Content recommendation method and device, computer equipment and storage medium

Abstract

The invention provides a content recommendation method and device, computer equipment and a storage medium, and the method comprises the steps: determining a keyword based on input information inputted by a user on a displayed first interface; based on the keyword and the first interface, obtaining first historical data generated by the user in the first interface; determining a second interface based on the keyword; obtaining second historical data generated by the user in the second interface, wherein the second historical data comprises historical content recommended to the user through the second interface and associated information corresponding to the historical content; based on the first historical data, the historical content and the associated information corresponding to the historical content, first recommendation content corresponding to the input information is determined, and the first recommendation content is displayed on the first interface. According to the method, intention understanding of a cross-interface scene can be realized, and the accuracy of recommending content to the user is improved.

Description

Technical Field

[0001] This application relates to the field of big data analytics, and more particularly to a content recommendation method, apparatus, computer device, and storage medium. Background Technology

[0002] In e-commerce, dedicated Artificial Intelligence (AI) models are typically deployed for independent scenarios such as search, recommendation, and customer service to achieve user question-and-answer or content recommendation within a single point. This approach is limited by the closed loop of a single scenario and cannot achieve cross-scenario collaboration and information flow. To improve the user experience, related technologies have further designed unified interaction modules (such as Software Development Kits, SDKs) that can be embedded in multiple pages. However, the underlying AI still strictly relies on the real-time functions and user behavior data of the current page when perceiving user needs and generating content. This results in a situation where, although the interaction entry point is unified in form, the decision-making during content generation still depends only on the local behavior and information of the current page. It cannot integrate the user's coherent intent across different scenarios, nor can it build a unified understanding across pages. Summary of the Invention

[0003] This application discloses a content recommendation method, apparatus, computer device, and storage medium, which solves the technical problem that existing AI models cannot achieve cross-scene intent understanding and service collaboration.

[0004] This application provides a content recommendation method, the method comprising: determining keywords based on input information entered by a user on a displayed first interface; obtaining first historical data generated by the user on the first interface based on the keywords and the first interface; determining a second interface based on the keywords, the second interface being at least one interface on which the user performed historical operations based on the keywords on a non-first interface; obtaining second historical data generated by the user on the second interface, the second historical data including historical content recommended to the user through the second interface and related information corresponding to the historical content, the historical content including historical recommended content that has not been viewed by the user; determining first recommended content corresponding to the input information based on the first historical data, the historical content and the related information corresponding to the historical content, the first recommended content being displayed on the first interface.

[0005] In some embodiments of this application, the method further includes: when the update conditions of the first recommended content are met, obtaining the association information corresponding to the first recommended content; generating an analysis result of the first recommended content based on the association information corresponding to the first recommended content; generating second recommended content based on the analysis result and third historical data, and displaying the second recommended content on the first interface.

[0006] In some embodiments of this application, the method further includes: if no viewing operation of the user on the first recommended content is detected within a preset time period of the first recommended content display, determining that the update condition is met; and / or, if the user is detected performing a preset operation on the first recommended content, determining that the update condition is met; wherein the preset operation includes one or more of page turning operations and swiping operations.

[0007] In some embodiments of this application, obtaining the second historical data generated by the user in the second interface includes: obtaining the historical content based on the keywords, wherein the historical content includes an object list constructed from multiple recommended objects; obtaining information related to each recommended object in the object list to obtain the association information corresponding to the historical content; and using the association information between the object list and the historical content as the second historical data.

[0008] In some embodiments of this application, the method further includes: using an intelligent agent to obtain the historical content and the associated information corresponding to the historical content, including: the intelligent agent calling the search function corresponding to the search control displayed on the second interface through a first data interface to search for the keywords and obtain the historical content; based on the multiple recommended objects in the historical content, the intelligent agent obtaining information related to each recommended object through a second data interface to obtain the associated information corresponding to the historical content.

[0009] In some embodiments of this application, determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the associated information corresponding to the historical content includes: if the input information is generated based on the user performing a query operation on the first interface, generating the query result based on the input information; predicting the user's intent based on the first historical data, the historical content, and the associated information corresponding to the historical content; generating a prediction control based on the intent; and displaying the query result and the prediction control through the first interface.

[0010] In some embodiments of this application, determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the associated information corresponding to the historical content includes: if the input information is generated based on the user performing a search operation on the first interface, generating a first result based on the first historical data, the historical content, the associated information corresponding to the historical content, and a preset sorting strategy; generating a second result based on the first historical data, the historical content, the associated information corresponding to the historical content, and the type of the keyword; and determining the first recommended content based on the first result and the second result.

[0011] This application also provides a content recommendation device, comprising: a determining module, configured to determine keywords based on input information entered by a user on a displayed first interface; an acquiring module, configured to acquire first historical data generated by the user on the first interface based on the keywords and the first interface; the determining module is further configured to determine a second interface based on the keywords, the second interface being at least one interface on which the user performed historical operations based on the keywords on a non-first interface; the acquiring module is further configured to acquire second historical data generated by the user on the second interface, the second historical data including historical content recommended to the user through the second interface and related information corresponding to the historical content, the historical content including historical recommended content that has not been viewed by the user; and a generating module, configured to determine first recommended content corresponding to the input information based on the first historical data, the historical content, and the related information corresponding to the historical content, the first recommended content being displayed on the first interface.

[0012] This application also provides a computer device including a processor and a memory, wherein the processor is used to implement the content recommendation method when executing a computer program stored in the memory.

[0013] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the content recommendation method.

[0014] In the content recommendation method provided in this application, keywords are determined based on the input information entered by the user on the first displayed interface, so that recommended content can be generated based on the keywords. Based on the keywords and the first interface, first historical data generated by the user on the first interface is obtained, thus obtaining the historical data of the interface where the keyword is currently located. A second interface different from the first interface is determined through the keywords, and second historical data generated by the user on the second interface is obtained, thus obtaining the historical data of the interface where the keyword is currently located. This achieves the purpose of obtaining historical data under different interface scenarios, improving the accuracy of subsequent generated recommended content. The second historical data includes historical content and corresponding association information. This historical content includes historical recommended content that has not been viewed by the user, so that the reasons why the user has not viewed it can be analyzed later. Based on the first historical data, historical content, and corresponding association information, the first recommended content corresponding to the input information is determined and displayed on the first interface, so as to deeply explore the user's intent under different interface scenarios, thereby achieving cross-scenario intent understanding and improving the efficiency and accuracy of content recommendation. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the application environment of the content recommendation method provided in the embodiments of this application.

[0016] Figure 2 This is a flowchart of the content recommendation method provided in the embodiments of this application.

[0017] Figure 3 This is a schematic diagram of entering the first interface provided in the embodiments of this application.

