in any of the preceding items.

[0054] In a fourth aspect of the embodiments of the present disclosure, there is provided a computing device, comprising:

[0055] processor;

[0056] a memory for storing instructions executable by the processor;

[0057] The processor is configured to execute the executable instructions to implement the content recommendation method as described in any of the preceding items.

[0058] According to the content recommendation solution provided by the embodiment of the present disclosure, by abstracting the business logic of several execution stages from the general recommendation logic, and then when accessing the target business, by configuring the configuration items corresponding to the business logic of each execution stage, the configured business logic of each execution stage is combined into the content recommendation logic of the target business like "building blocks". On the one hand, the development of content recommendation logic can be achieved quickly and at low cost through configuration; on the other hand, differentiated capabilities of different business scenarios can be achieved by modifying the configuration items. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:

[0060] Figure 1 The system architecture diagram of the content recommendation system provided by the present disclosure is schematically shown;

[0061] Figure 2 The following schematically shows a content recommendation method provided by the present disclosure;

[0062] Figure 3 A schematic diagram schematically shows a music service accessed by a service access layer provided by the present disclosure;

[0063] Figure 4 The schematic diagram of the pre-processing logic provided by the present disclosure is schematically shown;

[0064] Figure 5 A schematic diagram of the recall logic provided by the present disclosure is schematically shown;

[0065] Figure 6The schematic diagram of the sorting logic provided by the present disclosure is schematically shown;

[0066] Figure 7 The medium schematic diagram provided by the present disclosure is schematically shown;

[0067] Figure 8 The following schematically shows a content recommendation device provided by the present disclosure;

[0068] Fig. 9 The schematic diagram of the computing device provided by the present disclosure is schematically shown.

[0069] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0070] The principles and spirit of the present disclosure will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present disclosure, and are not intended to limit the scope of the present disclosure in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0071] Those skilled in the art will appreciate that the embodiments of the present disclosure may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0072] According to an embodiment of the present disclosure, a content recommendation method, a computer-readable storage medium, an apparatus, and a computing device are proposed.

[0073] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0074] The principle and spirit of the present disclosure are explained in detail below with reference to several representative embodiments of the present disclosure.

[0075] The data involved in the present disclosure may be data authorized by the user or fully authorized by all parties. The collection, dissemination, and use of the data shall comply with the requirements of relevant national laws and regulations. The implementation methods / examples of the present disclosure may be combined with each other. SUMMARY OF THE INVENTION

[0077] The present disclosure aims to provide a content recommendation solution, which abstracts the business logic of several execution stages from the general recommendation logic, and then when accessing the target business, by configuring the configuration items corresponding to the business logic of each execution stage, the configured business logic of each execution stage is combined into the content recommendation logic of the target business like "building blocks". On the one hand, the development of content recommendation logic can be achieved quickly and at low cost through configuration; on the other hand, differentiated capabilities of different business scenarios can be achieved by modifying the configuration items.

[0078] After introducing the basic principles of the present disclosure, various non-limiting embodiments of the present disclosure are described in detail below.

[0079] Application Scenario Overview

[0080] First reference Figure 1 A system architecture diagram of a content recommendation system is shown. In the system architecture diagram, the content recommendation system may include a service access layer, a service logic layer, and a data connection layer.

[0081] Among them, the service access layer is used to access various services in the content platform, such as Figure 1 The services 1, 2, 3, ..., and n are shown in FIG. These services may be old service scenarios that already exist in the content platform, new service scenarios that appear in the content platform, service scenarios that will be opened in the future content platform, or service test scenarios for content testing, etc. As long as there is a service for content recommendation, it can be added to the service access layer, so that the corresponding content recommendation logic can be quickly developed through the embodiments provided in this specification.

[0082] The business logic layer is used to develop content recommendation logic for access services. By abstracting the business logic of several execution stages from the general recommendation logic, these business logic of several execution stages are deployed as general modules in the business logic layer, for example Figure 1 The pre-processing logic, recall logic, sorting logic and / or post-processing logic shown in . When accessing the target business, by configuring the configuration items corresponding to the pre-processing logic, recall logic, sorting logic and / or post-processing logic, the content recommendation logic of the target business can be obtained by combining the configured pre-processing logic, recall logic, sorting logic and / or post-processing logic like "building blocks".

