Content recommendation method and apparatus

Abstract

The application discloses a content recommendation method and device. The method comprises the following steps: obtaining first data of a target object, wherein the first data at least comprises historical viewing content data and historical use behavior data; determining a preference feature of the target object according to the first data, wherein the preference feature comprises a first preference feature for reflecting temporary preference of the target object and a second preference feature for reflecting periodic preference of the target object; obtaining to-be-recommended content, and determining a first content in the to-be-recommended content based on the preference feature, wherein the first content is content that the target object is no longer interested in at a target period; and filtering the first content in the to-be-recommended content to obtain first target recommendation content. The application solves the technical problem that related technologies cannot effectively filter recommended content that a user is no longer interested in, resulting in low recommendation accuracy and poor user experience.

Description

Technical Field

[0001] This application relates to the field of content recommendation technology, and more specifically, to a content recommendation method and apparatus. Background Technology

[0002] In the field of content recommendation, content recommendation algorithms are the most crucial component of a recommendation system. Currently, to reduce duplicate recommendations, recommendation systems typically filter or block duplicate content based on customer reviews and comments made while using the product.

[0003] However, the above filtering methods are not suitable for recommendation systems with little or no interaction, such as telecom iTV products. Due to the inconvenience of remote control operation, customers rarely evaluate the content recommended by telecom iTV. Therefore, the recommendation system has difficulty filtering out temporary favorites and content that is no longer of interest to the customer, resulting in a large amount of repetitive content being recommended to the user. This leads to low accuracy of the recommended content and a poor user experience.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a content recommendation method and apparatus to at least solve the technical problem that related technologies cannot effectively filter recommended content that users are no longer interested in, resulting in low recommendation accuracy and poor user experience.

[0006] According to one aspect of the embodiments of this application, a content recommendation method is provided, comprising: acquiring first data of a target object, wherein the first data includes at least: historical viewed content data and historical usage behavior data; determining the target object's preference features based on the first data, wherein the preference features include: a first preference feature reflecting the target object's temporary preferences and a second preference feature reflecting the target object's periodic preferences; acquiring content to be recommended, and determining a first content among the content to be recommended based on the preference features, wherein the first content is content that the target object is no longer interested in during a target time period; filtering the first content among the content to be recommended to obtain a first target recommended content.

[0007] Optionally, determining the target object's preference characteristics based on the first data includes: determining the target object's first preference characteristics based on the first data, wherein the first preference characteristics include at least one of the following: temporary preference content, temporary preference type; and determining the target object's second preference characteristics based on the first data, wherein the second preference characteristics include at least one of the following: preference cycle, cycle preference content, cycle preference type.

[0008] Optionally, the first content includes: a first sub-content and a second sub-content. Determining the first content in the content to be recommended based on preference features includes: determining the content in the content to be recommended that matches the first preference feature as the first sub-content; and determining the content in the content to be recommended that matches the second preference feature as the second sub-content.

[0009] Optionally, determining the content in the recommended content that matches the second preference feature as the second sub-content includes: estimating the access data of the target audience for each content in the recommended content within the target time period based on the second preference feature, wherein the access data includes at least one of the following: number of accesses, access duration, and access percentage; determining the recommendation score for each content based on the access data; and determining the content with a recommendation score lower than a preset threshold as the second sub-content.

[0010] Optionally, the historical viewing content data is classified to obtain multiple content classification features; content association features are determined based on the correlation between the various content classification features; and second content that is associated with the first content in the first target recommended content is determined based on the content association features.

[0011] Optionally, the second content includes: a third sub-content and a fourth sub-content. The second content in the first target recommended content that is related to the first content is determined based on content association features, including: a third sub-content in the first target recommended content that is related to the first sub-content; and a fourth sub-content in the first target recommended content that is related to the second sub-content.

[0012] Optionally, the content to be recommended can be obtained, including: determining the content to be recommended based on the target audience's historical viewing data.

[0013] According to another aspect of the embodiments of this application, a content recommendation apparatus is also provided, comprising: an acquisition module, configured to acquire first data of a target object, wherein the first data includes at least: historical viewed content data and historical usage behavior data; a first determination module, configured to determine the target object's preference features based on the first data, wherein the preference features include: a first preference feature reflecting the target object's temporary preferences and a second preference feature reflecting the target object's periodic preferences; a second determination module, configured to acquire content to be recommended and determine a first content among the content to be recommended based on the preference features, wherein the first content is content that the target object is no longer interested in during a target time period; and a filtering module, configured to filter the first content among the content to be recommended to obtain a first target recommended content.