[0018] Figure 4 This is a schematic diagram illustrating the construction of a user profile provided in an embodiment of this application.

[0019] Figure 5 This is a schematic diagram of the module application provided in the embodiments of this application.

[0020] Figure 6 This is a schematic diagram of the interaction between the intelligent agent and the operation module provided in the embodiments of this application.

[0021] Figure 7 This is a flowchart illustrating the generation process of the second recommended content proposed in this application embodiment.

[0022] Figure 8 This is a structural diagram of the content recommendation device provided in the embodiments of this application.

[0023] Figure 9 This is a schematic diagram of the structure of the computer device provided in the embodiments of this application. Detailed Implementation

[0024] For ease of understanding, some concepts related to the embodiments of this application are illustrated and explained by way of example for reference.

[0025] It should be noted that in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and drawings of this application are used to distinguish similar objects, not to describe a specific order or sequence.

[0026] In e-commerce, dedicated artificial intelligence (AI) models are typically deployed for separate scenarios such as search, recommendation, and customer service. Specifically, dedicated AI models (e.g., search ranking models, recommendation recall models, customer service dialogue models) are trained and deployed separately for each scenario. This deployment model can achieve accurate user Q&A or content matching for specific task optimization goals (such as click-through rate and conversion rate). However, its limitation lies in the fact that each AI model can only perceive and process local information within its own scenario. For example, if a user repeatedly compares the performance parameters of a certain mobile phone in the search scenario, when they switch to the recommendation scenario, the recommendation model cannot recognize the purchase intention generated in the search scenario and still pushes recommendations based on the context information of the recommendation scenario, ultimately leading to a disjointed experience.

[0027] To improve this fragmented user experience through interactive methods, related technologies introduce unified interaction modules, such as embedded SDKs, to provide a unified intelligent assistant floating window across multiple pages, including product detail pages and shopping cart pages. However, this technology remains strictly bound to the context of the user's current page. This means that when a user is on a product page, the AI's decisions are based solely on the product information on that page and the user's real-time behavioral data, such as clicks and dwell times. It cannot proactively "recall" or "understand" the user's coherent sequence of intentions across all scenarios. Therefore, this essentially only physically aggregates multiple "data silos" rather than building a unified cognitive center capable of integrating information from across the entire domain and understanding the user's global intentions; its output remains partial and one-sided.

[0028] To address the technical problem that existing AI models cannot achieve cross-scenario intent understanding and service collaboration, embodiments of this application provide a content recommendation method, apparatus, computer device, and storage medium, which can achieve intent understanding across interface scenarios and improve the accuracy of content recommendations to users.

[0029] To better understand the content recommendation method, apparatus, computer equipment, and storage medium provided in the embodiments of this application, the application scenarios of the content recommendation method of this application are described below.

[0030] Figure 1 This is a schematic diagram illustrating the application environment of the content recommendation method provided in the embodiments of this application. For example... Figure 1 As shown, the content recommendation method described in this application uses the operating environment of an e-commerce platform 10 as its application environment. The e-commerce platform 10 can be an e-commerce service platform with open independent site services. The e-commerce platform 10 can be used to provide merchants' goods and services to users. Among them, users can be merchant users (e.g., sellers, retailers, wholesalers, or product providers), consumer users (e.g., buyers, purchasing agents, or product users), potential users (e.g., users who are browsing but have not yet committed to purchasing, etc.).

[0031] E-commerce platform 10 can be implemented through a processing facility including a processor and memory. When instructions stored in the memory are executed, the e-commerce platform 10 performs the e-commerce functions and supporting functions described herein. The processing facility can be a server, client, network infrastructure, mobile computing platform, cloud computing platform, or other computing platform, and provides electronic connectivity and communication between electronic components of e-commerce platform 10, merchant devices 20, client devices 30, etc. E-commerce platform 10 can be implemented as cloud computing services, Software as a Service (SaaS), Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Desktop as a Service (DaaS), Managed Software as a Service (MSaaS), Mobile Backend as a Service (MBaaS), Information Technology Management as a Service (ITMaaS), etc.

[0032] E-commerce platform 10 can provide a centralized system for offering merchants online resources and facilities to manage their businesses. E-commerce platform 10 includes communication facilities 101 and a merchant data server 102, which maintains data related to each merchant, such as merchant profiles, merchant settings, transaction data, product data, collected data, etc. In an optional embodiment, communication facilities 101 may be deployed on merchant data server 102, which is not limited herein. Merchant data server 102 includes a memory 1021 for storing all merchant data. Merchant data server 102 further includes a processor 1022 for directly executing or directing all operations performed by merchant data server 102. Processor 1022 may be implemented by one or more processors that execute instructions stored in memory 1021.

[0033] In some optional embodiments, the e-commerce platform 10 receives data requests from merchant device 20 or client device 30 via communication facility 101. Merchant data server 102 processes the data requests via processor 1022, retrieves the data requested by merchant device 20 or client device 30 from memory 1021, and provides the requested data to communication facility 101 via network.

[0034] E-commerce platform 10 may be configured with a business management engine 104 for content management, task automation, and data management to support and serve multiple online stores 103 (e.g., independent website stores). The business management engine 104 includes basic or "core" functionalities of e-commerce platform 10 (e.g., common to most online store activities, such as cross-channel, administrator interface, merchant location, industry, product type, etc.) that can be reused across online stores 103 (e.g., functions that can be reused / modified across core functionalities). In one example, the business management engine 104 may be used to extract relevant features for each product in online store 103 and store these features in the memory 1021 of merchant data server 102.

[0035] Merchants can manage one or more storefronts in online store 103 through merchant device 20 and offer goods to customers through various channels (e.g., online store 103). The terms "online store 103" and "storefront" can be used synonymously to refer to online e-commerce provided by the merchant through e-commerce platform 10.

[0036] Merchants can log in to the administrator 105 via merchant device 20 (such as from a desktop computer or mobile device) and manage various aspects of their online store 103. For example, merchants can customize the appearance of the online store 103 through a theme system, or customize the presentation content of the product detail page (such as product recommendation list, product title, product details, product images, product attributes, frequently asked questions about products, and other types of content).

[0037] Online store 103 can support a large number of independently managed stores and process a large amount of transaction data for various products every day. Transaction data may include customer contact information, information about the purchased goods, information about the services presented, and any other information related to the business conducted through e-commerce platform 10. In some embodiments, e-commerce platform 10 may store the transaction data in merchant data server 102.

[0038] In some embodiments, customers can interact via customer device 30 (e.g., computer, mobile terminal, etc.) or any other business interface device known in the art. E-commerce platform 10 enables merchants to contact customers through online store 103, facilitate business with customers through dialogue via electronic communication facility 101, etc., thereby providing a system for contacting customers and facilitating merchant services for real or virtual paths that can be used to contact and interact with customers.

