Intelligent recommendation method and system based on artificial intelligence

Abstract

The invention is suitable for the technical field of commodity recommendation, and particularly relates to an intelligent recommendation method and system based on artificial intelligence, and the method comprises the steps: intercepting a friend list of a user in an e-commerce platform, carrying out the statistics of the chat frequency of the user with each friend, obtaining the personal data uploaded by the user, and storing the personal data in the e-commerce platform; wherein the personal data at least comprises a preference label, integrating the personal data and chat frequency, marking close friends, and clustering the close friends into a plurality of anonymous groups by using the preference label. By dividing the core layer and the edge layer, the exposure rate of preferred commodities, especially new commodities or unpopular commodities, can be remarkably improved, so that information cocoon rooms are avoided, meanwhile, the core layer focuses on group sharing more, the edge layer focuses on privacy protection and personal experience more, balance between friend recommendation and personal privacy is achieved, and user experience is improved. Therefore, the user satisfaction and the platform benefit are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of commodity recommendation, and in particular to an intelligent recommendation method and system based on artificial intelligence. Background Art

[0002] In e-commerce shopping, e-commerce platforms generally recommend products that users may be interested in by analyzing their purchases, browsing histories or common relationships between products. However, this behavior-based recommendation method is often more mechanical and lacks "warmth", making it difficult to trigger users' emotional resonance. In addition, if only products similar to the user's historical behavior are recommended, there is a lack of diversity, which may lead to a monotonous user experience, making it difficult to bring freshness and surprises, and may even cause information cocoons, greatly reducing the user experience.

[0003] Therefore, “how to recommend products based on friend relationships” is a technical problem that the present invention needs to solve. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent recommendation method and system based on artificial intelligence to solve the problem of "how to recommend products based on friend relationships" raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: An intelligent recommendation method based on artificial intelligence, the method comprising: S100: intercepting a friend list of a user on an e-commerce platform, counting the chat frequency between the user and each friend, obtaining a personal profile uploaded by the user, wherein the personal profile at least includes: a preference tag, integrating the personal profile and the chat frequency, marking close friends, and clustering the close friends into a number of anonymous groups using the preference tags, defining the preference tags corresponding to the anonymous groups as preferences, creating nodes corresponding to the anonymous groups one by one, and writing the preferences into the nodes; S200: construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output a recommendation list, and determine the browsing items; S300: Divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

[0006] Furthermore, the S100 includes: Configure a time window, count the number of chats between the user and each friend within the time window, and calculate the chat frequency; The chat frequencies are integrated, a relationship network is constructed, and the close friends are corrected.

[0007] Furthermore, the S100 further includes: Traversing the influencing factors of close friends, wherein the influencing factors at least include: order records and chat duration; Based on the influencing factors, the relationship network is dynamically adjusted.

[0008] Further, the S200 includes: In the anonymous group, all close friends are sorted according to the chat frequency; Based on the sorting results, priority processing rules are constructed and multicast to all nodes.

[0009] Furthermore, the S200 further includes: Based on the order records, the number of orders placed by close friends in each node is counted, and the anonymous group is divided into a number of segments; The recommendation frequency is set, and a mapping between the recommendation frequency and the segment is established.

[0010] Furthermore, the multi-level recommendation mechanism is: Using the instant request, determining a recommendation list and browsing items, defining the close friend who initiated the instant request as a target friend, determining whether the target friend belongs to the core layer, and if so, pushing the browsing item to all close friends in the core layer; If not, the recommendation list is pushed to all close friends in the edge layer.

[0011] Further, the S300 includes: Integrate the recommendation list or browsing items into a preset template to generate a recommendation pop-up window; A trigger mechanism is established and embedded in the node, and the recommendation pop-up windows are pushed in parallel using the trigger mechanism.