[0014] According to another aspect of the embodiments of this application, a processor is also provided, which is used to run a program, wherein the program executes the above-described content recommendation method during runtime.

[0015] According to another aspect of the embodiments of this application, an electronic device is also provided, the electronic device including: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-described content recommendation method through the computer program.

[0016] In this embodiment, firstly, first data of the target object is obtained, including at least historical viewing content data and historical usage behavior data. Then, based on the first data, the target object's preference characteristics are determined, including a first preference characteristic reflecting the target object's temporary preferences and a second preference characteristic reflecting the target object's periodic preferences. Next, content to be recommended is obtained, and based on the preference characteristics, a first content within the content to be recommended is determined, wherein the first content is content that the target object is no longer interested in during the target time period. Finally, the first content in the content to be recommended is filtered to obtain the first target recommended content. By analyzing the target object's temporary and periodic preferences and filtering the corresponding first and second preference characteristics, the recommendation of content that the target object is no longer interested in is effectively reduced, improving the accuracy of the recommended content and achieving the goal of enhancing the competitiveness of the recommendation system product. This solves the technical problem that related technologies cannot effectively filter recommended content that users are no longer interested in, resulting in low recommendation accuracy and a poor user experience. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a flowchart illustrating a content recommendation method according to an embodiment of this application;

[0019] Figure 2 This is a flowchart illustrating an optional content recommendation method according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of the structure of a content recommendation device according to an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Example 1

[0024] Currently, in the field of content recommendation, to reduce duplicate recommendations, recommendation systems typically filter or block duplicate content based on customer comments and actions while using the product. However, this filtering method is not suitable for recommendation systems with little or no interaction. For example, with telecom iTV products, due to the inconvenience of remote control operation, customers rarely comment on the content recommended. Therefore, the recommendation system struggles to filter out temporarily liked content and content that is no longer of interest, leading to a large amount of duplicate content being recommended to users and resulting in a poor user experience.

[0025] To address the aforementioned issues, this application provides a content recommendation method embodiment. First, it analyzes a customer's historical viewing data, analyzing their periodic and temporary preference characteristics based on historical usage behavior data such as time, day, week, month, and holidays. Then, it filters all content matching the temporary preference characteristics from the acquired content to be recommended, and further filters content with a recommendation score lower than a system preset threshold, obtaining the target recommended content. This process performs the first filtering of the recommended content. To avoid the system repeatedly recommending content that the target audience is no longer interested in, this application embodiment further performs a second filtering of the target recommended content based on the above content recommendation method, effectively reducing the repeated recommendation of content that the customer is no longer interested in and improving the accuracy of the recommended content.

[0026] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0027] Figure 1 This is a flowchart illustrating an optional content recommendation method according to an embodiment of this application, such as... Figure 1 As shown, the method includes at least steps S102-S108, wherein:

[0028] Step S102: Obtain the first data of the target object, wherein the first data includes at least: historical viewing content data and historical usage behavior data.

[0029] Optionally, the first data may also include: basic data of the target object and historical viewing behavior data.

[0030] Currently, with the rapid development of the Internet and network communication technologies, operators can use big data platforms to load, transform, clean, process, and store users' primary data, so that their systems can recommend content that users may be interested in based on this data.

[0031] Step S104: Determine the target object's preference characteristics based on the first data, wherein the preference characteristics include: a first preference characteristic reflecting the target object's temporary preferences and a second preference characteristic reflecting the target object's periodic preferences.

[0032] Optionally, a first preference feature of the target object is determined based on the first data, wherein the first preference feature includes at least one of the following: temporary preference content, temporary preference type; and a second preference feature of the target object is determined based on the first data, wherein the second preference feature includes at least one of the following: preference cycle, periodic preference content, periodic preference type.

[0033] For example, using a user's historical viewing content data and historical usage behavior data as access data, a temporary preference algorithm is executed to generate temporary preference features (i.e., the first preference feature).

[0034] The temporary preference feature algorithm determines a user's temporary preference features based on the type and frequency of content accessed and viewed on the Internet. Temporary preference features include: user ID, user age, user gender, network access time, number of products, temporarily preferred content, and frequency of temporarily preferred content.