[0039] Online store 103 can be provided to customers by e-commerce platform 10, where customers can browse and purchase various available goods. For example, online store 103 can be served by a webpage provided by the server of e-commerce platform 10 to customer device 30. The server can receive requests for webpages from a browser or other application installed on customer device 30, and in response, the server sends back the requested webpage, which is then displayed by the web browser (or other application) of customer device 30. For example, a product search box can be displayed to customers on the store's homepage of online store 103, which can accept text input and graphic input, etc.

[0040] It is understood that the application scenarios illustrated in the embodiments of this application do not constitute specific limitations. In other embodiments of this application, the e-commerce platform 10 may also include more or fewer components, such as network interfaces, data analysis components, etc.

[0041] Figure 2 This is a flowchart of a content recommendation method provided in an embodiment of this application, applied to a computer device (e.g., Figure 9 The computer device 90 shown is deployed as follows: Figure 1 In the e-commerce platform 10 shown, the order of steps in this flowchart can be changed, and some steps can be omitted, depending on different needs.

[0042] Step S201: Determine keywords based on the input information entered by the user on the first displayed interface.

[0043] In some embodiments of this application, the computer device can detect user interactions on an e-commerce platform, such as login, clicking, and swiping. The computer device can then present different display interfaces to the user based on these interactions.

[0044] In one example, taking a computer as the client device, the computer's display device is typically a monitor. Users can manipulate the cursor or controls on the display interface using input devices connected to the computer (e.g., a mouse, keyboard, or touchpad). These input devices can detect the user's intentions and translate the actions into instructions for specific locations or elements on the screen; that is, they can recognize the user's input operations on the graphical user interface. Input operations can be cursor movement, clicking, double-clicking, and dragging performed with a mouse; character input and shortcut key triggering performed with a keyboard; or gesture operations performed with a touchpad, etc., and this application does not limit these possibilities.

[0045] The computer displays a primary interface to the user. This primary interface can be the one that appears after the user clicks a desktop shortcut or launches a local software program through File Explorer. This software program can be a desktop application that needs to be downloaded and installed, such as a shopping website application or an e-commerce platform application.

[0046] The first interface can also be the display interface after entering the application. Taking the first interface of a shopping website application displayed to a user on a computer as an example, this first interface can be the main page of the e-commerce platform when the user enters, the main page of any online store within the e-commerce platform when the user enters, or the page entered after the user triggers any platform element of the e-commerce platform. This platform element can be a control, an input box, any product, etc. This application does not limit the specific type of platform element.

[0047] The first interface accessed through different methods represents different scenarios. In one example, the first interface accessed through a customer service control can be recorded as a customer service scenario; the first interface accessed through an input box on the store's main page can be recorded as a search / query scenario within the store; and the first interface accessed through an input box on the platform's main page can be recorded as a search / query scenario within the platform's main page.

[0048] After entering the first interface, users can control the computer to display the first interface in windowed mode, full-screen mode, or other custom window sizes according to their operating habits.

[0049] In some embodiments of this application, various types of interface elements can be displayed on the first interface, and the interface elements can be divided into three dimensions: content presentation type, interactive operation type, and status and information type.

[0050] Content presentation can include product media content, such as main product images, detail images, display videos, panoramic images, 3D models, AR try-on / trial views, and live streams.

[0051] Interactive operation classes include controls, navigation and scrolling components. Controls may include command controls (such as "Add to Cart", "Buy Now", "Order Inquiry", "Cancel Order", "Apply for After-sales Service"), navigation controls (such as favorites, share), overlays / pop-ups, steppers (quantity increase / decrease), and input controls (such as search boxes). Navigation and scrolling components may include pull-to-refresh components, sliders, carousel indicators, and tabs.

[0052] Status and information categories can include dynamic status information, social information, and notification information. Dynamic status information can include order status, inventory status, price and promotion information (coupon information), and real-time updates (such as "XXX just purchased"). Social information can include user reviews (with pictures / videos), Q&A sections, recommendation tags such as "You May Also Like", sales volume display, store / product ratings, and quality assurance labels. Notification information can include operational banners / announcements, contextualized notification text (such as "Buy XX more and get free shipping"), loading status, etc.

[0053] In some embodiments of this application, the computer device can detect input operations that a user can perform on a displayed first interface and obtain the user's input information. To better understand the process of generating input information, the following describes... Figure 3 The following are descriptions of (a) to (c) shown.

[0054] like Figure 3 As shown in (a), the computer device displays multiple controls on its interface, such as control 1, control 2, and control 3. This interface can be a user's personal account page, a store page, or a product page. The multiple controls can be presented in a floating format, which is not limited in this application. The first interface is the display interface entered by the user clicking on the customer service control (assuming it is control 1). If the computer device detects that the user enters "query order status" in the input box of the first interface, then "query order status" is the input information. If the computer device detects that the user enters "query air conditioner functions" in the input box of the first interface, then the input information obtained is "query air conditioner functions".

[0055] like Figure 3 As shown in (b), the main interface of the computer device displays multiple applications, such as Application 1, Application 2, and Application 3. The first interface is the display interface that the user enters after logging into the application corresponding to the shopping software (assuming it is Application 1). If the computer device detects that the user enters "Midea Air Conditioner" in the input box of the first interface, then the input information obtained is "Midea Air Conditioner".

[0056] like Figure 3As shown in (c), the computer device displays the main interface of the online store, which includes input boxes and store products (such as product X1 and product X2). If the computer device detects that the user has entered "air conditioner" in the input box, then the input information obtained is "air conditioner".

[0057] like Figure 3 The examples shown (a) to (c) are merely examples, and this application does not limit the input format of the input information.

[0058] In some embodiments of this application, after a computer device receives input information, it can parse the input information and extract keywords from it. This application does not limit the method of keyword extraction.

[0059] Step S202: Based on keywords and the first interface, obtain the first historical data generated by the user in the first interface.

[0060] In some embodiments of this application, before acquiring the first historical data, a memory module can be constructed to record user profiles and behavioral trajectory data, stored in a Hadoop database. The behavioral trajectory data can be time-series data divided according to behavior time and scenario. For example, if a user sequentially enters a search scenario and a customer service scenario, the recorded behavior time in the search scenario is earlier than that in the customer service scenario. Furthermore, data generated within the search scenario and data generated within the customer service scenario also need to be recorded according to behavior time. This allows for subsequent querying of different scenarios and behavioral data generated within different scenarios by time.