[0012] Furthermore, the system comprises: A writing module is used to intercept the user's friend list in the e-commerce platform, count the chat frequency between the user and each friend, obtain the personal information uploaded by the user, wherein the personal information at least includes: a preference tag, integrate the personal information and the chat frequency, mark the close friends, and use the preference tags to cluster the close friends into several anonymous groups, define the preference tags corresponding to the anonymous groups as preferences, create nodes corresponding to the anonymous groups one by one, and write the preferences into the nodes; An output module is used to construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output the recommendation list, and determine the browsing items; The activation module is used to divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

[0013] Furthermore, the writing module includes: A calculation unit, used to configure a time window, count the number of chats between the user and each friend in the time window, and calculate the chat frequency; A construction unit, used to integrate the chat frequencies, construct a relationship network, and correct the close friends; A traversal unit, used to traverse the influencing factors of close friends, wherein the influencing factors at least include: order records and chat time; An adjustment unit is used to dynamically adjust the relationship network according to the influencing factors.

[0014] Furthermore, the output module includes: A sorting unit, used to sort all close friends in the anonymous group according to the chat frequency; A multicast unit, used for constructing a priority processing rule according to the sorting result, and multicasting the priority processing rule to all nodes; A segmentation unit, for counting the number of orders placed by close friends in each node according to the order records, and dividing the anonymous group into a plurality of segments; The mapping unit is used to set the recommendation frequency and establish a mapping between the recommendation frequency and the segment.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By identifying close friends, the relevance and persuasiveness of recommendations can be greatly improved, providing users with a warmer recommendation experience. At the same time, friend recommendations will increase emotional connections and enhance the shopping experience. By building anonymous groups, it is possible to bring together user groups with similar interests and improve the accuracy of recommendations, while also avoiding privacy and security issues caused by direct exposure of personal information or friend relationships. By dividing the core layer and the edge layer, the exposure rate of preferred products can be significantly improved, especially new or unpopular products, thereby avoiding the emergence of information cocoons. At the same time, the core layer focuses more on group sharing, while the edge layer pays more attention to privacy protection and personal experience, achieving a balance between friend recommendations and personal privacy, greatly improving user satisfaction and platform benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of an artificial intelligence-based intelligent recommendation method provided by an embodiment of the present invention; Figure 2 A first sub-flow chart of the intelligent recommendation method based on artificial intelligence provided by an embodiment of the present invention; Figure 3 A second sub-flow diagram of the intelligent recommendation method based on artificial intelligence provided by an embodiment of the present invention; Figure 4 A third sub-flow chart of the intelligent recommendation method based on artificial intelligence provided by an embodiment of the present invention; Figure 5 A block diagram of the composition of an artificial intelligence-based intelligent recommendation system provided by an embodiment of the present invention; Figure 6 A block diagram of the composition of a writing module in an artificial intelligence-based intelligent recommendation system provided by an embodiment of the present invention; Figure 7 A block diagram of the composition of an output module in an artificial intelligence-based intelligent recommendation system provided by an embodiment of the present invention; Figure 8 A block diagram of the composition of the activation module in the artificial intelligence-based intelligent recommendation system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In Example 1, Figure 1 The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown and described in detail below: S100: intercepting the user's friend list on the e-commerce platform, counting the chat frequency between the user and each friend, obtaining the personal information uploaded by the user, wherein the personal information at least includes: preference tags, integrating the personal information and chat frequency, marking close friends, and using the preference tags to cluster close friends into several anonymous groups, defining the preference tags corresponding to the anonymous groups as preferences, creating nodes corresponding to the anonymous groups one by one, and writing the preferences into the nodes.