[0035] Using users' historical viewing content data and historical usage behavior data as access data, the periodic preference feature algorithm is executed to generate periodic preference features (i.e., second preference features).

[0036] The periodic preference feature algorithm determines a user's periodic preference features by analyzing the number of times a user repeats similar preferences, the content of periodic preferences, and the period of preference. The periodic preference features include: user ID, user age, user gender, network access time, number of products, content of periodic preferences, number of times of periodic preference content, type of periodic preference, and number of times of periodic preference type.

[0037] Step S106: Obtain the content to be recommended, and determine the first content among the content to be recommended based on preference features, wherein the first content is the content that the target audience is no longer interested in during the target time period.

[0038] Optionally, content to be recommended can be determined based on the target audience's historical viewing data.

[0039] As an optional implementation, the recommendation system can employ content recommendation algorithms to generate content to be recommended. Common content recommendation algorithms include Content-Based (CB) and Collaborative Filtering (CF). This application does not limit the content recommendation algorithm used.

[0040] Here's a simple explanation using a content recommendation algorithm as an example. The recommendation system extracts basic attributes and content information from a list of historical content that user A likes or frequently watches, creating a tag list and assigning a weight to each tag. Then, the tag list corresponding to all content is inverted and stored in an inverted index server. When content recommendation is needed, the tag lists corresponding to each content are retrieved and concatenated into a search system query expression. The retrieved results are then sorted and output as the content to be recommended. The content to be recommended can include: user ID, content ID, content name, content quality, content value, validity period, and content features.

[0041] Optionally, the first content includes: a first sub-content and a second sub-content. The first content can be determined by: determining the content in the content to be recommended that matches the first preference feature as the first sub-content; determining the content in the content to be recommended that matches the second preference feature as the second sub-content.

[0042] As an optional implementation, the access data of the target object for each content in the recommended content during the target time period can be estimated based on the second preference feature. The access data includes at least one of the following: number of visits, access duration, and access percentage. The recommendation score for each content is determined based on the access data. Content with a recommendation score lower than a preset threshold is identified as the second sub-content.

[0043] Specifically, the first sub-content is the user's temporary preferences, and the second sub-content is the user's periodic preferences.

[0044] For example, a user cycle preference trend algorithm can be used to determine the second sub-content. The user cycle preference trend algorithm calculates the number of times a user visits a content within a cycle, the duration of visits, the historical average number of visits, and the percentage of visits within a cycle based on the user's cycle preference characteristics, in order to evaluate the user's recommendation rating for that preference.

[0045] Specifically, in this embodiment of the application, the second preference feature can be used as access data. A user periodic preference trend algorithm is used to estimate the user's access data for each content in the recommended content within the target time period. The access data includes: user ID, preference feature, preference period, number of periodic accesses, duration of periodic accesses, number of historical accesses, duration of historical accesses, percentage of periodic accesses, etc. The user's recommendation rating for each content is evaluated based on the access data. Finally, the recommendation rating is compared with the preset recommendation rating threshold of the recommendation system, and the content with a recommendation rating less than the preset threshold is determined as periodic preference content.

[0046] Step S108: Filter the first content in the content to be recommended to obtain the first target recommended content.

[0047] For example, a temporary preference content filtering algorithm can be used to filter out the first sub-content in the content to be recommended; a periodic preference content filtering algorithm can be used to filter out the second sub-content in the content to be recommended.

[0048] After determining the primary target content for recommendation, further steps are taken to reduce the redundancy of the recommended content and improve its accuracy.

[0049] Optionally, the historical viewing content data can be classified to obtain multiple content classification features; content association features can be determined based on the correlation between the various content classification features; and second content that is associated with the first content can be determined based on the content association features.

[0050] Specifically, the second content includes a third sub-content and a fourth sub-content. The second content can be determined in the following ways: based on content association features, determine the third sub-content in the first target recommended content that is related to the first sub-content; based on content association features, determine the fourth sub-content in the first target recommended content that is related to the second sub-content.