[0061] Taking buyers as an example, a user profile can include user attribute information, such as gender and age. Behavioral trajectory data can include price sensitivity, product category preference, brand preference, and preference for new products. The recording format for behavioral trajectory data is as follows: Time | Scenario | Behavior Type | Specific Behavior | Result. Example record: (2025-01-01 10:10 | Input box on store homepage | Search | Search keywords | No clicks). The following section combines... Figure 4 Describe the process of building user profiles.

[0062] like Figure 4 As shown, the process of building a user profile consists of several steps: data source, data collection and storage, and profile tag data modeling. Figure 4 The data applications shown represent examples of scenarios where user profiles can be applied.

[0063] The data sources include internal and external sources. Internal data sources include business data, server-side (which can be computer equipment) log data, and client-side (which can be client devices) reported data. Business data includes user registration and product data; server-side log data includes scenario data and transaction data; and client-side reported data includes behavioral data and interaction data. External data sources include third-party data and public data. Third-party data includes external web crawlers and user research data; and public data includes public materials and shared data.

[0064] Data acquisition and storage includes offline / real-time data accessed from internal and / or external data sources. The accessed data is extracted to obtain structured and unstructured data. Streaming and offline computations are then performed on the structured and unstructured data. Streaming computations include Flink, Spark Streaming, and Storm; offline computations include Hadoop, Hive, and Spark. The processed data is stored in a persistent layer, which can include one or more of HDFS, HBase, Redis, MongoDB, Elasticsearch, and MySQL.

[0065] In profile tag data modeling, modeling methods can be pre-deployed, categorized into rule-based, model-based, and algorithm-based methods. Rule-based modeling methods can include basic statistics, mean analysis, numerical analysis, probability distribution, and analysis of variance; model-based modeling methods can include multiple regression, cluster analysis, time series analysis, entropy method, and analytic hierarchy process; and algorithm-based modeling methods can include neural networks, NLP, image recognition, audio recognition, and LTR.

[0066] A tagging system is constructed by creating labels for the stored data based on modeling methods. This system includes core user attributes and user preference attributes. Core attributes can include both natural and social attributes. Natural attributes can include gender, height, age, and weight; social attributes can include consumption level, family situation, social status, and occupation.

[0067] Preference attributes can include interest preferences and behavioral consumption preferences. Interest preferences can include whether one likes sports, food, games, or music; behavioral consumption preferences can include internet literacy, industry category preferences, payment method preferences, and shopping channel preferences.

[0068] Once the tagging system is built, it can provide support for subsequent data applications. For example, it can be applied internally, such as for user insights, operational analysis, risk control, and promotion systems; and it can be applied externally, such as for platform openness and collaborative operations.

[0069] At least some of the aforementioned data requires user permission to access, such as data obtained from external data sources or data required by external applications.

[0070] In other embodiments of this application, the computer device can also construct a store profile for the online store that has completed registration. The store profile may include store attributes, store sales, store operations, etc., wherein store attributes include main category, store type (site cluster / boutique), etc.; store sales include monthly sales, cumulative sales, average price of products for sale, etc.; store operations may include monthly active users, monthly retained users, first order completion time, plugin type preference, product performance (such as product click-through rate, purchase rate), etc.

[0071] In some embodiments of this application, the first historical data refers to the digital collection of all recordable behavioral trajectories, attribute information, and interaction results retained by the user on the first interface before the current time point. The first historical data is a structured or unstructured dataset that is ordered by time and continuously accumulated, forming the temporal basis of the user's digital profile and is also the core production data for user behavior analysis and business intelligence decision-making.

[0072] Specifically, the scope of first-historical data is extremely broad. Its core components include user attribute data (such as static registration information and account levels), behavioral interaction data (such as implicit intent sequences like page views, clickstreams, and search queries), transaction fulfillment data (complete records of orders, payments, and after-sales service), and multimodal interaction data (such as customer service dialogues and reviews of media content). Furthermore, supporting device and environmental data (device identification, network environment, etc.) also provide important dimensions for understanding the context of user behavior. This data, linked vertically by timestamps, forms a complete view of the user's lifecycle.

[0073] User behavior data generated on the first interface can be retrieved based on user profiles. This data can then be filtered using keywords to select behavior patterns related to those keywords. This keyword-related behavior data generated on the first interface is then used as the primary historical data.

[0074] In one example, assuming the first interface is the one presented in a customer service scenario, the system determines whether the user entered information such as "inquiring about air conditioning functions" or "inquiring about order status" on the first interface during the historical time period prior to the current moment. If the keyword includes "air conditioning," then all data related to the input information "inquiring about air conditioning functions" is retrieved from the trajectory behavior data. For example, the historical query time for "inquiring about air conditioning functions" and the corresponding response content (also known as the recommended content presented to the user) are recorded as the first historical data. The above is just an example; the first historical data can also include multiple response contents generated during the historical time period, the interval between two adjacent response contents, etc.

[0075] In other embodiments of this application, the first historical data may further include unviewed display content and associated information of the display content. In one example, assuming the first interface is a customer service scenario, it is determined that the user entered "the best-selling air conditioner brand" on the first interface during a historical period prior to the current moment. The corresponding response content includes a list of brands, and no user click was detected on any brand displayed in the brand list. Therefore, the unviewed display content in the first historical data includes the brand list, and the associated information of the display content includes data on the air conditioners corresponding to each brand (which may be the top 10 air conditioners in terms of sales).

[0076] Step S203: Determine the second interface based on keywords.

[0077] In some embodiments of this application, the second interface may include at least one scenario-corresponding interface, which is at least one interface for users to perform historical operations on a non-first interface based on keywords. That is, the first interface and the second interface are generated based on different functional modules and belong to different pages / interfaces, performing at least some different functions.

[0078] In some embodiments of this application, keywords are used to search for data not related to the first interface in the user's behavior trajectory data, thereby determining the second interface that is associated with the user.

[0079] In one example, assuming the first interface is the one presented in a customer service scenario, starting from the current moment, the system retrieves user behavior data within a preset period. For example, if the preset period is 7 days and the current moment is 00:00 on the 8th, then the system can retrieve behavior data generated from the 1st to the 7th. If, within the preset period (i.e., from the 1st to the 7th), the user enters information containing keywords in the search box on the main page of the shopping app, then the interface containing the search box on the main page of the shopping app is recorded as the second interface. If, within the preset period (i.e., from the 1st to the 7th), the user also enters information containing keywords in the search box within an online store, then both the interface containing the search box on the main page of the shopping app and the interface containing the search box within the online store are recorded as the second interface.

[0080] In another example, if no second interface associated with the keyword is detected within a preset period, meaning the user is only active in one interface scenario within the preset period. For example, suppose the first interface is the search box on the main page of a shopping app, and the user only used the search box on the main page of the shopping app within the preset period. In such scenarios, the search scope of historical data can be expanded. The preset period can be changed from 7 days to 14 days to obtain behavioral trajectory data from the 14th day before the current moment to the current moment, so as to query the second interface associated with the keyword from this behavioral trajectory data.

