A topic intelligent recommendation method and device, equipment and medium

Abstract

This application relates to the field of data mining technology, specifically to an intelligent topic recommendation method, apparatus, device, and medium, used to address the problem of low user response rates in traditional topic recommendation methods. The method involves: based on the target user's user information and topic filtering conditions, obtaining various topic categories for communication with the target user; and based on each topic category, a topic library, and preset importance of each topic category, obtaining a predicted score for each topic content within each category; then, sorting the predicted scores for each topic content to obtain the target topic content for communication with the target user. The topic library includes each topic content within each topic category and the rules for determining the predicted score; a higher predicted score indicates a higher response rate for the topic content. Thus, using the target topic content for communication can effectively attract the target user and improve the user response rate.

Description

Technical Field

[0001] This application relates to the field of data mining technology, specifically to an intelligent recommendation method, apparatus, device, and medium for a specific topic. Background Technology

[0002] In some application scenarios, a service provider may offer consulting, agency, and other services to numerous users. Due to the large number of users, service providers often struggle with choosing communication topics, or the topics used during communication may not accurately address the user's concerns, resulting in the communication failing to achieve the desired effect, such as maintaining relationships or recommending products.

[0003] However, with the relevant technologies, topic recommendation methods typically involve direct communication with users through chatbots and intelligent Q&A systems. Since these methods usually communicate with a large number of users on topics such as birthday reminders, they often fail to elicit positive user responses and are not effective in attracting customers, resulting in a low user response rate. Summary of the Invention

[0004] This application provides a method, apparatus, device, and medium for intelligent topic recommendation, which is used to improve the user response rate of topics.

[0005] In a first aspect, an intelligent topic recommendation method provided in the embodiments of this application includes:

[0006] Based on the target user's user information and topic filtering conditions, various topic categories for communication with the target user are obtained, wherein the user information includes some or all of the target user's virtual resource information, various reminder information and behavioral information;

[0007] Based on the topic categories, the topic library, and the preset importance of each topic category, a predicted score is obtained for each topic content included in each topic category. The topic library includes each topic content included in each topic category and the prediction score determination rules. The higher the prediction score, the higher the response rate of the topic content.

[0008] The predicted scores for each topic are sorted to obtain the target topic content for communication with the target user.

[0009] In one possible embodiment, before obtaining the topic categories used for communication with the target user based on the target user's user information and topic filtering conditions, the method further includes:

[0010] After determining that preset conditions are met, at least one target user is selected from the user database based on the preset conditions, wherein the preset conditions include some or all of the following: periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions; or,

[0011] In response to a user filtering instruction, based on the service identification information contained in the user filtering instruction, at least one target user associated with the service identification information is filtered from the user database, wherein the service identification information is the identification information of the service provider that provides services to the target user.

[0012] In one possible embodiment, the topic filtering criteria include some or all of the following criteria:

[0013] The expiration time of any virtual resource information of the target user is less than the first threshold from the current time;

[0014] The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold;

[0015] The reminder time of any reminder message for the target user is less than the second threshold from the current time.

[0016] In one possible embodiment, obtaining the predicted score corresponding to each topic content included in each topic category based on the topic categories, the topic library, and the preset importance of each topic category includes:

[0017] Based on the aforementioned topic categories, determine the topic content contained in each topic category from the topic categories included in the topic library;

[0018] Obtain the number of times each topic's content has been cited and the number of times it has received user responses;

[0019] Based on the predicted score determination rules, the preset importance of each topic category, and the number of times each topic content is cited and the number of user responses, the predicted score for each topic content included in each topic category is obtained.

[0020] In one possible embodiment, sorting the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target user includes:

[0021] The predicted scores of each topic are sorted in descending order of predicted scores and according to preset sorting optimization rules, wherein the preset sorting optimization rules include topic category priority rules and / or topic content priority configuration rules.

[0022] The topic that appears first in the ranking information will be used as the target topic for communication with the target user.

[0023] In one possible embodiment, after obtaining the target topic content for communication with the target user, the method further includes:

[0024] Output the topic category to which the target topic content belongs, and the user information corresponding to the target topic content.