[0083] The data docking layer is used to dock with various data storage objects that provide content for the content platform, such as Figure 1 The data storage object 1, data storage object 2, data storage object 3, ..., data storage object m are shown in FIG. Each data storage object can store a plurality of contents.

[0084] After the business logic layer in the content recommendation system generates the content recommendation logic for the target business, the data storage object in the data docking layer can perform content recommendation for the recommendation request generated in the target business.

[0085] Exemplary Methods

[0086] Combine the following Figure 1 The application scenario shown is shown in Figure 2 To describe the method of content recommendation according to the exemplary embodiment of the present disclosure. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principle of the present disclosure, and the embodiments of the present disclosure are not limited in this respect. On the contrary, the embodiments of the present disclosure can be applied to any applicable scenario.

[0087] like Figure 2 As shown, the content recommendation method may include the following steps:

[0088] Step 210: In response to the access request of the target service, obtain the service logic of several execution stages abstracted from the general recommendation logic;

[0089] Step 220: configuring the configuration items corresponding to the business logic of each execution stage, and generating content recommendation logic for the target business according to the configured business logic of each execution stage;

[0090] Step 230: Based on the generated content recommendation logic, content recommendation is performed for the recommendation request generated in the target business.

[0091] In an exemplary embodiment, the method can be applied to the business logic layer of the aforementioned universal recommendation system, and the universal recommendation system can also include a business access layer and a data docking layer; wherein the business access layer is used to access a number of target businesses, and the data docking layer is used to dock with a number of data storage objects storing content;

[0092] Accordingly, step 210, when responding to the access request of the target service, may further include:

[0093] Responding to an access request initiated by a target service accessed in the service access layer;

[0094] In addition, when performing content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic, step 230 may further include:

[0095] Based on the generated content recommendation logic, the data storage object in the data docking layer performs content recommendation for the recommendation request generated in the target business.

[0096] In this embodiment, the target business in the content platform initiates an access request to join the access business layer, thereby triggering the business logic layer to configure the configuration items corresponding to the business logic of several execution stages, and further combining the configured business logic of several execution stages to obtain the content recommendation logic of the target business. Then, after the business logic layer generates the content recommendation logic of the target business, the data storage object in the data docking layer can perform content recommendation for the recommendation request generated in the target business.

[0097] Please refer to the following Figure 3 The schematic diagram of the music service accessed by the service access layer shown in the figure; in the music platform, the recommendation form can be divided into list type and stream type, and different services are provided under different pages (such as the cover of the homepage, the details page of the song, etc.), such as radio station recommendation, daily recommendation, roaming and other services on the cover, style daily recommendation, original song list recommendation, etc. on the details page. After different services are connected to the content recommendation system, the content recommendation logic development corresponding to each service can be completed quickly and at low cost through the embodiments provided in this specification, and content recommendation can be performed for each service based on the content recommendation logic corresponding to each service.

[0098] To summarize, by abstracting the business logic of several execution stages from the general recommendation logic, the standardization of the content recommendation logic process is completed; and as long as the access protocols of various target businesses accessing the business access layer are consistent, and the docking protocols of various data storage objects in the data docking layer are consistent, it can be ensured that a set of general recommendation logic processes can be used regardless of the number of business scenarios to realize a general song recommendation system.

[0099] Regarding the business logic of several execution stages abstracted from the general recommendation logic in this specification, as mentioned above Figure 1 As shown in , the business logic of several execution stages may include at least one of pre-processing logic, recall logic, and sorting logic, or at least one of pre-processing logic, recall logic, sorting logic, and post-processing logic.

[0100] These logics are introduced below in an exemplary manner.

[0101] [Preprocessing logic]

[0102] In an exemplary embodiment, the pre-processing logic includes execution logic related to parsing the recommendation request, filtering the recommendation request based on the filtering policy configured in the configuration item, and pulling user behavior data;

[0103] The configuration items corresponding to the business logic of each execution stage include:

[0104] In response to a target general filtering policy and / or a newly added custom filtering policy selected from the general filtering policies, the target general filtering policy and / or the custom filtering policy configuration is configured in a configuration item corresponding to the preprocessing logic.