[0081] The second interface is at least one interface on which the user performs historical operations based on keywords within a preset period, on a non-first interface. The range of the preset period changes dynamically according to the interface scenario to ensure that there is at least one second interface on a non-first interface within the preset period, providing a basis for obtaining data from different interface scenarios and achieving the goal of cross-scenario.

[0082] Combination Figure 3 Assuming the user's first screen is as follows: Figure 3 As shown in (b), and the keyword is "air conditioning"; such as Figure 3 The interface shown in (c) is the interface for users to perform historical operations (such as search operations) based on keywords within a preset period. It will then be as follows: Figure 3 The interface shown in (c) serves as the second interface.

[0083] Step S204: Obtain the second historical data generated by the user in the second interface.

[0084] In some embodiments of this application, based on the second interface, if it is determined that there is an interface other than the first interface where historical operations were performed, then data associated with the second interface can be obtained to obtain second historical data. The second historical data includes historical content recommended to the user through the second interface and the associated information corresponding to the historical content. The historical content includes historical recommended content that has not been viewed by the user. Additionally, the second historical data also includes historical recommended content that has been viewed by the user. The historical recommended content that has not been viewed by the user includes an object list constructed from multiple recommended objects, and the associated information corresponding to the historical content can include information about the association between each recommended object in the object list.

[0085] Specifically, the second interface presents users with historical recommended content related to the keywords. For example, if the keywords include air conditioners, then the historical recommended content displayed on the second interface will show multiple recommended objects, which will be multiple air conditioners, forming an air conditioner list.

[0086] If no user clicks or views any of the displayed air conditioners are detected, then the multiple air conditioner displays are determined to be historical recommendations that have not been viewed by the user. Viewing can be a user's voluntary action via a gesture control interface, which can include one or more of the following: head shaking, nodding, blinking, or sticking out the tongue. For example, a user nodding can trigger a click action on the computer device; therefore, if no head nodding is detected when multiple air conditioners are displayed, then it is determined that those air conditioners have not been viewed by the user.

[0087] Once the historical content is identified, related information can be obtained. Continuing with the example above, if the historical content includes multiple air conditioners, the related information can include detailed product information for each air conditioner (e.g., price, title, functions, etc.), click-through rate, sales volume, etc.

[0088] Since the second interface includes at least one interface where the user performs historical operations based on keywords on a platform other than the first interface, the acquired second historical data includes the historical content corresponding to at least one interface and the associated information corresponding to the historical content.

[0089] In some embodiments of this application, since the user does not view the displayed historical content, the associated information corresponding to the historical content will not be recorded in the user's corresponding behavior trajectory data. Therefore, the historical content can be obtained from the user's corresponding historical trajectory data, and the associated information related to the historical content can be found from the store profile or store database based on the historical content.

[0090] To identify unvisited historical recommendations and obtain unrecorded data as described above, agents can be integrated into the application. For example, an agent can be integrated into a shopping website application and associated with each interface scene in the application.

[0091] If the user's memory module only records data related to search terms and not historical content recommended to the user based on those terms, then the agent needs to use the first data interface to call the search function corresponding to the search control displayed on the second interface to search for keywords and obtain historical content. It's understandable that to ensure the displayed historical content matches the user's historical actions, the agent can generate historical content based on the historical time when the keywords appeared.

[0092] In one example, if the current time is 0:00 on the 8th, and a keyword associated with the second interface is confirmed to have appeared at 0:00 on the 7th, the agent retrieves the behavioral trajectory data prior to 0:00 on the 7th through the first data interface and triggers the search function corresponding to the search control to search for the keyword and generate search candidate data. Then, based on the behavioral trajectory data, the search candidate data is filtered and sorted to generate historical content.

[0093] After determining the historical content, the agent can also obtain information related to each recommended object through the second data interface to obtain the association information corresponding to the historical content.

[0094] In one example, the recommendation object can be a product. The agent can obtain information related to each product through the second data interface, and then use the set of related information corresponding to each product as the related information corresponding to the historical content.

[0095] Through the above embodiments, it is possible to use intelligent agents to trigger corresponding functions to obtain historical unrecorded data, as well as data under different interface scenarios, providing data support for the subsequent purpose of in-depth analysis of user intent.

[0096] Step S205: Based on the first historical data, historical content and the corresponding association information, determine the first recommended content corresponding to the input information, and display the first recommended content on the first interface.

[0097] In some embodiments of this application, the first historical data and the input information generated by the user on the first interface belong to the same interface scenario and can be recorded as historical data under the same scenario. The second historical data includes historical content and the associated information corresponding to the historical content. The second historical data and the input information generated by the user on the first interface belong to different interface scenarios and can be recorded as historical data across scenarios.

[0098] By analyzing historical data from both the same and cross-scenario contexts, it's possible to gain a deeper understanding of user intent. In particular, since cross-scenario historical data includes previously unviewed recommendations, analyzing this data can determine why users didn't view those recommendations—for example, a significant price discrepancy with their expectations. Furthermore, analyzing both primary and secondary historical data can help identify recent user preferences.

[0099] Based on the determined reasons and recent preferences, a primary recommendation is generated and displayed on a primary interface. This primary recommendation may include one or more of the following: a list of recommended products, query information, and predictive controls for the next action.

[0100] The process of generating the first recommended content is described below using two scenarios.

[0101] The first scenario: The input information is generated based on the user's query operation on the first interface. For example, if the system detects that the user entered "query order progress" in the input box in a customer service scenario, it confirms that the user has performed a query operation. The query result is generated based on the input information, for example, the query result could be "Your order has not been shipped and is expected to ship tomorrow".

[0102] Furthermore, based on the first historical data, historical content, and the corresponding correlation information, the user's intent is predicted; that is, the reasons why the user entered "check order progress" are analyzed. For example, if both the first and second historical data indicate that the user has recently engaged in frequent purchases followed by refunds, then the intent may be to request a refund.

[0103] Predictive controls are generated based on the user's intent. For example, the predictive control could be a button with a refund function. Therefore, in the first scenario, the first recommended content displayed on the first interface includes the query information and the predictive control corresponding to the next action. That is, the first interface displays "Your order has not been shipped and is expected to ship tomorrow" and a button with a refund function.

[0104] The second scenario: The input information is generated based on the user's search operation on the first interface. For example, if the user enters "Midea air conditioner" in the input box in the search scenario of the page, it is determined that the user has performed a search operation.

[0105] Based on the first historical data, historical content, the corresponding related information of the historical content, and the preset sorting strategy, a first result is generated. This involves pre-setting multiple initial sorting strategies, obtaining the click-through rate and conversion rate for each initial sorting strategy, and determining the final sorting strategy from these initial strategies based on the click-through rate and conversion rate.