[0025] Secondly, an intelligent topic recommendation device provided in this application includes:

[0026] The topic category determination module is used to obtain various topic categories for communication with the target user based on the target user's user information and topic filtering conditions, wherein the user information includes some or all of the target user's virtual resource information, various reminder information and behavioral information;

[0027] The score prediction module is used to obtain a predicted score for each topic content contained in each topic category based on the topic categories, the topic library, and the preset importance of each topic category. The topic library includes each topic content contained in each topic category and the prediction score determination rules. The higher the prediction score, the higher the response rate of the topic content.

[0028] The sorting module is used to sort the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target user.

[0029] In one possible embodiment, before obtaining the topic categories used for communication with the target user based on the target user's user information and topic filtering conditions, the topic category determination module is further configured to:

[0030] After determining that preset conditions are met, at least one target user is selected from the user database based on the preset conditions, wherein the preset conditions include some or all of the following: periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions; or,

[0031] In response to a user filtering instruction, based on the service identification information contained in the user filtering instruction, at least one target user associated with the service identification information is filtered from the user database, wherein the service identification information is the identification information of the service provider that provides services to the target user.

[0032] In one possible embodiment, the topic filtering criteria include some or all of the following criteria:

[0033] The expiration time of any virtual resource information of the target user is less than the first threshold from the current time;

[0034] The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold;

[0035] The reminder time of any reminder message for the target user is less than the second threshold from the current time.

[0036] In one possible embodiment, the score prediction module is specifically used for:

[0037] Based on the aforementioned topic categories, determine the topic content contained in each topic category from the topic categories included in the topic library;

[0038] Obtain the number of times each topic's content has been cited and the number of times it has received user responses;

[0039] Based on the predicted score determination rules, the preset importance of each topic category, and the number of times each topic content is cited and the number of user responses, the predicted score for each topic content included in each topic category is obtained.

[0040] In one possible embodiment, the sorting module is specifically used for:

[0041] The predicted scores of each topic are sorted in descending order of predicted scores and according to preset sorting optimization rules, wherein the preset sorting optimization rules include topic category priority rules and / or topic content priority configuration rules.

[0042] The topic that appears first in the ranking information will be used as the target topic for communication with the target user.

[0043] In one possible embodiment, after obtaining the target topic content for communication with the target user, the sorting module is further configured to:

[0044] Output the topic category to which the target topic content belongs, and the user information corresponding to the target topic content.

[0045] Thirdly, this application provides an electronic device, comprising:

[0046] Memory, used to store program instructions;

[0047] A processor is configured to invoke program instructions stored in the memory and execute the steps of the method described in any one of the first aspects according to the obtained program instructions.

[0048] Fourthly, this application provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the method described in any one of the first aspects.

[0049] Fifthly, this application provides a computer program product comprising: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of the first aspects.

[0050] The embodiments of this application have the following beneficial effects:

[0051] By using topic filtering criteria, we can obtain various topic categories suitable for communication with target users. Then, based on the topic categories included in the topic library, the content of each topic within each category, the prediction score determination rules, and the importance of each topic category, we can obtain the predicted scores for the content of each topic within each category for communication with target users. A higher predicted score indicates a higher user response rate for the topic content. Then, by sorting the predicted scores for each topic content, we can obtain the target topic content for communication with target users. In this way, because the process of obtaining the target topic content comprehensively considers the various topic types and content included in the topic library, as well as the importance of each topic category, the target topic content is more aligned with the interests of target users. Communicating with target users based on this target topic content can effectively attract them, thereby improving the user response rate. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application;

[0053] Figure 2 A flowchart illustrating an intelligent topic recommendation method provided in an embodiment of this application;

[0054] Figure 3 A flowchart illustrating a method for determining a predicted score, as provided in an embodiment of this application;

[0055] Figure 4 A flowchart illustrating a method for sorting predicted scores, provided in an embodiment of this application;

[0056] Figure 5 A structural diagram of a topic-based intelligent recommendation device provided in an embodiment of this application;

[0057] Figure 6 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0059] The terms "first" and "second" in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising" and any variations thereof are intended to cover non-exclusive protection. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. The term "multiple" in this application can mean at least two, for example, two, three, or more, and the embodiments of this application do not impose limitations.

