Recommendation method and recommendation device for activity scheme

Abstract

The invention provides an activity scheme recommendation method and device, and the method comprises the steps: providing at least one service for a user through an application program by a service provider, and generating a demand feature vector used for representing a service demand corresponding to a target user according to first service information corresponding to at least one first service related dimension; generating a scheme feature vector for each candidate activity scheme according to second service information corresponding to the at least one second service related dimension; and matching the demand feature vector corresponding to the target user with the scheme feature vector of each candidate activity scheme, and screening out a target activity scheme from the plurality of candidate activity schemes. In this way, the activity scheme meeting the personalized service requirements of the target user can be screened out; indirect, through the target activity scheme, the activity participation enthusiasm of the user can be improved, idle service resources are reduced, and the utilization rate of the service resources is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method and device for recommending an activity plan. Background Art

[0002] Existing activity plans are typically manually preconfigured and generated. For example, a user's check-in behavior data is recorded and a preset incentive is issued if a preset threshold is reached. However, this approach lacks flexibility and cannot recommend personalized activity plans for users. Summary of the Invention

[0003] In view of this, the purpose of the present application is to provide a method and device for recommending an activity plan, so as to solve the problem in the prior art that it is impossible to match and recommend personalized activity plans for users.

[0004] The present embodiment provides a method for recommending an activity plan, wherein a service provider provides at least one service to a user through an application program; the recommendation method includes:

[0005] generating, based on the first service information corresponding to the at least one first service-related dimension, a demand feature vector for characterizing the service demand corresponding to the target user;

[0006] generating a solution feature vector for each candidate activity solution based on the service information corresponding to the at least one second service-related dimension;

[0007] The demand feature vector corresponding to the target user is matched with the solution feature vector of each candidate activity solution, and a target activity solution is selected from the multiple candidate activity solutions.

[0008] Furthermore, the first service-related dimension includes a user dimension, and the second service-related dimension includes a service provider dimension; and the recommendation method further includes:

[0009] Obtaining historical operation data of the target user for the application; determining first service information corresponding to the user dimension based on data related to service usage in the historical operation data; wherein the data related to service usage is used to characterize the target user's willingness to use the service;

[0010] According to the resource usage of each service, the service guidance demand of the service provider is determined; the service guidance demand is determined as the second service information corresponding to the service provider dimension; the service guidance demand refers to the demand for guiding users to use specific services.

[0011] Furthermore, the first service-related dimension and / or the second service-related dimension further includes: a timeliness dimension and / or a prediction dimension; and the recommendation method further includes:

[0012] For the timeliness dimension, determining the current time information and / or the hot spot information at the current time as the first service information and / or the second service information corresponding to the timeliness dimension;

[0013] For the prediction dimension, predict at least one of the predicted resource usage of the service provider for each service, the predicted service usage of the target user, and the predicted hotspot information within a predetermined time period in the future; determine the first service information and / or second service information corresponding to the prediction dimension based on at least one of the predicted resource usage, the predicted service usage, and the predicted hotspot information.

[0014] Furthermore, generating a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension includes:

[0015] Determining task content and incentive content based on second service information corresponding to at least one second service-related dimension;

[0016] Determine candidate activity plans based on task content and incentive content;

[0017] Based on the determined candidate activity plans, a plan feature vector is generated for each candidate activity plan.

[0018] Furthermore, generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes:

[0019] For each task and each incentive that can be provided by the service provider, generate a task feature vector for each task and an incentive feature vector for each incentive according to the second service information corresponding to at least one second service-related dimension;

[0020] The task feature vector of each task and the incentive feature vector of each incentive are combined to obtain a scheme feature vector of each candidate activity scheme; wherein each candidate activity scheme includes at least one task and at least one incentive.

[0021] Furthermore, generating a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension includes:

[0022] Combining each task and each incentive that can be provided by the service provider to generate each candidate activity plan; wherein each candidate activity plan includes at least one task and at least one incentive;

[0023] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to at least one second service-related dimension.

[0024] Furthermore, generating a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension includes:

[0025] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to the at least one second service-related dimension and the multi-dimensional calculation weight.

[0026] Furthermore, generating a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension includes:

[0027] When any second service information and / or candidate activity plan meets a predetermined condition, a dynamic weight adjustment factor corresponding to the second service information and / or candidate activity plan is generated;

[0028] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to each second service-related dimension, the multi-dimensional calculation weight, and the dynamic weight adjustment factor corresponding to the second service information and / or the candidate activity solution.