[0106] The second result is generated based on the first historical data, historical content, corresponding related information, and keyword types. Keyword types include product terms (e.g., dress), brand terms (e.g., apple), model terms (e.g., iPhone 16), and specification terms (e.g., 256GB).

[0107] In the second scenario, the first recommended content is generated by using data from two dimensions: ranking strategy and keyword type. This makes the second recommended content more in line with the user's intent, avoids misjudgment of intent, and improves the user experience.

[0108] Through the above embodiments, keywords are determined based on the input information entered by the user on the first displayed interface, so that recommended content can be generated subsequently based on the keywords. Based on the keywords and the first interface, first historical data generated by the user on the first interface is obtained, thus obtaining the historical data of the interface where the keyword is currently located. A second interface different from the first interface is determined through the keywords, and second historical data generated by the user on the second interface is obtained, thus obtaining the historical data of the interface where the keyword is currently located. This achieves the purpose of obtaining historical data under different interface scenarios, improving the accuracy of subsequent recommended content generation. The second historical data includes historical content and corresponding association information. This historical content includes historical recommended content that has not been viewed by the user, so that the reasons why the user has not viewed it can be analyzed later. Based on the first historical data, historical content, and corresponding association information, the first recommended content corresponding to the input information is determined and displayed on the first interface, so as to deeply explore the user's intent under different interface scenarios, thereby achieving cross-scenario intent understanding and improving the efficiency and accuracy of content recommendation.

[0109] In other embodiments of this application, such as Figure 5 As shown, multiple modules can be integrated into the application, such as intelligent agent, memory module, perception and planning module, tool module, and action module.

[0110] The intelligent agent integrates at least one algorithmic model, such as one or more of the following: Large Language Model (LLM), knowledge graph, reinforcement learning model, causal inference model, machine learning model, and multimodal model. The intelligent agent connects to the memory module, perception and planning module, tool module, and action module respectively, and controls these modules to perform corresponding functions.

[0111] The memory module records user behavior data and user profiles. The perception and planning module is used to perceive the interface scene, break down tasks, and perform self-reflection. It can also retrieve the data stored in the memory module.

[0112] The tool module integrates multiple data interfaces. Based on the MCP rules, it encapsulates and opens the ports of the e-commerce system and uses tokens to implement interface authentication capabilities. This allows intelligent agents to freely call the tool module based on the output of the perception and planning modules, thereby enabling them to acquire information and perform operations on the e-commerce system.

[0113] like Figure 5 As shown, multiple data interfaces may include search APIs, recommendation APIs, order cancellation APIs, logistics APIs, query APIs, etc. The first data interface mentioned above can be a search API, and the second data interface can be a query API. The tool module can be used to call the functions of interface elements, for example, to get a list of products under keywords through the search API, to cancel an order through the order cancellation API, and to query the real-time logistics information of an order through the query API.

[0114] Combination Figure 6 As shown, when the agent detects a user's query operation, it sends a "Query Receipt Status" command + token to the tool module. The tool module responds by calling the "Query Receipt Status" command + token to retrieve the receipt status API. When the agent detects a user's return operation, it sends a "Return Only" command + token to the tool module. The tool module responds by calling the "Return Only" command + token to execute the operation and complete the return. Next, the agent sends an "Order Cancellation" command + token to the tool module. The tool module responds by calling the "Order Cancellation" command + token to execute the operation and cancel the order. Additionally, when the agent predicts that the probability of user churn is greater than a preset value, it can send a "Distribute Discount Coupon" command + token to the tool module. The tool module responds by calling the "Distribute Discount Coupon" command + token to execute the operation, such as distributing a discount coupon or free shipping coupon.

[0115] The action module is used to present the primary recommended content, such as the results of informational Q&A, order cancellation prompts, and a list of recommended products. In addition to performing corresponding functions based on the agent, at least some of the action module's functions can be implemented by invoking the utility module.

[0116] In one example, the agent detects that a user is searching for "Midea air conditioner" on the first interface presented in a customer service scenario. The agent uses its memory module to obtain the store profile (assuming the user has entered a store) and user profile from the first interface. Based on the store profile and user profile, the agent calls the perception and planning modules.

[0117] Next, the perception and planning module perceives the scene, determines that the user is on the first interface, and obtains contextual information (including first historical data). The perception and planning module obtains the user's behavioral trajectory data (including second historical data) from the memory module. For example, the user may have inquired about the air conditioner's functions in a customer service scenario, or searched for Midea air conditioners on the search page but did not click on the air conditioner list displayed based on the Midea air conditioner search results.

[0118] Based on the information obtained above, the perception and planning modules reflect on themselves and determine the task plan. The agent, based on the task plan, acquires first and second historical data through the tool module, and then outputs the final decision, generating the first recommended content.

[0119] Specifically, the agent detected that the user searched for "air conditioner" on the second interface presented in the search scenario, but did not detect the air conditioner list generated based on "air conditioner". To analyze why the user did not check, the agent obtained historical content from the second historical data through the first data interface of the tool module. This historical content includes the product list. The agent then obtained the association information of the historical content through the second data interface of the tool module. This association information includes product information corresponding to each product in the product list, such as click-through rate, sales volume, price, title, and features.

[0120] The perception and planning module interacts with the agent multiple times, analyzing second-level historical data to determine why the user did not browse. Based on the reason for the user's failure to browse and the first-level historical data, the agent provides guidance in the customer service scenario, generating primary recommended content. The agent then displays this primary recommended content to the user through the action module; for example, it might show the user three recommended Midea air conditioners.

[0121] After displaying the first recommended content to the user, the memory module tracks the user's activity in real time and feeds the activity back to the agent, enabling the agent to continuously understand the user's needs based on the feedback information.

[0122] Figure 7 This is a flowchart illustrating the generation process of the second recommended content proposed in an embodiment of this application. For example... Figure 7 As shown, the steps include the following.

[0123] Step S701: If the update conditions of the first recommended content are met, obtain the associated information corresponding to the first recommended content.

[0124] In some embodiments of this application, if no user interaction with the first recommended content is detected within a preset time period (this interaction could be a click or similar action), it indicates that the user may be dissatisfied with the displayed first recommended content and therefore chooses not to view it. In such cases, the update condition is determined to be met. The preset time period can be set according to actual circumstances, and this application does not impose any restrictions on it.

[0125] If a user is detected performing a preset action on the first recommended content, which could be one or more actions such as page turning or swiping, it indicates that the user may not be interested in the first recommended content and is looking at other content. In such cases, the update condition is determined to be met.

[0126] If no user viewing of the first recommended content is detected within the preset time period for displaying the first recommended content, but a preset operation is detected on the first recommended content, then the update condition is determined to be met.