[0060] The data collection, dissemination, and use in this application all comply with relevant national laws and regulations.

[0061] Before introducing the intelligent recommendation method for topics provided in the embodiments of this application, the technical background of the embodiments of this application will be described in detail below for ease of understanding.

[0062] In some application scenarios, a service provider may need to provide consulting, agency, and other services to numerous users. Due to the large number of users and the busy workload, if there has been no communication with a particular user for a period of time, the service provider may become anxious about how to choose topics to discuss with that user, or the topics used during the communication may not accurately address the user's concerns, resulting in the communication failing to achieve the expected results, such as maintaining relationships or recommending products, or even causing business loss.

[0063] However, with the relevant technologies, topic recommendation methods typically involve direct communication with users through chatbots and intelligent Q&A systems. Since these methods usually communicate with a large number of users on topics such as birthday reminders, they often fail to elicit positive user responses and are not effective in attracting customers, resulting in a low user response rate.

[0064] To address this, this application provides an intelligent topic recommendation method. Based on the target user's user information and topic filtering criteria, various topic categories for communication with the target user are obtained. The user information includes some or all of the target user's virtual resource information, various reminder messages, and behavioral information. Then, based on each topic category, a topic library, and preset rules for determining the importance prediction score of each topic category, a predicted score is obtained for each topic content within each category. The predicted scores for each topic content are then sorted to obtain the target topic content for communication with the target user. The topic library includes the topic content within each topic category and the prediction score determination rules; a higher prediction score indicates a higher response rate. This method comprehensively considers the topic types and content included in the topic library, as well as the importance of each topic category, to predict topic content, making the obtained target topic content more aligned with the target user's interests, thereby effectively attracting the target user and improving the user response rate.

[0065] The following is a brief introduction to the application scenarios to which the technical solutions of the embodiments of this application are applicable. It should be noted that the application scenarios described below are only for illustrating the embodiments of this application and are not intended to limit the scope. In specific implementation, the technical solutions provided by the embodiments of this application can be flexibly applied according to actual needs.

[0066] See Figure 1 This is a schematic diagram illustrating an application scenario of the intelligent recommendation method for topics provided in this application embodiment. The application scenario includes multiple terminal devices 101 (including terminal device 101-1, terminal device 101-2, ..., terminal device 101-n) and a server 102. The terminal devices 101 and the server 102 are connected via a wireless or wired network. The terminal devices 101 include, but are not limited to, desktop computers, mobile phones, portable computers, tablet computers, and other electronic devices. The server 102 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0067] The service provider triggers a user filtering instruction to the server 102 via terminal device 101. This instruction includes service identification information, which is the identifier of the service provider offering services to the target user. The server 102 responds to the instruction by filtering at least one target user associated with the service identification information from the user database. Then, based on the target user's information and topic filtering criteria, it obtains the topic categories used for communication with the target user. Furthermore, based on each topic category, the topic library, and preset importance levels, it obtains a predicted score for each topic content within each category. The predicted scores for each topic content are then sorted to obtain the target topic content for communication with the target user. The user information includes some or all of the target user's virtual resource information, various reminder messages, and behavioral information. The topic library includes the topic content within each topic category and the rules for determining the predicted score; a higher predicted score indicates a higher response rate. The server 102 sends this target topic content to the terminal device 101, enabling the terminal device 101 to communicate with the target user based on the target topic content.

[0068] In some feasible embodiments, the server 102 may also send the topic category to which the target topic content belongs, as well as the user information corresponding to the target topic content, to the terminal device 101, so as to prompt the service provider to provide the recommendation reasons for the target topic content and the content that the target user should pay attention to recently.