[0029] Furthermore, the demand feature vector corresponding to the target user is matched with the solution feature vector of each candidate activity solution, and the target activity solution is screened out from the multiple candidate activity solutions, including:

[0030] Determining respectively the vector matching degree between the demand feature vector and the solution feature vector of each candidate activity solution;

[0031] The target activity plan that best matches the service demand corresponding to the target user is selected from the multiple candidate activity plans according to the vector matching degree.

[0032] Furthermore, before matching the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, the recommendation method further includes:

[0033] Unify the dimensions of the demand feature vector and the solution feature vector of each candidate activity solution,

[0034] The present application also provides an activity plan recommendation device, wherein a service provider provides at least one service to a user through an application program; the recommendation device includes:

[0035] A first generating module, configured to generate a demand feature vector for representing a service demand corresponding to a target user based on first service information corresponding to at least one first service-related dimension;

[0036] A second generating module, configured to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension;

[0037] The matching module is used to match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, and select the target activity solution from the multiple candidate activity solutions.

[0038] An embodiment of the present application also provides an electronic device, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for recommending an activity plan as described above are performed.

[0039] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for recommending an activity plan as described above are executed.

[0040] The embodiments of the present application provide a method and device for recommending an activity plan, which generates a demand feature vector representing the service demand corresponding to a target user from at least one first service-related dimension, and generates a plan feature vector for each candidate activity plan from at least one second service-related dimension; by matching the two, a target activity plan is screened out; thereby, an activity plan that meets the personalized service demand of the target user can be screened out; indirectly, the target activity plan can improve the user's enthusiasm for activity participation, reduce idle service resources, and improve the utilization rate of service resources.

[0041] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0043] Figure 1 A flowchart showing a method for recommending an activity plan provided in an embodiment of the present application is shown;

[0044] Figure 2 A schematic diagram showing the structure of a device for recommending an activity plan provided in an embodiment of the present application is shown;

[0045] Figure 3 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work falls within the scope of protection of the present application.

[0047] Research has found that various activity plans in existing technologies are generally pre-configured manually. For example, they record user check-in data and then issue a preset incentive if a preset threshold is reached. However, this approach lacks flexibility and cannot recommend personalized activity plans for users.

[0048] Based on this, the embodiment of the present application provides a method for recommending activity plans to screen out activity plans that meet the personalized service needs of target users, increase users' enthusiasm for participating in activities, reduce idle service resources, and improve the utilization rate of service resources.

[0049] In an embodiment of the present application, a service provider provides services to users through an application. The service provider can provide at least one service based on its own functional design and can provide corresponding services in response to user requests when in use. Among them, applications include local applications that are installed and run locally, cloud applications that can be accessed through a browser and deployed on a cloud platform, and mini-programs (light applications) that rely on other APPs or operating systems to run. For example, an application is a conversational intelligent entity that can provide users with services such as intelligent conversation and content generation.

[0050] See also Figure 1 , Figure 1 This is a flow chart of a method for recommending an activity plan provided in an embodiment of the present application. Figure 1 As shown in , the recommended method provided in the embodiment of the present application includes:

[0051] S101. Generate a demand feature vector for characterizing a service demand corresponding to a target user based on first service information corresponding to at least one first service-related dimension.

[0052] S102: Generate a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension.

[0053] Here, service-related dimensions refer to the dimensions involved in the entire process of service provision.

[0054] In a first possible implementation, the first service-related dimension includes the user dimension, and the first service information includes information related to the target user's service needs, such as the target user's login data, chat duration, created activities, and hot topics. The second service-related dimension includes the service provider dimension. The service provider focuses on overall data, and the second service information includes information related to the process of providing services to all users, such as the number of users logged in, chat service duration, created activities, hot topic rankings, resource constraints, and service guidance requirements.

[0055] In specific implementation, the first service information and the second service information can be obtained through the following steps:

[0056] For the first service information corresponding to the user dimension, step a1: obtain historical operation data of the target user on the application.

[0057] It should be noted that the embodiment of the present application obtains the historical operation data of the target user on the application program with the authorization of the user and the relevant organization in advance, wherein the operation data includes the data corresponding to the operations triggered by the target user when using the application program.

[0058] Step a2: Determine the first service information corresponding to the user dimension based on the data related to service usage in the historical operation data.