[0105] In this example, the process of the preprocessing logic may include parsing a standard format recommendation request initiated by the target business accessed in the business access layer to obtain the recommended user account or identifier, the business identifier of the target business, etc., and may also pull the user's real-time or offline behavior data based on the parsed user account or identifier, and finally assemble the parsed data into user data for subsequent content recommendation; for example, the user's content preferences can be determined based on the behavior data, and then the recommended content that meets the content preferences can be determined.

[0106] Please combine the following Figure 4 The schematic diagram of the preprocessing logic is shown in Figure 4 As shown in the figure, the preprocessing logic also includes filtering configuration items, and is configured with several common filtering configuration strategies; taking the music platform as an example, common filtering strategies may include blacklist, offline log, real-time log, interface cache, scene specialization, trash can, red heart, etc. In addition, the filtering configuration items also support the addition of custom filtering strategies. Regardless of whether it is a common filtering strategy or a custom filtering strategy, it is necessary to ensure that these filtering strategies are registered and used in accordance with the defined unified strategy format.

[0107] By configuring the filtering strategy of the preprocessing logic, it is possible to filter out non-compliant recommendation requests, that is, to avoid filtering non-compliant recommendation requests (for example, adding user accounts that have been manipulated to the blacklist, thereby filtering these user accounts that frequently search for specific content in order to improve the ranking of the specific content) and to reduce the pressure on the content recommendation system.

[0108] [Recall Logic]

[0109] In an exemplary embodiment, the recall logic includes execution logic for determining candidate recommended content based on a recall policy configured in the configuration item;

[0110] The configuration items corresponding to the business logic of each execution stage include:

[0111] In response to a target recall strategy selected from the general recall strategies, the target recall strategy is configured in a configuration item corresponding to the recall logic.

[0112] In this example, the recall phase can roughly select several candidate recommended contents for the user from the massive content according to the user's content preferences as shown above, so that the subsequent sorting phase can sort these candidate recommended contents.

[0113] Please combine the following Figure 5 The schematic diagram of the recall logic is shown in Figure 5 As shown, the recall logic may include recall configuration items and be configured with several common recall strategies.

[0114] Exemplarily, the general recall strategy includes at least one of the following:

[0115] Recall strategies for content recall from premium content,

[0116] Recall strategies for content recall based on content type,

[0117] Recall strategy for content recall based on users’ long-term and short-term behavior data;

[0118] A recall strategy for content recall based on a preset recall model.

[0119] Similar to the aforementioned filtering policy, the recall configuration item can also support the addition of custom recall policies. Regardless of whether it is a general recall policy or a custom recall policy, it is necessary to ensure that these recall policies are registered, put into use, and shut down in accordance with the defined unified policy format.

[0120] If multiple (two or more) recall strategies are configured in the recall configuration item, the content can be recalled for each recall strategy, and the content recalled by each recall strategy can be aggregated to obtain the final recall result. In addition, the source of each recalled content can be marked (specifically which recall strategy it comes from), so as to facilitate the evaluation of the performance (accuracy, diversity, etc.) of the recall strategy. If the performance does not meet expectations, the strategy can be modified or offline.

[0121] By configuring the recall strategy of the recall logic, accurate recall of candidate recommendation content can be achieved, and the diversity of candidate recommendation content can be achieved, thereby improving the effect of content recommendation.

[0122] [Sorting logic]

[0123] In an exemplary embodiment, the sorting logic includes execution logic related to a coarse sorting strategy and a fine sorting strategy configured in the configuration item; wherein the coarse sorting strategy is used to perform coarse-grained sorting on the recall results, and the fine sorting strategy is used to perform fine-grained sorting on a preset number of recall results in the coarse sorting results;

[0124] The configuration items corresponding to the business logic of each execution stage include:

[0125] In response to a first sorting model for a coarse sorting strategy and a second sorting model for a fine sorting strategy selected from a general sorting model, the first sorting model and the second sorting model are configured in a configuration item corresponding to the sorting logic.

[0126] In this example, the sorting stage can be used to sort the candidate recommendation contents recalled in the recall stage according to actual business needs to obtain the final recommendation results.