[0069] Of course, the methods provided in the embodiments of this application are not limited to... Figure 1 The application scenarios shown can also be used in other possible scenarios, and this application does not impose any limitations. Figure 1 The functions that each device in the application scenario shown can achieve will be described in subsequent method embodiments, and will not be elaborated on here.

[0070] To further illustrate the technical solutions provided by the embodiments of this application, the technical solutions provided by the embodiments of this application will be described in detail below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for illustration and explanation of this application and are not intended to limit this application. Furthermore, the embodiments of this application and the features in the embodiments can be combined with each other without conflict.

[0071] refer to Figure 2 This application provides an intelligent topic recommendation method, which includes the following steps:

[0072] S201. Based on the target user's user information and topic filtering conditions, obtain the various topic categories used for communication with the target user. The user information includes some or all of the target user's virtual resource information, various reminder information, and behavioral information.

[0073] S202. Based on each topic category, the topic library, and the preset importance of each topic category, obtain the prediction score corresponding to each topic content contained in each topic category. The topic library includes each topic content contained in each topic category and the prediction score determination rules. The higher the prediction score, the higher the response rate of the topic content.

[0074] S203. Sort the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target users.

[0075] In this embodiment of the application, before executing step S201, a topic library is constructed, which includes multiple topic categories. Taking communication for the purpose of maintaining relationships as an example, the topic library includes multiple topic categories, and each topic category contains multiple topic contents.

[0076] In practice, the above-mentioned topics can be categorized into three main types: care-related topics, resource incentive topics, and follow-up topics. Each main category contains multiple subcategories. For example, care-related topics include topics for specific dates (greetings), recent reminders, virtual resource reminders / redeems, and most recent notifications; resource incentive topics include topics for XX resources (incentives) and YY resources (incentives); and follow-up topics include topics for resource allocation communication, XX resource materials, and user follow-up. Each subcategory also contains multiple topic contents. For example, topics for specific dates (greetings) include topic content 1 – “Happy Holidays,” topic content 2 – “Happy XX Holidays,” and topic content 3 – “Wishing you good health and a happy holiday on this special day.”

[0077] In this embodiment of the application, after introducing the topic library and before executing step S201, the target user can be determined specifically in the following way:

[0078] Method 1: After determining that the preset conditions are met, at least one target user is selected from the user database based on the preset conditions. The preset conditions include some or all of the periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions.

[0079] Method 2: Respond to the user filtering instruction and, based on the service identification information contained in the user filtering instruction, filter out at least one target user from the user database that is associated with the service identification information, where the service identification information is the identification information of the service provider that provides services to the target user.

[0080] As can be seen from the above methods one and two, target users can be determined through the two dimensions mentioned above. In specific implementation, corresponding to method one, the server can automatically filter users in the user database to determine the target user; corresponding to method two, when the service provider wants to communicate with its users in a targeted manner, the terminal device can trigger a user filtering command to the server, thereby enabling the server to filter multiple users associated with the service provider in the user database to determine the target user.

[0081] In step S201, in this embodiment of the application, after determining the target user, step S201 is executed. Based on the topic filtering conditions, the user information of the target user is queried to see if it involves topic categories that meet the topic filtering conditions, thereby obtaining various topic categories that can be used to communicate with the target user. The aforementioned user information includes some or all of the target user's virtual resource information, various reminder messages, and behavioral information. For example, whether the various reminder messages of the target user involve care-related topics (such as due date reminders), and whether the virtual resource information of the target user involves follow-up topics (such as resource allocation plan communication topics), etc.

[0082] In this embodiment of the application, the topic filtering conditions may include, but are not limited to, the following conditions:

[0083] Condition 1: The expiration time of any virtual resource information of the target user is less than the first threshold from the current time;

[0084] Condition 2: The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold;

[0085] Condition 3: The reminder time of any reminder message for the target user is less than the second threshold from the current time.

[0086] For example, regarding condition one, assuming the target user's information meets condition one and the first threshold is 3 days, then, in step S201, based on the target user's information and topic filtering conditions, it is determined that the target user's virtual resource information A will expire in three days. This yields a topic category for communication with the target user: a recent reminder topic. In other words, when communicating with the target user, the conversation can begin with the expiration of virtual resource information A in three days, and then gradually progress to more in-depth topics.