[0019] A social module is integrated into the e-commerce platform, and friend relationships are established to determine the user's friend list; the number of chats the user has with each friend within a preset time period is recorded, and the chat frequency within the preset time period is calculated, wherein conversations with a reply time less than a preset threshold are counted as one chat; the personal information uploaded by the user is obtained, and the personal information includes: city of residence, age, and preference tags, etc., wherein the preference tags are the basis for product recommendations, and the preference tags are composed of information such as the user's purchase history, browsing habits, and brands or categories of interest on the e-commerce platform, which can reflect the user's interests and preferences; close friends are selected from the friend list based on the chat frequency, and all close friends are clustered according to the preference tags corresponding to each close friend to obtain several groups, and all users in the group are anonymized; wherein each anonymous group corresponds to a preference tag, in other words, the users in the anonymous group have a common preference; a node is created for each anonymous group, and a corresponding relationship between anonymous groups, preferences, and nodes is established, wherein the node is the basic structure for storing anonymous group information.

[0020] S200: Construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output the recommendation list, and determine the browsing items.

[0021] Based on the preference, create a keyword set for each node, where the keyword set is a combination of all characteristic words about the preference; for example, if the preference of all users in a certain node is football supplies, then the keyword set should include characteristic words such as football shoes, autumn clothes, training equipment and fan accessories; obtain the query request of close friends, where the query request is the user's query data in the e-commerce platform, which contains one or more keywords, input the query data into the AI ​​recommendation model, generate a recommendation list, record the products browsed by close friends, and determine them as browsed items.

[0022] For example, if the query request is football, the football is input into the AI ​​recommendation model to obtain a recommended list of football products. When close friends browse, the browsed products are recorded to obtain browsed items.

[0023] S300: Divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

[0024] The close friends in each node are divided into the core layer and the edge layer, and a multi-level recommendation mechanism is constructed. When an instant request uploaded by a friend is received, the multi-level recommendation mechanism is activated.

[0025] For example, user A's friend list is divided into three anonymous groups, A, B and C, where A prefers football, B prefers basketball, and C prefers cycling. Each anonymous group is divided into a core layer and an edge layer. B is a member of the core layer in anonymous group A. B initiates an instant request on the e-commerce platform. The instant request is for products related to football cleaning. The instant request is input into the AI ​​recommendation model, and a recommended list of products related to football cleaning is output. When B browses one of the products, the product is defined as a browsing item; using the browsing item, a recommendation pop-up window is generated, and yes and no buttons are integrated into the recommendation pop-up window. When B clicks the yes button, this recommendation pop-up window is sent to all members in the core layer of anonymous group A. If B is a member of the edge layer, the recommendation list is sent to all members in the edge layer of anonymous group A.

[0026] In Example 2, Figure 2 The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown, and S100 is described in detail as follows: S101: configuring a time window, counting the number of chats between the user and each friend in the time window, and calculating the chat frequency.

[0027] Set the initial time and end time, determine the time window, count the number of chats between the user and each friend within the time window, combine the number of chats with the time window to generate the chat frequency; for example, the chat frequency is once every three days, or five times a day, etc.

[0028] S102: Integrate the chat frequencies, build a relationship network, and correct the close friends.

[0029] A relationship network is constructed with each friend as a point and the relationship between friends as an edge. The relationship network is used to show the relationships between friends, such as colleague relationships and relative relationships. The advantage of doing this is to improve the accuracy of the definition of close friends. In real life, it is one-sided to determine the degree of intimacy based solely on the frequency of chats. For example, the chat frequency between friends with business contacts is high, but the degree of intimacy may be average.

[0030] In Example 3, Figure 2The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown, and S100 is further described in detail below, as follows: S103: Traverse out influencing factors of close friends, where the influencing factors at least include: order records and chat duration.

[0031] Determine the influencing factors of close friends, where the influencing factors include order records and chat duration, etc.; for example, a user chats frequently with a close friend, but the close friend has never placed an order on the e-commerce platform, so the close friend should not be classified into any anonymous group.

[0032] S104: Dynamically adjust the relationship network based on the influencing factors.

[0033] According to the influencing factors, the relationship network is adjusted and non-potential users in the e-commerce platform are deleted from the relationship network.

[0034] In Example 4, Figure 3 The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown, and S200 is described in detail below, as follows: S201: In the anonymous group, all close friends are sorted according to the chat frequency.