[0127] Please combine the following Figure 6 The schematic diagram of the sorting logic is shown in Figure 6 As shown, the sorting logic may include sorting configuration items. Sorting generally includes two processes: rough sorting and fine sorting. Therefore, the sorting configuration items further include configuration items of the rough sorting strategy and the fine sorting strategy.

[0128] Among them, the coarse sorting strategy can solve the coarse-grained sorting from the recall results (such as the sorting selection from 5k to 1k), and the fine sorting strategy can solve the fine-grained sorting (such as the sorting selection of 1k recall results). Although the coarse sorting and fine sorting have different focuses, the sorting logic is the same. Figure 6 As shown, several general ranking models can be configured, such as TF-IDF (term frequency-inverse document frequency) model, word embedding model, etc. During configuration, according to actual business needs, the first ranking model is configured for the rough ranking strategy from these general ranking models, and the second ranking model is configured for the fine ranking strategy.

[0129] In addition, the sorting configuration items can also support the addition of custom sorting models. Through the configuration management of model name + version number, the model can be quickly registered and managed online and offline.

[0130] By configuring the sorting strategy of the sorting logic, accurate sorting of candidate recommendation content can be achieved, thereby improving the effect of content recommendation.

[0131] [Post-processing logic]

[0132] The post-processing logic includes execution logic of a rearrangement strategy and a diversity strategy.

[0133] In this example, the reordering strategy is used to reorder the sorting results. For example, after receiving the recommended content, if the recommended user is not satisfied, he or she will usually refresh the results in order to obtain new recommended content. At this time, the post-processing logic can reorder the sorting results according to the reordering strategy, for example, remove the first preset number of contents in the sorting results of the aforementioned sorting stage and reorder the remaining contents, so as to recommend the new sorting results to the user. Alternatively, the reordering results are classified according to content type according to the diversity strategy and recommended to the user.

[0134] Since the content in the existing sorting results is of interest to users, but the content in the front row that is displayed to users first does not meet their expectations, by configuring the post-processing logic, there is no need to recall the content again, and the existing sorting results are processed again to provide users with different recommendation results. In this way, the computing resources in the recall and sorting stages can be reduced.

[0135] To summarize, according to any of the above-mentioned content recommendation embodiments, by abstracting the business logic of several execution stages from the general recommendation logic, and then when accessing the target business, by configuring the configuration items corresponding to the business logic of each execution stage, the configured business logic of each execution stage is combined into the content recommendation logic of the target business just like "building blocks".

[0136] On the one hand, through a configurable approach, fast and low-cost content recommendation logic development can be achieved; the time required for traditional content recommendation logic development is reduced from days to minutes, greatly improving business iteration efficiency.

[0137] On the other hand, deploying common policies in the configuration items of each execution stage and managing policies dynamically through configuration-policy mapping greatly improves the universality of policies in each execution stage.

[0138] On the other hand, by defining a unified policy format, whether you choose a general policy or add a new custom policy, you can achieve differentiated capabilities for different business scenarios.

[0139] Exemplary Media

[0140] After introducing the method of the exemplary embodiment of the present disclosure, next, refer to Figure 7 A medium according to an exemplary embodiment of the present disclosure is described.

[0141] In this exemplary embodiment, the above method can be implemented by a program product, such as a portable compact disk read-only memory (CD-ROM) and including program code, and can be run on a device, such as a personal computer. However, the program product of the present disclosure is not limited thereto, and in this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, apparatus, or device.

[0142] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0143] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0144] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RE, etc., or any suitable combination of the foregoing.

[0145] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user computing device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0146] In summary, the present disclosure may provide a computer-readable storage medium. When instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device may be enabled to execute the aforementioned content recommendation method embodiment.

[0147] Exemplary Devices

[0148] After introducing the medium of the exemplary embodiment of the present disclosure, next, reference is made to Figure 8A device according to an exemplary embodiment of the present disclosure is described.