[0087] Regarding step S202, in this embodiment of the application, when performing step S202, refer to... Figure 3 As shown, the predicted score for each topic content within each topic category can be obtained by performing the following steps, where each topic category is used for communication with the target user:

[0088] Step S2021: Based on each topic category, determine the topic content contained in each topic category from the topic categories included in the topic library;

[0089] Step S2022: Obtain the number of times each topic's content has been cited and the number of times it has received user responses;

[0090] Step S2023: Based on the prediction score determination rules, the preset importance of each topic category, and the number of times each topic content is cited and the number of user responses, the prediction score corresponding to each topic content included in each topic category is obtained.

[0091] Regarding S2022, in this embodiment of the application, the server can pre-deploy a small plugin for collecting user behavior information and store the statistical results in a preset database, such as a user database. Then, when executing step S2022, the server can obtain the number of citations and user responses corresponding to each topic content from the preset database. The number of citations and user responses includes the number of citations and user responses corresponding to the topic categories to which the target user's topic content belongs, the number of citations and user responses corresponding to the topic content of all users in the user database, and part or all of the number of citations and user responses corresponding to the topic categories to which the topic content of all users in the user database belongs.

[0092] In some feasible embodiments, when executing step S2022, the server can specifically perform the following operations to obtain the number of times each topic content is cited and the number of times users respond:

[0093] Operation 1: Based on user information, obtain the first number of citations and the number of user responses corresponding to the topic category to which each topic content belongs; and / or.

[0094] Operation two: Based on the user information of all users contained in the user database, obtain the second citation count and user response count for each topic content; and / or,

[0095] Operation 3: Based on the user information of all users contained in the user database, obtain the number of times the third application and the number of user responses corresponding to the topic category to which each topic content belongs.

[0096] Regarding step S2023, in this embodiment of the application, the server can pre-respond to the configuration operation for the importance of each topic category, so that when executing step S2023, intelligent topic recommendation can be performed based on the preset importance of each topic category. In this way, it can better meet the expected purpose of the service provider and better recommend topics for the service provider.

[0097] In some embodiments, step S2023 is still involved, and the above-mentioned prediction score determination rule can be expressed by the following formula:

[0098]

[0099] Where y1 is the predicted score of any topic content; td is the time indicator, which is the absolute value of the difference between the occurrence time of the topic indicator corresponding to the current topic content and the current time; i is the importance indicator of each topic category, which can be between 0 and 1; r is the number of user responses to the topic content of all users in the user database; c is the number of times the topic content of all users in the user database is cited; tr is the number of user responses to the topic category to which the topic content belongs of all users in the user database; tc is the number of times the topic category to which the topic content belongs of all users in the user database is cited; utr is the number of user responses to the topic category to which the target user's topic content belongs; ut is the number of times the topic category to which the target user's topic content belongs is cited; {a1, a2, a3, a4, a5, a6, b} is the topic indicator weight array, representing the weight value corresponding to each of the above topic indicators.

[0100] In this embodiment of the application, the above {a1, a2, a3, a4, a5, a6, b} can be specifically limited based on the actual situation.

[0101] In some feasible embodiments, a model can be used to determine the specific value (i.e., the optimal value) of each parameter in {a1, a2, a3, a4, a5, a6, b}. Specifically, a communication sample set is collected, which includes multiple communication samples. If a communication sample has a response, it is considered a positive sample; if it has no response, it is considered a negative sample. Then, the response rate is used as the model optimization objective, and a loss function is constructed for a1, a2, a3, a4, a5, a6, b, such as the maximum likelihood function or the cross-entropy loss function. In this embodiment, the above loss function can be represented by the following formula:

[0102]

[0103] Where y is the true score of any communication sample in the communication sample set, i.e., 1 for positive samples and 0 for negative samples; td, i, r, tr, tc, utr, ut are similar to the prediction score determination rules mentioned above, and will not be repeated here; {a1, a2, a3, a4, a5, a6, b} is the topic index weight array to be determined as the optimal value, and the initial value of each parameter in {a1, a2, a3, a4, a5, a6, b} is a preset value.