[0035] Sort all close friends in an anonymous group by chat frequency, from highest to lowest.

[0036] S202: Based on the sorting result, construct a priority processing rule, and multicast the priority processing rule to all nodes.

[0037] Construct a priority processing rule, where the priority processing rule is: when receiving instant requests uploaded by multiple users, give priority to instant requests corresponding to close friends with higher chat frequency.

[0038] In Example 5, Figure 3 The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown, and S200 is further described in detail below, as follows: S203: Based on the order records, count the number of orders placed by close friends in each node, and divide the anonymous group into several segments.

[0039] Record the number of orders placed by each close friend in the node, and divide the close friends into several segments based on the number of orders.

[0040] S204: Setting the recommendation frequency, and establishing a mapping between the recommendation frequency and the segment.

[0041] Configure a recommendation frequency for each segment, where the recommendation frequency is determined by the user; if the average number of orders placed by close friends in an anonymous group is high, a larger recommendation frequency should be set, that is, recommendation pop-ups should be sent more frequently to all close friends in the anonymous group.

[0042] In Example 6, the multi-level recommendation mechanism is: Using the instant request, determining a recommendation list and browsing items, defining the close friend who initiated the instant request as a target friend, determining whether the target friend belongs to the core layer, and if so, pushing the browsing item to all close friends in the core layer; If not, the recommendation list is pushed to all close friends in the edge layer.

[0043] As described in the example of S300, if the target friend belongs to the core layer, the browsing item is pushed to all close friends in the core layer, and if the target friend belongs to the edge layer, the recommendation list is sent to all close friends in the edge layer.

[0044] In Example 7, Figure 4 The implementation process of the intelligent recommendation method based on artificial intelligence provided by the embodiment of the present invention is shown, and S300 is further described in detail as follows: S301: Integrate the recommendation list or browsing items into a preset template to generate a recommendation pop-up window.

[0045] When the user retrieves the recommended list and browsing items through instant request, the search results are integrated into the preset template to generate the recommendation pop-up window. The preset template is pre-made by professionals.

[0046] S302: Establish a trigger mechanism and embed it into the node, and use the trigger mechanism to push the recommendation pop-up windows in parallel.

[0047] A trigger mechanism is embedded in the node, where the trigger mechanism is a specific push method for the recommended pop-up window, such as pushing within a preset time period.

[0048] Figure 5 The figure shows a structural block diagram of an intelligent recommendation system based on artificial intelligence provided by an embodiment of the present invention. The intelligent recommendation system based on artificial intelligence 1 includes: The writing module 11 is used to intercept the friend list of the user in the e-commerce platform, count the chat frequency between the user and each friend, obtain the personal information uploaded by the user, wherein the personal information at least includes: preference tags, integrate the personal information and chat frequency, mark close friends, and use the preference tags to cluster the close friends into several anonymous groups, define the preference tags corresponding to the anonymous groups as preferences, create nodes corresponding to the anonymous groups one by one, and write the preferences into the nodes; Output module 12, used to construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output the recommendation list, and determine the browsing items; The activation module 13 is used to divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

[0049] Figure 6 The structure diagram of the composition of the intelligent recommendation system based on artificial intelligence provided by the embodiment of the present invention is shown, and the writing module 11 includes: The calculation unit 111 is used to configure a time window, count the number of chats between the user and each friend in the time window, and calculate the chat frequency; A construction unit 112 is used to integrate the chat frequencies, construct a relationship network, and divide the close friends into high frequency, medium frequency, and low frequency; A traversal unit 113, used to traverse the influencing factors of close friends, wherein the influencing factors at least include: order records and chat time; The adjusting unit 114 is used to dynamically adjust the relationship network according to the influencing factors.