[0149] Figure 8 The block diagram of a content recommendation device according to an embodiment of the present disclosure is schematically shown, corresponding to the aforementioned Figure 2 The content recommendation device may include:

[0150] The acquisition unit 710 acquires the business logic of several execution stages abstracted from the general recommendation logic in response to the access request of the target business;

[0151] The configuration unit 720 configures the configuration items corresponding to the business logic of each execution stage, and generates content recommendation logic for the target business according to the configured business logic of each execution stage;

[0152] The recommendation unit 730 performs content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic.

[0153] Optionally, the business logic of the several execution stages includes at least one of preprocessing logic, recall logic, and sorting logic.

[0154] Optionally, the pre-processing logic includes execution logic related to parsing the recommendation request, filtering the recommendation request based on the filtering policy configured in the configuration item, and pulling user behavior data;

[0155] The configuration unit 720 is further configured to configure the target general filtering policy and / or the newly added custom filtering policy in the configuration item corresponding to the preprocessing logic in response to the target general filtering policy and / or the newly added custom filtering policy selected from the general filtering policies.

[0156] Optionally, the recall logic includes an execution logic for determining candidate recommended content based on a recall strategy configured in the configuration item;

[0157] The configuration unit 720 is further configured to configure the target recall strategy in the configuration item corresponding to the recall logic in response to the target recall strategy selected from the general recall strategy.

[0158] Optionally, the general recall strategy includes at least one of the following:

[0159] Recall strategies for content recall from premium content,

[0160] Recall strategies for content recall based on content type,

[0161] Recall strategy for content recall based on users’ long-term and short-term behavior data;

[0162] A recall strategy for content recall based on a preset recall model.

[0163] Optionally, the sorting logic includes execution logic related to a coarse sorting strategy and a fine sorting strategy configured in the configuration item; wherein the coarse sorting strategy is used to perform coarse-grained sorting on the recall results, and the fine sorting strategy is used to perform fine-grained sorting on a preset number of recall results in the coarse sorting results;

[0164] The configuration unit 720 is further configured to configure the first sorting model and the second sorting model in the configuration items corresponding to the sorting logic in response to a first sorting model for a coarse sorting strategy and a second sorting model for a fine sorting strategy selected from the general sorting model.

[0165] Optionally, the device is applied to a business logic layer of a general recommendation system, and the general recommendation system further includes a business access layer and a data docking layer; wherein the business access layer is used to access a number of target businesses, and the data docking layer is used to dock with a number of data storage objects storing content;

[0166] The acquisition unit 710 is further configured to respond to an access request initiated by a target service accessed in the service access layer;

[0167] The recommendation unit 730 is further configured to perform content recommendation for the recommendation request generated in the target business by the data storage object in the data docking layer based on the generated content recommendation logic.

[0168] Optionally, access protocols of various target services in the service access layer are consistent, and docking protocols of various data storage objects in the data docking layer are consistent.

[0169] Exemplary Computing Devices

[0170] After introducing the method, medium and apparatus of the exemplary embodiments of the present disclosure, next, reference is made to Fig. 9 A computing device according to an exemplary embodiment of the present disclosure is described.

[0171] Fig. 9 The computing device 1500 shown is merely an example and should not bring any limitation to the functionality and scope of use of the embodiments of the present disclosure.

[0172] like Fig. 9 As shown, computing device 1500 is in the form of a general computing device. Components of computing device 1500 may include, but are not limited to: at least one processing unit 1501, at least one storage unit 1502, and a bus 1503 connecting different system components (including processing unit 1501 and storage unit 1502).

[0173] The bus 1503 includes a data bus, a control bus, and an address bus.

[0174] The storage unit 1502 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 15021 and / or a cache memory 15022 , and may further include a readable medium in the form of a non-volatile memory, such as a read-only memory (ROM) 15023 .

[0175] The storage unit 1502 may also include a program / utility 15025 having a set (at least one) of program modules 15024, such program modules 15024 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0176] The computing device 1500 may also communicate with one or more external devices 1504 (eg, a keyboard, pointing device, etc.).

[0177] Such communication may be performed via input / output (I / O) interface 1505. In addition, computing device 1500 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via network adapter 1506. Fig. 9 As shown, network adapter 1506 communicates with other modules of computing device 1500 via bus 1503. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with computing device 1500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0178] Through Fig. 9 The computing device 1500 shown can implement the aforementioned content recommendation method. More specifically, the storage unit 1502 stores instructions executable by the processing unit 1501. When the processing unit 1501 executes the instructions, the aforementioned content recommendation method is implemented.