[0104] After accurately determining the communication sample set and the aforementioned loss function, the loss function is iterated multiple times based on the communication sample set and the aforementioned prediction score determination rule until the total loss value is minimized. The specific value of each parameter in {a1, a2, a3, a4, a5, a6, b} when the total loss value is minimized is taken as the aforementioned optimal value, that is, as {a1, a2, a3, a4, a5, a6, b} in the aforementioned prediction score determination rule in practical applications. In specific implementation, during one round of iterative calculation, for each communication sample included in the communication sample set, the prediction score is determined using the aforementioned prediction score determination rule. Then, the loss value of each communication sample is obtained using the aforementioned loss function. The loss values ​​of each communication sample included in the communication sample set are summed and averaged to obtain the total loss value corresponding to the communication sample set. Based on the total loss value, the specific values ​​of each parameter in {a1, a2, a3, a4, a5, a6, b} are updated. The updated {a1, a2, a3, a4, a5, a6, b} are substituted into the prediction score determination rule used in the next round of iterative calculation.

[0105] Involving step S203, see the embodiments of this application. Figure 4 As shown, the target topic content for communication with the target user is obtained by performing the following steps:

[0106] Step S2031: Sort the predicted scores of the above topics in descending order of predicted scores and according to the preset sorting optimization rules. The preset sorting optimization rules include the topic category priority rule and / or the topic content priority configuration rule.

[0107] Step S2032: Select the topic content that ranks first in the sorting information as the target topic content for communication with the target user.

[0108] In step S2031, the service provider can pre-set topic category priority rules and / or topic content priority configuration rules. In this way, when the server executes the intelligent topic recommendation method provided in this application embodiment, it can combine the service provider's daily work habits to make targeted topic recommendations, thereby better matching the actual application scenario and improving the user response rate.

[0109] In step S2032, after obtaining the target topic content for communication with the target user, the topic category to which the target topic content belongs, as well as the user information corresponding to the target topic content, can be output; so as to prompt the service provider server to select the recommendation reason for the target topic content, as well as the content that the target user needs to pay attention to recently.

[0110] In some feasible embodiments, when performing step S2022, the number of times each topic content is applied and the number of user responses can be obtained for different communication channels. Thus, when performing steps S2031 to S2032, the different communication channels can be sorted separately to obtain the target topic content for different communication channels. This can maximize the accuracy of topic recommendations and further improve the user response rate.

[0111] Based on the same inventive concept, embodiments of this application provide a topic-based intelligent recommendation device, please refer to... Figure 5 The device includes:

[0112] The topic category determination module 501 is used to obtain various topic categories for communication with the target user based on the target user's user information and topic filtering conditions, wherein the user information includes part or all of the target user's virtual resource information, various reminder information and behavior information;

[0113] The score prediction module 502 is used to obtain a predicted score for each topic content included in each topic category based on the topic categories, the topic library, and the preset importance of each topic category. The topic library includes each topic content included in each topic category and the prediction score determination rules. The higher the prediction score, the higher the response rate of the topic content.

[0114] The sorting module 503 is used to sort the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target user.

[0115] In one possible embodiment, before obtaining the topic categories used for communication with the target user based on the target user's user information and topic filtering conditions, the topic category determination module 501 is further configured to:

[0116] After determining that preset conditions are met, at least one target user is selected from the user database based on the preset conditions, wherein the preset conditions include some or all of the following: periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions; or,

[0117] In response to a user filtering instruction, based on the service identification information contained in the user filtering instruction, at least one target user associated with the service identification information is filtered from the user database, wherein the service identification information is the identification information of the service provider that provides services to the target user.

[0118] In one possible embodiment, the topic filtering criteria include some or all of the following criteria:

[0119] The expiration time of any virtual resource information of the target user is less than the first threshold from the current time;

[0120] The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold;

[0121] The reminder time of any reminder message for the target user is less than the second threshold from the current time.