[0051] For example, firstly, a content feature algorithm is used to classify historical viewing content data, resulting in multiple content classification features such as content ID, content name, content type, and sub-category features. Then, content association features are determined based on the relationships between these content classification features. These content association features include: content ID, associated content ID, content name, content type, and sub-category features. Next, a content association algorithm is used to identify the third sub-content related to temporary preferences and the fourth sub-content related to periodic preferences within the first target recommended content. Both the third and fourth sub-contents are content that the user is not interested in during the target time period. Finally, an associated content filtering algorithm is used to filter the third and fourth sub-contents within the first target recommended content to obtain the second target recommended content.

[0052] The associated content filtering algorithm involves associating the filtered content with relevant data, then matching it with the association features of the recommended content data for a second filtering process. This filters out associated content that the user is no longer interested in. The second target recommended content includes: user ID, content ID, content name, content quality, and content validity period. Finally, the second target recommended content is pushed to the application platform of the recommendation system, which can improve the accuracy of the application platform's recommended content and thus enhance the competitiveness of the recommendation system product.

[0053] As an optional implementation method, the process of the above content recommendation method can be illustrated using a telecommunications iTV recommendation system as an example. Figure 2 As shown, a detailed explanation is provided through steps S1-S10.

[0054] S1, Obtain the user's iTV first data, which includes: iTV historical viewing content data and iTV historical usage behavior data;

[0055] S2, uses a content recommendation algorithm to generate iTV content to be recommended;

[0056] S3, the access information is the user's iTV basic data, and the user's preference characteristics are determined by the user feature algorithm. The user feature algorithm includes a temporary preference feature algorithm and a periodic preference feature algorithm. The temporary preference feature algorithm can determine the user's temporary preference characteristics (i.e., the first preference characteristic); the periodic preference feature algorithm can determine the user's periodic preference characteristics (i.e., the second preference characteristic).

[0057] S4 uses a periodic preference trend algorithm to calculate the access data of each content in the content to be recommended;

[0058] S5, the access information is iTV content to be recommended, and a temporary preference content filtering algorithm is used to filter temporary preference content (i.e. the first sub-content);

[0059] S6 uses a periodic preference content filtering algorithm to filter periodic preference content (i.e., the second sub-content);

[0060] S7, obtain the first target recommended content;

[0061] S8, the access information is iTV historical viewing content data, and the iTV historical viewing content data is classified by content feature algorithm to obtain multiple content classification features;

[0062] S9, using a content association algorithm to determine the third sub-content in the first target recommended content that is associated with temporary preference content, and the fourth sub-content that is associated with periodic preference content;

[0063] S10, use the related content filtering algorithm to filter the third and fourth sub-contents in the first target recommended content;

[0064] S11, obtain the second target recommended content;

[0065] S12, push the second target recommended content to the iTV application platform.

[0066] In this embodiment, firstly, first data of the target object is obtained, including at least historical viewing content data and historical usage behavior data. Then, based on the first data, the target object's preference characteristics are determined, including a first preference characteristic reflecting the target object's temporary preferences and a second preference characteristic reflecting the target object's periodic preferences. Next, content to be recommended is obtained, and based on the preference characteristics, a first content within the content to be recommended is determined, wherein the first content is content that the target object is no longer interested in during the target time period. Finally, the first content in the content to be recommended is filtered to obtain the first target recommended content. By analyzing the target object's temporary and periodic preferences and filtering the corresponding first and second preference characteristics, the recommendation of content that the target object is no longer interested in is effectively reduced, improving the accuracy of the recommended content and achieving the goal of enhancing the competitiveness of the recommendation system product. This solves the technical problem that related technologies cannot effectively filter recommended content that users are no longer interested in, resulting in low recommendation accuracy and a poor user experience.

[0067] Example 2

[0068] According to embodiments of this application, a content recommendation apparatus for implementing the above-described content recommendation method is also provided, such as... Figure 3 As shown, the device includes at least an acquisition module 31, a first determination module 32, a second determination module 33, and a filtering module 34, wherein:

[0069] The acquisition module 31 is used to acquire the first data of the target object, wherein the first data includes at least: historical viewing content data and historical usage behavior data.

[0070] Optionally, the first data may also include: basic data of the target object and historical viewing behavior data.

[0071] Currently, with the rapid development of the Internet and network communication technologies, operators can use big data platforms to load, transform, clean, process, and store users' primary data, so that their systems can recommend content that users may be interested in based on this data.

[0072] The first determining module 32 is used to determine the preference characteristics of the target object based on the first data, wherein the preference characteristics include: a first preference characteristic reflecting the temporary preferences of the target object and a second preference characteristic reflecting the periodic preferences of the target object.