[0179] It should be noted that although several units / modules or sub-units / modules of the content recommendation device are mentioned in the above detailed description, such division is merely exemplary and not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more units / modules described above may be embodied in one unit / module. Conversely, the features and functions of one unit / module described above may be further divided to be embodied by multiple units / modules.

[0180] In addition, although the operations of the disclosed method are described in a specific order in the drawings, this does not require or imply that the operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0181] Although the spirit and principle of the present disclosure have been described with reference to several specific embodiments, it should be understood that the present disclosure is not limited to the disclosed specific embodiments, and the division of various aspects does not mean that the features in these aspects cannot be combined to benefit, and such division is only for the convenience of expression. The present disclosure is intended to cover various modifications and equivalent arrangements included in the spirit and scope of the attached claims.

Claims

1. A content recommendation method, comprising: In response to an access request of a target business, obtain business logic of several execution stages abstracted from the general recommendation logic; Configure the configuration items corresponding to the business logic of each execution stage, and generate content recommendation logic for the target business according to the configured business logic of each execution stage; Based on the generated content recommendation logic, content recommendation is performed for the recommendation request generated in the target business.

2. According to the method of claim 1, the business logic of the several execution stages includes at least one of pre-processing logic, recall logic, and sorting logic.

3. The method according to claim 2, wherein the pre-processing logic includes execution logic related to parsing the recommendation request, filtering the recommendation request based on the filtering policy configured in the configuration item, and pulling the user behavior data; The configuration items corresponding to the business logic of each execution stage include: In response to a target general filtering policy and / or a newly added custom filtering policy selected from the general filtering policies, the target general filtering policy and / or the custom filtering policy configuration is configured in a configuration item corresponding to the preprocessing logic.

4. The method according to claim 2, wherein the recall logic comprises an execution logic for determining candidate recommended content based on a recall strategy configured in a configuration item; The configuration items corresponding to the business logic of each execution stage include: In response to a target recall strategy selected from the general recall strategies, the target recall strategy is configured in a configuration item corresponding to the recall logic.

5. The method according to claim 4, wherein the general recall strategy comprises at least one of the following: Recall strategies for content recall from premium content, Recall strategies for content recall based on content type, Recall strategy for content recall based on users’ long-term and short-term behavior data; A recall strategy for content recall based on a preset recall model.

6. The method according to claim 2, wherein the sorting logic includes execution logic related to the coarse sorting strategy and the fine sorting strategy configured in the configuration item; wherein: The coarse sorting strategy is used to perform coarse-grained sorting on the recall results, and the fine sorting strategy is used to perform fine-grained sorting on a preset number of recall results in the coarse sorting results; The configuration items corresponding to the business logic of each execution stage include: In response to a first sorting model for a coarse sorting strategy and a second sorting model for a fine sorting strategy selected from a general sorting model, the first sorting model and the second sorting model are configured in a configuration item corresponding to the sorting logic.

7. The method according to claim 1 is applied to the business logic layer of a general recommendation system, wherein the general recommendation system further comprises a business access layer and a data docking layer; wherein: The service access layer is used to access a number of target services, and the data docking layer is used to dock with a number of data storage objects storing content; The responding to the access request of the target service includes: Responding to an access request initiated by a target service accessed in the service access layer; The step of performing content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic includes: Based on the generated content recommendation logic, the data storage object in the data docking layer performs content recommendation for the recommendation request generated in the target business.

8. A content recommendation device, comprising: An acquisition unit, in response to an access request of a target service, acquires service logic of several execution stages abstracted from the general recommendation logic; A configuration unit, which configures configuration items corresponding to the business logic of each execution stage, and generates content recommendation logic for the target business according to the configured business logic of each execution stage; The recommendation unit performs content recommendation for the recommendation request generated in the target business based on the generated content recommendation logic.

9. A computer-readable storage medium, comprising: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the content recommendation method as described in any one of claims 1 to 7.

10. A computing device comprising: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the executable instructions to implement the content recommendation method as described in any one of claims 1-7.

Images