[0122] In one possible embodiment, the score prediction module 502 is specifically used for:

[0123] Based on the aforementioned topic categories, determine the topic content contained in each topic category from the topic categories included in the topic library;

[0124] Obtain the number of times each topic's content has been cited and the number of times it has received user responses;

[0125] Based on the predicted score determination rules, the preset importance of each topic category, and the number of times each topic content is cited and the number of user responses, the predicted score for each topic content included in each topic category is obtained.

[0126] In one possible embodiment, the sorting module 503 is specifically used for:

[0127] The predicted scores of each topic are sorted in descending order of predicted scores and according to preset sorting optimization rules, wherein the preset sorting optimization rules include topic category priority rules and / or topic content priority configuration rules.

[0128] The topic that appears first in the ranking information will be used as the target topic for communication with the target user.

[0129] In one possible embodiment, after obtaining the target topic content for communication with the target user, the sorting module 503 is further configured to:

[0130] Output the topic category to which the target topic content belongs, and the user information corresponding to the target topic content.

[0131] Based on the same inventive concept, embodiments of this application provide an electronic device that can realize the function of an intelligent recommendation device for the topics discussed above. Please refer to... Figure 6 The device includes one or more processors 601 and memory 602.

[0132] Processor 601 may include one or more central processing units (CPUs) or digital processing units, etc. Processor 601 is used to implement the intelligent recommendation method for the above-mentioned topics when invoking a computer program stored in memory 602.

[0133] The memory 602 is used to store computer programs executed by the processor 601. The memory 602 may mainly include a program storage area and a data storage area. The program storage area may store the operating system and programs required to run instant messaging functions, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.

[0134] Memory 602 may be volatile memory, such as random-access memory (RAM); memory 602 may also be non-volatile memory, such as read-only memory, flash memory, hard disk drive (HDD), or solid-state drive (SSD); or memory 602 may be any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 602 may be a combination of the above-described memories.

[0135] This application embodiment does not limit the specific connection medium between the processor 601 and the memory 602. This application embodiment... Figure 6 The processor 601 and memory 602 are connected via a bus 603, and the bus 603 is in Figure 6 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. The 603 bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0136] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium. The computer program product includes computer program code, which, when executed on a computer, causes the computer to perform an intelligent recommendation method for any of the topics discussed above. Since the principle by which the above-described computer-readable storage medium solves the problem is similar to that of the intelligent recommendation method for topics, the implementation of the above-described computer-readable storage medium can be found in the implementation of the method, and repeated details will not be elaborated further.

[0137] Based on the same inventive concept, this application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to execute an intelligent recommendation method for any of the topics discussed above. Since the principle by which the above computer program product solves the problem is similar to the intelligent recommendation method for topics, the implementation of the above computer program product can refer to the implementation of the method, and repeated details will not be elaborated further. The program product can be used in any combination of one or more readable media.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of user-operated steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the function specified in one or more processes in the flowchart and / or one or more blocks in the block diagram.

[0142] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for intelligent topic recommendation, characterized in that, include: Based on the target user's user information and topic filtering conditions, various topic categories that can be used to communicate with the target user are obtained, wherein the user information includes some or all of the target user's virtual resource information, various reminder information and behavioral information; Based on the topic categories and the multiple topic categories included in the topic library, determine the topic content included in each topic category, where any topic category includes multiple topic contents. Based on the preset importance of each topic category, the topic indicator weight array corresponding to each topic content, and the number of citations and user responses corresponding to each topic content, the user response rate of each topic content is predicted and scored to obtain a predicted score for each topic content. The number of citations and user responses for any topic content includes the number of citations and user responses for the topic category to which the target user's topic content belongs, the number of citations and user responses for the topic content to which all users in the user database belong, and the number of citations and user responses for the topic category to which the topic content to which all users in the user database belong. The predicted scores corresponding to each topic are sorted to obtain the target topic content for communication with the target user.