[0050] Figure 7 The structure diagram of the composition of the intelligent recommendation system based on artificial intelligence provided by the embodiment of the present invention is shown, and the output module 12 includes: A sorting unit 121, configured to sort all close friends in the anonymous group according to the chat frequency; The multicast unit 122 is used to construct a priority processing rule according to the sorting result, and multicast the priority processing rule to all nodes; A segmentation unit 123, configured to count the number of orders placed by the user in each node according to the order record, and to segment the anonymous group into a plurality of segments; The mapping unit 124 is used to set the recommendation frequency and establish a mapping between the recommendation frequency and the segment.

[0051] Figure 8 The structure diagram of the composition of the intelligent recommendation system based on artificial intelligence provided by the embodiment of the present invention is shown, and the activation module 13 includes: A generating unit 131, configured to integrate the recommendation list or browsing items into a preset template to generate a recommendation pop-up window; The push unit 132 is used to establish a trigger mechanism and embed it into the node, and use the trigger mechanism to push the recommendation pop-up windows in parallel.

[0052] The writing module 11 is mainly used to complete step S100, the output module 12 is mainly used to complete step S200, and the activation module 13 is mainly used to complete step S300; The calculation unit 111 is mainly used to complete step S101, the construction unit 112 is mainly used to complete step S102, the traversal unit 113 is mainly used to complete step S103, and the adjustment unit 114 is mainly used to complete step S104; The sorting unit 121 is mainly used to complete step S201, the multicast unit 122 is mainly used to complete step S202, the segmentation unit 123 is mainly used to complete step S203, and the mapping unit 124 is mainly used to complete step S204; The generating unit 131 is mainly used to complete step S301, and the pushing unit 132 is mainly used to complete step S302.

[0053] The functions that can be achieved by the above-mentioned artificial intelligence-based intelligent recommendation system are all completed by a computer device, which includes one or more processors and one or more memories, and at least one program code is stored in the one or more memories. The program code is loaded and executed by the one or more processors to realize the functions of the artificial intelligence-based intelligent recommendation system.

[0054] The processor takes out instructions from the memory one by one, analyzes the instructions, and then completes the corresponding operations according to the instruction requirements, generating a series of control commands, so that the various parts of the computer can automatically, continuously and coordinately move to become an organic whole, realize the input of programs, the input of data, and the calculation and output of results. The arithmetic operations or logical operations generated in this process are all completed by the operator; the memory includes a read-only memory, which is used to store computer programs, and a protection device is provided outside the memory.

[0055] Exemplarily, the computer program may be divided into one or more modules, one or more modules are stored in a memory and executed by a processor to implement the present invention. One or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in a terminal device.

[0056] Those skilled in the art will understand that the description of the above service equipment is merely an example and does not constitute a limitation on the terminal equipment. It may include more or fewer components than described above, or a combination of certain components, or different components, for example, it may include input and output devices, network access devices, buses, etc.

[0057] The processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire user terminal.

[0058] The above-mentioned memory can be used to store computer programs and / or modules. The above-mentioned processor realizes various functions of the above-mentioned terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as an information collection template display function, a product information release function, etc.), etc.; the data storage area can store data created according to the use of the berth status display system (such as product information collection templates corresponding to different product types, product information that different product providers need to release, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0059] If the module / unit integrated in the terminal device is implemented in the form of 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 present invention implements all or part of the modules / units in the above-mentioned embodiment system, and can also be completed by instructing the relevant hardware through a computer program. The above-mentioned computer program can be stored in a computer-readable storage medium, and the computer program can realize the functions of the above-mentioned various system embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0060] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0061] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent recommendation method based on artificial intelligence, characterized in that: The method comprises: S100: intercepting a friend list of a user on an e-commerce platform, counting the chat frequency between the user and each friend, obtaining a personal profile uploaded by the user, wherein the personal profile at least includes: a preference tag, integrating the personal profile and the chat frequency, marking close friends, and clustering the close friends into a number of anonymous groups using the preference tags, defining the preference tags corresponding to the anonymous groups as preferences, creating nodes corresponding to the anonymous groups one by one, and writing the preferences into the nodes; S200: construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output a recommendation list, and determine the browsing items; S300: Divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

2. The intelligent recommendation method based on artificial intelligence according to claim 1, characterized in that: The S100 includes: Configure a time window, count the number of chats between the user and each friend within the time window, and calculate the chat frequency; The chat frequencies are integrated, a relationship network is constructed, and the close friends are corrected.