[0127] Retrieve the associated information corresponding to the first recommended content. In one example, taking a list of air conditioners as the first recommended content, the agent uses the query API function of the tool module to retrieve information such as click-through rate, sales volume, price, title, and features for each air conditioner in the list as the associated information corresponding to the first recommended content.

[0128] Step S702: Based on the association information corresponding to the first recommended content, generate the analysis results for the first recommended content.

[0129] In some embodiments of this application, the associated information corresponding to the first recommended content is analyzed to obtain the reasons why the user did not view or click on the first recommended content, thereby generating the analysis results of the first recommended content. For example, it may be because the recommended product is too expensive or there are no discounts available.

[0130] Step S703: Based on the analysis results and the third historical data, generate second recommended content, which is then displayed on the first interface.

[0131] In some embodiments of this application, the analysis results include the reasons why the user did not view the first recommended content. The third historical data includes first historical data, second historical data, and the user's recent behavior data. This behavior data can be data generated by the user on the first interface after the first recommended content is generated. For example, if the first recommended content is displayed on the first interface, but the user performs a page-turning operation without viewing the first recommended content, then the data generated after the page-turning operation is the user's recent behavior data.

[0132] Based on the analysis results and third-party historical data, the intelligent agent can re-recommend content to the user, which is called the second recommended content. This second recommended content is then displayed on the first interface, replacing the first recommended content.

[0133] The above embodiments enable timely understanding of user intent and behavior, improving the accuracy and efficiency of recommendations, and enhancing the user experience.

[0134] Figure 8 This is a structural diagram of the content recommendation device provided in an embodiment of this application. The content recommendation device 800 may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the content recommendation device 800 may be stored in the memory of a computer device and executed by at least one processor to perform the content recommendation function.

[0135] In this embodiment, the content recommendation device 800 can be divided into multiple functional modules according to the functions it performs. These functional modules may include: a determination module 810, an acquisition module 820, and a generation module 830.

[0136] The determining module 810 is used to determine keywords based on the input information entered by the user on the first displayed interface; the obtaining module 820 is used to obtain first historical data generated by the user on the first interface based on the keywords and the first interface; the determining module 810 is also used to determine a second interface based on the keywords, the second interface being at least one interface on which the user performed historical operations based on the keywords on a non-first interface; the obtaining module 820 is also used to obtain second historical data generated by the user on the second interface, the second historical data including historical content recommended to the user through the second interface and the associated information corresponding to the historical content, the historical content including historical recommended content that has not been viewed by the user; the generating module 830 is used to determine first recommended content corresponding to the input information based on the first historical data, the historical content and the associated information corresponding to the historical content, the first recommended content being displayed on the first interface.

[0137] Based on any embodiment of this application, the method further includes: when the update conditions of the first recommended content are met, obtaining the associated information corresponding to the first recommended content; generating an analysis result of the first recommended content based on the associated information corresponding to the first recommended content; generating second recommended content based on the analysis result and third historical data, and displaying the second recommended content on the first interface.

[0138] Based on any embodiment of this application, it further includes: if no viewing operation of the first recommended content by the user is detected within a preset time period for displaying the first recommended content, it is determined that the update condition is met; and / or, if the user is detected performing a preset operation on the first recommended content, it is determined that the update condition is met; wherein, the preset operation includes one or more of page turning operations and swiping operations.

[0139] Based on any embodiment of this application, obtaining the second historical data generated by the user in the second interface includes: obtaining the historical content based on the keywords, wherein the historical content includes an object list constructed from multiple recommended objects; obtaining information related to each recommended object in the object list to obtain the association information corresponding to the historical content; and using the association information between the object list and the historical content as the second historical data.

[0140] Based on any embodiment of this application, the method further includes: using an intelligent agent to obtain the historical content and the associated information corresponding to the historical content, including: the intelligent agent calling the search function corresponding to the search control displayed on the second interface through a first data interface to search for the keywords and obtain the historical content; based on the multiple recommended objects in the historical content, the intelligent agent obtaining information related to each recommended object through a second data interface to obtain the associated information corresponding to the historical content.

[0141] Based on any embodiment of this application, determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the associated information corresponding to the historical content includes: if the input information is generated based on the user performing a query operation on the first interface, generating the query result based on the input information; predicting the user's intent based on the first historical data, the historical content, and the associated information corresponding to the historical content; generating a prediction control based on the intent; and displaying the query result and the prediction control through the first interface.

[0142] Based on any embodiment of this application, determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the associated information corresponding to the historical content includes: if the input information is generated based on the user performing a search operation on the first interface, generating a first result based on the first historical data, the historical content, the associated information corresponding to the historical content, and a preset sorting strategy; generating a second result based on the first historical data, the historical content, the associated information corresponding to the historical content, and the type of the keyword; and determining the first recommended content based on the first result and the second result.

[0143] The term "module" as used in this application refers to a series of computer program segments that can be executed by at least one processor and perform a fixed function, and which are stored in memory. In this embodiment, the limitations of the content recommendation device 800 can be found in the above-described limitations of the content recommendation method, and will not be repeated in detail here.

[0144] Another embodiment of this application also provides a computer device. Figure 9 This is a schematic diagram of the structure of the computer device provided in the embodiments of this application, such as... Figure 9 As shown, in one embodiment of this application, the computer device 90 can be a tablet computer, an augmented reality (AR) / virtual reality (VR) device, a laptop computer, a netbook, a server cluster, a large server, etc. This application embodiment does not limit the specific type of the computer device 90. The computer device 90 is used to carry devices such as… Figure 1 The e-commerce platform 10 shown.

[0145] like Figure 9 As shown, the computer device 90 may include, but is not limited to, a communication module 910, a memory 920, a processor 930, an input / output (I / O) interface 940, and a bus 950. The processor 930 is coupled to the communication module 910, the memory 920, and the I / O interface 940 via the bus 950.

[0146] Those skilled in the art will understand that the schematic diagram is merely an example of the computer device 90 and does not constitute a limitation on the computer device 90. It may include more or fewer components than shown, or combine certain components, or different components. For example, the computer device 90 may also include network access devices, etc.

[0147] The communication module 910 may include a wired communication module and / or a wireless communication module. The wired communication module may provide one or more wired communication solutions, such as Universal Serial Bus (USB) and Controller Area Network (CAN). The wireless communication module may provide one or more wireless communication solutions, such as Wireless Fidelity (Wi-Fi), Bluetooth (BT), mobile communication networks, Frequency Modulation (FM), Near Field Communication (NFC), and Infrared (IR) technologies.