2. The method as described in claim 1, characterized in that, Before obtaining the topic categories that can be used to communicate with the target user based on the target user's user information and topic filtering conditions, the method further includes: After determining that preset conditions are met, at least one target user is selected from the user database based on the preset conditions, wherein the preset conditions include some or all of the following: periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions; or, In response to a user filtering instruction, based on the service identification information contained in the user filtering instruction, at least one target user associated with the service identification information is filtered from the user database, wherein the service identification information is the identification information of the service provider that provides services to the target user.

3. The method as described in claim 1, characterized in that, The topic filtering criteria include some or all of the following criteria: The expiration time of any virtual resource information of the target user is less than the first threshold from the current time; The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold; The reminder time of any reminder message for the target user is less than the second threshold from the current time.

4. The method according to any one of claims 1-3, characterized in that, The step of sorting the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target user includes: The predicted scores of each topic are sorted in descending order of predicted scores and according to preset sorting optimization rules, wherein the preset sorting optimization rules include topic category priority rules and / or topic content priority configuration rules. The topic that appears first in the ranking information will be used as the target topic for communication with the target user.

5. The method as described in claim 4, characterized in that, After obtaining the target topic content for communication with the target user, the process further includes: Output the topic category to which the target topic content belongs, and the user information corresponding to the target topic content.

6. A topic-based intelligent recommendation device, characterized in that, include: The topic category determination module is used to obtain various topic categories that can be used to communicate with the target user based on the target user's user information and topic filtering conditions, wherein the user information includes some or all of the target user's virtual resource information, various reminder information and behavioral information; The score prediction module is used to determine the topic content included in each topic category based on the topic categories and multiple topic categories included in the topic library. Each topic category includes multiple topic contents. Based on the preset importance of each topic category, the topic indicator weight array corresponding to each topic content, and the number of citations and user responses corresponding to each topic content, the module predicts and scores the user response rate of each topic content to obtain a predicted score for each topic content. The number of citations and user responses of any topic content includes the number of citations and user responses corresponding to the topic category to which the topic content belongs for the target user, the number of citations and user responses corresponding to the topic content of all users in the user database, and the number of citations and user responses corresponding to the topic category to which the topic content belongs for all users in the user database. The sorting module is used to sort the predicted scores corresponding to each topic content to obtain the target topic content for communication with the target user.

7. The apparatus as claimed in claim 6, characterized in that, Before obtaining the available topic categories for communication with the target user based on the target user's user information and topic filtering conditions, the topic category determination module is further configured to: After determining that preset conditions are met, at least one target user is selected from the user database based on the preset conditions, wherein the preset conditions include some or all of the following: periodic filtering conditions, expiration reminder conditions, and virtual resource change conditions; or, In response to a user filtering instruction, based on the service identification information contained in the user filtering instruction, at least one target user associated with the service identification information is filtered from the user database, wherein the service identification information is the identification information of the service provider that provides services to the target user.

8. The apparatus as claimed in claim 6, characterized in that, The topic filtering criteria include some or all of the following criteria: The expiration time of any virtual resource information of the target user is less than the first threshold from the current time; The virtual resources corresponding to the target user's virtual resource information are not less than the resource threshold; The reminder time of any reminder message for the target user is less than the second threshold from the current time.

9. The apparatus according to any one of claims 6-8, characterized in that, The sorting module is specifically used for: The predicted scores of each topic are sorted in descending order of predicted scores and according to preset sorting optimization rules, wherein the preset sorting optimization rules include topic category priority rules and / or topic content priority configuration rules. The topic that appears first in the ranking information will be used as the target topic for communication with the target user.

10. The apparatus as claimed in claim 9, characterized in that, After obtaining the target topic content for communication with the target user, the sorting module is further configured to: Output the topic category to which the target topic content belongs, and the user information corresponding to the target topic content.

11. An electronic device, characterized in that, include: Memory, used to store program instructions; A processor is configured to invoke program instructions stored in the memory and execute the steps of the method according to any one of claims 1-5.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a computer, cause the computer to perform the method as described in any one of claims 1-5.

13. A computer program product, characterized in that, The computer program product includes: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of claims 1-5.

Images