[0073] Optionally, a first preference feature of the target object is determined based on the first data, wherein the first preference feature includes at least one of the following: temporary preference content, temporary preference type; and a second preference feature of the target object is determined based on the first data, wherein the second preference feature includes at least one of the following: preference cycle, periodic preference content, periodic preference type.

[0074] For example, using a user's historical viewing content data and historical usage behavior data as access data, a temporary preference algorithm is executed to generate temporary preference features (i.e., the first preference feature).

[0075] The temporary preference algorithm determines a user's temporary preference characteristics based on the type and frequency of content accessed and viewed on the internet. These temporary preference characteristics include: user ID, user age, user gender, network access time, number of products, temporarily preferred content, and frequency of temporarily preferred content.

[0076] Using users' historical viewing content data and historical usage behavior data as access data, the periodic preference feature algorithm is executed to generate periodic preference features (i.e., second preference features).

[0077] The periodic preference feature algorithm determines a user's periodic preference features by analyzing the number of times a user repeats similar preferences, the content of periodic preferences, and the period of preference. The periodic preference features include: user ID, user age, user gender, network access time, number of products, content of periodic preferences, number of times of periodic preference content, type of periodic preference, and number of times of periodic preference type.

[0078] The second determining module 33 is used to obtain the content to be recommended and determine the first content in the content to be recommended based on the preference features, wherein the first content is content that the target object is no longer interested in during the target time period.

[0079] Optionally, content to be recommended can be determined based on the target audience's historical viewing data.

[0080] As an optional implementation, the recommendation system can employ content recommendation algorithms to generate content to be recommended. Common content recommendation algorithms include Content-Based (CB) and Collaborative Filtering (CF). This application does not limit the content recommendation algorithm used.

[0081] Here's a simple explanation using a content recommendation algorithm as an example. The recommendation system extracts basic attributes and content information from a list of historical content that user A likes or frequently watches, creating a tag list and assigning a weight to each tag. Then, the tag list corresponding to all content is inverted and stored in an inverted index server. When content recommendation is needed, the tag lists corresponding to each content are retrieved and concatenated into a search system query expression. The retrieved results are then sorted and output as the content to be recommended. The content to be recommended can include: user ID, content ID, content name, content quality, content value, validity period, and content features.

[0082] Optionally, the first content includes: a first sub-content and a second sub-content. The first content can be determined by: determining the content in the content to be recommended that matches the first preference feature as the first sub-content; determining the content in the content to be recommended that matches the second preference feature as the second sub-content.

[0083] As an optional implementation, the access data of the target object for each content in the recommended content during the target time period can be estimated based on the second preference feature. The access data includes at least one of the following: number of visits, access duration, and access percentage. The recommendation score for each content is determined based on the access data. Content with a recommendation score lower than a preset threshold is identified as the second sub-content.

[0084] Specifically, the first sub-content is the user's temporary preferences, and the second sub-content is the user's periodic preferences.

[0085] The filtering module 34 is used to filter the first content in the content to be recommended to obtain the first target recommended content.

[0086] For example, a temporary preference content filtering algorithm can be used to filter out the first sub-content in the content to be recommended; a periodic preference content filtering algorithm can be used to filter out the second sub-content in the content to be recommended.

[0087] After determining the primary target content for recommendation, further steps are taken to reduce the redundancy of the recommended content and improve its accuracy.

[0088] Optionally, the historical viewing content data can be classified to obtain multiple content classification features; content association features can be determined based on the correlation between the various content classification features; and second content that is associated with the first content can be determined based on the content association features.

[0089] Specifically, the second content includes a third sub-content and a fourth sub-content. The second content can be determined in the following ways: based on content association features, determine the third sub-content in the first target recommended content that is related to the first sub-content; based on content association features, determine the fourth sub-content in the first target recommended content that is related to the second sub-content.

[0090] It should be noted that each module in the content recommendation device in this application embodiment corresponds one-to-one with each implementation step of the content recommendation method in embodiment 1. Since embodiment 1 has been described in detail, some details not shown in this embodiment can be referred to embodiment 1, and will not be elaborated further here.

[0091] Example 3

[0092] According to an embodiment of this application, a non-volatile storage medium is also provided, which includes a stored program, wherein, when the program is running, it controls the device where the non-volatile storage medium is located to execute the content recommendation method in Embodiment 1.