[0148] The memory 920 can be used to store computer-readable instructions and / or modules. The processor 930 implements various functions of the computer device 90 by running or executing the computer-readable instructions and / or modules stored in the memory 920 and by calling the data stored in the memory 920. The memory 920 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device 90, etc. The memory 920 may include non-volatile and volatile memory, such as: hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other storage devices.

[0149] The memory 920 can be the external memory and / or internal memory of the computer device 90. Furthermore, the memory 920 can be a memory in physical form, such as a memory stick, a TF card (Trans-flash Card), etc.

[0150] Processor 930 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. Processor 930 is the computing core and control center of computer device 90, connecting various parts of the computer device 90 through various interfaces and lines, and executing the operating system of computer device 90, as well as various installed application programs and program code.

[0151] For example, computer-readable instructions can be divided into one or more modules / submodules / units, one or more of which are stored in memory 920 and executed by processor 930 to complete this application. One or more modules / submodules / units can be a series of computer-readable instruction segments capable of performing a specific function, which describe the execution process of the computer-readable instructions in computer device 90.

[0152] If the modules / units integrated in the computer device 90 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can also be implemented by instructing related hardware through computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when executed by a processor, the computer-readable instructions can implement the steps of the various method embodiments described above.

[0153] Computer-readable instructions include computer-readable instruction code, which can be in the form of source code, object code, executable file, or some intermediate form. Computer-readable media can include: any entity or device capable of carrying computer-readable instruction code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), and random access memory (RAM).

[0154] The memory 920 in the computer device 90 stores computer-readable instructions. The specific implementation method of the processor 930 for these computer-readable instructions can be found in [reference needed]. Figure 2 , Figure 7 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0155] I / O interface 940 is used to provide a channel for user input or output. For example, I / O interface 940 can be used to connect various input and output devices, such as mouse, keyboard, touch device, display screen, etc., so that users can enter information or visualize information.

[0156] Bus 950 is used at least to provide a channel for communication between communication modules 910, memory 920, processor 930 and I / O interface 940 in computer device 90.

[0157] Furthermore, the functional modules 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 in the form of hardware plus software functional modules.

[0158] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within this application. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0159] Furthermore, it is clear that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices can also be implemented by a single unit or device through software or hardware. Terms such as "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0160] This application also provides a computer-readable storage medium storing a computer program, which includes program instructions. When the program instructions are executed, the method implemented can refer to the methods in the above embodiments of this application.

[0161] The computer-readable storage medium can be the internal memory of the electronic device described in the above embodiments, such as the hard disk or memory of the electronic device. Alternatively, the computer-readable storage medium can be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., provided on the electronic device.

[0162] In some embodiments, a computer-readable storage medium may include a stored program area and a stored data area, wherein the stored program area may store an operating system, an application program required for at least one function, etc.; and the stored data area may store data created based on the use of the electronic device, etc.

[0163] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0164] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0165] In the embodiments provided in this application, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0166] 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 network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A content recommendation method, characterized in that, The method includes: Keywords are determined based on the input information entered by the user on the first displayed interface; Based on the keywords and the first interface, obtain the first historical data generated by the user on the first interface; A second interface is determined based on the keyword, and the second interface is at least one interface on which the user performs historical operations based on the keyword on a non-first interface. Acquire the second historical data generated by the user in the second interface. The second historical data includes historical content recommended to the user through the second interface and related information corresponding to the historical content. The historical content includes historical recommended content that has not been viewed by the user. Based on the first historical data, the historical content and the associated information corresponding to the historical content, the first recommended content corresponding to the input information is determined, and the first recommended content is displayed on the first interface.

2. The content recommendation method according to claim 1, characterized in that, The method further includes: If the update conditions for the first recommended content are met, obtain the associated information corresponding to the first recommended content; Based on the association information corresponding to the first recommended content, an analysis result of the first recommended content is generated; Based on the analysis results and the third historical data, a second set of recommended content is generated and displayed on the first interface.

3. The content recommendation method according to claim 2, characterized in that, The method further includes: If no user viewing operation on the first recommended content is detected within the preset time period for displaying the first recommended content, it is determined that the update condition is met; and / or, If it is detected that the user performs a preset operation on the first recommended content, it is determined that the update condition is met; wherein, the preset operation includes one or more of page turning operations and swiping operations.

4. The content recommendation method according to claim 1, characterized in that, The step of obtaining the second historical data generated by the user in the second interface includes: The historical content is obtained based on the keywords, and the historical content includes an object list constructed from multiple recommended objects; Obtain information related to each recommended object in the object list to obtain the association information corresponding to the historical content; The association information between the object list and the historical content is used as the second historical data.

5. The content recommendation method according to claim 4, characterized in that, The method further includes: using an intelligent agent to obtain the association information between the historical content and the corresponding historical content, including: The intelligent agent calls the search function corresponding to the search control displayed on the second interface through the first data interface to search for the keywords and obtain the historical content; Based on the multiple recommended objects in the historical content, the intelligent agent obtains information related to each recommended object through the second data interface to obtain the association information corresponding to the historical content.

6. The content recommendation method according to claim 1, characterized in that, The step of determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the association information corresponding to the historical content includes: If the input information is generated based on the user performing a query operation on the first interface, the query result is generated based on the input information; Based on the first historical data, the historical content, and the associated information corresponding to the historical content, the user's intent is predicted; Generate a prediction control based on the stated intent; The query results and the prediction control are displayed on the first interface.

7. The content recommendation method according to claim 1, characterized in that, The step of determining the first recommended content corresponding to the input information based on the first historical data, the historical content, and the association information corresponding to the historical content includes: If the input information is generated based on the user performing a search operation on the first interface, a first result is generated based on the first historical data, the historical content, the associated information corresponding to the historical content, and a preset sorting strategy; A second result is generated based on the first historical data, the historical content, the associated information corresponding to the historical content, and the type of the keyword. Based on the first result and the second result, the first recommended content is determined.

8. A content recommendation device, characterized in that, The content recommendation device includes: The determination module is used to determine keywords based on the input information entered by the user on the first displayed interface; The acquisition module is used to acquire the first historical data generated by the user on the first interface based on the keywords and the first interface; The determining module is further configured to determine a second interface based on the keyword, wherein the second interface is at least one interface on which the user performs historical operations based on the keyword on a non-first interface; The acquisition module is further configured to acquire second historical data generated by the user in the second interface. The second historical data includes historical content recommended to the user through the second interface and related information corresponding to the historical content. The historical content includes historical recommended content that has not been viewed by the user. The generation module is used to determine the first recommended content corresponding to the input information based on the first historical data, the historical content and the association information corresponding to the historical content, and the first recommended content is displayed on the first interface.

9. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, and the processor, when executing the computer program, implements the content recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction that, when executed by a processor, implements the content recommendation method as described in any one of claims 1 to 7.

Images