3. The intelligent recommendation method based on artificial intelligence according to claim 2, characterized in that: The S100 further includes: Traversing the influencing factors of close friends, wherein the influencing factors at least include: order records and chat duration; Based on the influencing factors, the relationship network is dynamically adjusted.

4. The intelligent recommendation method based on artificial intelligence according to claim 1, characterized in that: The S200 includes: In the anonymous group, all close friends are sorted according to the chat frequency; Based on the sorting results, priority processing rules are constructed and multicast to all nodes.

5. The intelligent recommendation method based on artificial intelligence according to claim 3, characterized in that: The S200 further includes: Based on the order records, the number of orders placed by close friends in each node is counted, and the anonymous group is divided into a number of segments; The recommendation frequency is set, and a mapping between the recommendation frequency and the segment is established.

6. The intelligent recommendation method based on artificial intelligence according to claim 1, characterized in that: The multi-level recommendation mechanism is: Using the instant request, determining a recommendation list and browsing items, defining the close friend who initiated the instant request as a target friend, determining whether the target friend belongs to the core layer, and if so, pushing the browsing item to all close friends in the core layer; If not, the recommendation list is pushed to all close friends in the edge layer.

7. The intelligent recommendation method based on artificial intelligence according to claim 1, characterized in that: The S300 includes: Integrate the recommendation list or browsing items into a preset template to generate a recommendation pop-up window; A trigger mechanism is established and embedded in the node, and the recommendation pop-up windows are pushed in parallel using the trigger mechanism.

8. An intelligent recommendation system based on artificial intelligence, characterized in that: The system comprises: A writing module is used to intercept the user's friend list in the e-commerce platform, count the chat frequency between the user and each friend, obtain the personal information uploaded by the user, wherein the personal information at least includes: a preference tag, integrate the personal information and the chat frequency, mark the close friends, and use the preference tags to cluster the close friends into several anonymous groups, define the preference tags corresponding to the anonymous groups as preferences, create nodes corresponding to the anonymous groups one by one, and write the preferences into the nodes; An output module is used to construct a keyword set, configure the corresponding relationship between the keyword set and the node, obtain the read permission of all close friends' query requests in the e-commerce platform, input the query request into the pre-built AI recommendation model, output the recommendation list, and determine the browsing items; The activation module is used to divide each of the nodes into a core layer and an edge layer, integrate the recommendation list, browsing items, the core layer and the edge layer, build a multi-level recommendation mechanism, receive instant requests uploaded by the close friends, and activate the multi-level recommendation mechanism.

9. The artificial intelligence-based intelligent recommendation system according to claim 8, characterized in that: The writing module comprises: A calculation unit, used to configure a time window, count the number of chats between the user and each friend in the time window, and calculate the chat frequency; A construction unit, used to integrate the chat frequencies, construct a relationship network, and correct the close friends; A traversal unit, used to traverse the influencing factors of close friends, wherein the influencing factors at least include: order records and chat time; An adjustment unit is used to dynamically adjust the relationship network according to the influencing factors.

10. The intelligent recommendation system based on artificial intelligence according to claim 9, characterized in that: The output module comprises: A sorting unit, used to sort all close friends in the anonymous group according to the chat frequency; A multicast unit, used for constructing a priority processing rule according to the sorting result, and multicasting the priority processing rule to all nodes; A segmentation unit, for counting the number of orders placed by close friends in each node according to the order records, and dividing the anonymous group into a plurality of segments; The mapping unit is used to set the recommendation frequency and establish a mapping between the recommendation frequency and the segment.

Images