[0093] According to an embodiment of this application, a processor is also provided for running a program, wherein the program executes the content recommendation method in Embodiment 1 during runtime.

[0094] According to an embodiment of this application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the content recommendation method of Embodiment 1 through the computer program.

[0095] Optionally, during program execution, the device containing the non-volatile storage medium is controlled to perform the following steps: acquiring first data of the target object, wherein the first data includes at least: historical viewing content data and historical usage behavior data; determining the target object's preference characteristics based on the first data, wherein the preference characteristics include: a first preference characteristic reflecting the target object's temporary preferences and a second preference characteristic reflecting the target object's periodic preferences; acquiring content to be recommended, and determining a first content among the content to be recommended based on the preference characteristics, wherein the first content is content that the target object is no longer interested in during the target time period; filtering the first content among the content to be recommended to obtain the first target recommended content.

[0096] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0097] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0098] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0101] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0102] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A content recommendation method, characterized in that, include: Obtain first data of the target object, wherein the first data includes at least: historical viewed content data and historical usage behavior data; The preference characteristics of the target object are determined based on the first data, wherein the preference characteristics include: a first preference characteristic reflecting the temporary preferences of the target object and a second preference characteristic reflecting the periodic preferences of the target object, wherein the second preference characteristic includes at least one of the following: preference period, periodic preference content, and periodic preference type; The process involves acquiring content to be recommended and determining a first content within that content based on the preference features. The first content is content that the target object is no longer interested in during a target time period. The first content includes a first sub-content and a second sub-content. The first sub-content is content within the content to be recommended that matches the first preference features. The determination of the second sub-content includes: estimating the target object's access data for each content within the content to be recommended during the target time period based on the second preference features; determining the recommendation score for each content based on the access data; and identifying content with a recommendation score less than a preset threshold as the second sub-content. The first content is filtered out from the content to be recommended to obtain the first target recommended content.

2. The method according to claim 1, characterized in that, Determining the target object's preference characteristics based on the first data includes: The first preference feature of the target object is determined based on the first data, wherein the first preference feature includes at least one of the following: temporary preference content and temporary preference type; The second preference feature of the target object is determined based on the first data.

3. The method according to claim 1, characterized in that, The access data includes at least one of the following: number of accesses, access duration, and access percentage.

4. The method according to claim 1, characterized in that, The method further includes: The historical viewing content data is classified to obtain multiple content classification features; Content association features are determined based on the relationships between the various content classification features; Based on the content association features, determine the second content in the first target recommended content that is associated with the first content.

5. The method according to claim 4, characterized in that, The second content includes: a third sub-content and a fourth sub-content. Based on the content association features, the second content within the first target recommended content that is associated with the first content includes: Based on the content association features, the third sub-content in the first target recommended content that is associated with the first sub-content is determined; Based on the content association features, the fourth sub-content in the first target recommended content that is associated with the second sub-content is determined.

6. The method according to claim 1, characterized in that, Obtain content to be recommended, including: Based on the target object's historical viewing data, the content to be recommended is determined.

7. A content recommendation device, characterized in that, include: The acquisition module is used to acquire first data of the target object, wherein the first data includes at least: historical viewing content data and historical usage behavior data; The first determining module is used to determine the preference characteristics of the target object based on the first data, wherein the preference characteristics include: a first preference characteristic reflecting the temporary preferences of the target object and a second preference characteristic reflecting the periodic preferences of the target object, wherein the second preference characteristic includes at least one of the following: preference period, periodic preference content, and periodic preference type; The second determining module is used to acquire the content to be recommended and determine the first content in the content to be recommended based on the preference features. The first content is content that the target object is no longer interested in during the target time period. The first content includes: a first sub-content and a second sub-content. The first sub-content is content in the content to be recommended that matches the first preference features. The method for determining the second sub-content includes: estimating the access data of the target object to each content in the content to be recommended during the target time period based on the second preference features, determining the recommendation score of each content based on the access data, and determining the content with the recommendation score less than a preset threshold as the second sub-content. The filtering module is used to filter the first content in the content to be recommended to obtain the first target recommended content.

8. A processor for running a program, wherein, When the program runs, it executes the content recommendation method described in any one of claims 1 to 6.

9. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the content recommendation method of any one of claims 1 to 6 through the computer program.

Images