[0059] Data related to service usage may include data on user usage of various services (such as the frequency of user login services), the duration of user service usage, and the type of service used (chat services, create services). Data related to service usage can reflect user historical usage habits and be used to characterize the target user's historical service usage intentions. This data can be identified as primary service information. For example, if login services are rare and chat services are frequent, this indicates that the target user prefers limited, in-depth discussions rather than extensive, frequent social interactions.

[0060] For the second service information corresponding to the service provider dimension, step b1: determining the service guidance requirements of the service provider according to the resource usage of each service.

[0061] It should be noted that when an application provides multiple complex services, the computing power of a single computing device may be insufficient. Therefore, the services in the embodiments of the present application can be implemented based on a distributed cluster architecture, with different services deployed on different computing devices in the cluster. Based on the status of at least one computing device corresponding to each service, the service provider's resource usage for each service can be determined, thereby determining the service provider's service guidance requirements.

[0062] Step b2: Determine the service guidance requirement as the second service information corresponding to the service provider dimension.

[0063] Service guidance refers to the need to guide users to use specific services. For example, when a service resource is available, you can guide users to increase the provision of that service, such as by increasing demand for login services, chat services, and creation services. This prevents idle service resources and improves service resource utilization. For another example, if a service's resources are already heavily utilized, it indicates that users are highly motivated to participate in that service. To increase user engagement, if resources for that service remain available, you can continue to guide users to increase the provision of that service.

[0064] In this way, the first service information and the second service information respectively reflect the corresponding service needs of the target user and the service provider from two directions, so that the target activity plan determined by subsequent matching meets both the user's personal usage needs and the service provision needs of the service provider.

[0065] In a second possible implementation, the first service-related dimension and / or the second service-related dimension further includes: a timeliness dimension and / or a prediction dimension.

[0066] With respect to the timeliness dimension, the current time information and / or the hot spot information at the current time is determined as the first service information and / or the second service information corresponding to the timeliness dimension.

[0067] For example, the current time information may include whether the date is a weekday or a holiday, whether the time is daytime or nighttime, etc. The current time information is related to user needs. Generally speaking, users tend to have higher service usage needs during holidays or at night.

[0068] Hot information includes hot words in various social media, hot search rankings, as well as participation, number of followers, number of clicks, etc. Hot information is also related to user needs. Generally speaking, the emergence of hot information often promotes users to use related services. For the first service information, the hot information at the current time refers to the hot information that the target user is concerned about. The hot information that the target user is concerned about can be filtered out from the public hot information at the current time by pre-setting keywords and field preferences by the target user. The application can also learn and filter out the hot information that the target user is concerned about based on the target user's historical attention to public hot information. For the second service information, the hot information at the current time refers to the public hot information at the current time that the public is concerned about.

[0069] For the prediction dimension, predict at least one of the predicted resource usage of the service provider for each service, the predicted service usage of the target user, and the predicted hotspot information within a predetermined time period in the future; determine the first service information and / or second service information corresponding to the prediction dimension based on at least one of the predicted resource usage, the predicted service usage, and the predicted hotspot information.

[0070] In this step, the pre-trained model can predict at least one of the above-mentioned prediction information within a predetermined time period in the future based on historical information. Afterwards, the service information corresponding to the prediction dimension is determined based on the at least one of the above-mentioned prediction information obtained by prediction. Among them, the first service information corresponding to the prediction dimension includes prediction information related to the target user, such as the service usage operations that the target user may perform within the prediction time period, and the hot information that the target user may be concerned about; the second service information corresponding to the prediction dimension includes prediction information related to the overall data, such as the prediction of the service usage operations (resource usage) that all users may perform within the prediction time period, and the public hot information that all users may be concerned about.

[0071] The introduction of timeliness and prediction dimensions can improve the timeliness of the demand feature vector, thereby making the subsequent target activity plan more timely and effectively enhancing user participation in activities within the current and future timeframes. Furthermore, it can facilitate timely evaluation of the effectiveness of activity plans and enable timely adjustments if ineffective implementation is detected.

[0072] After obtaining the first service information, in step S101, a demand feature vector representing the service demand corresponding to the target user can be generated based on the first service information corresponding to each first service-related dimension and the calculated weights. The calculated weights can be determined by referring to relevant methods in the prior art, such as model optimization training, and this application does not impose any restrictions on this.

[0073] After obtaining the second service information, each activity plan in step S102 includes at least one task and at least one incentive. A task is a task that the user is required to perform, such as using a service of an application. An incentive is a reward that the user receives after completing the task. Step S102 may include:

[0074] Method 1: Determine the task content and incentive content based on the second service information corresponding to at least one second service-related dimension; determine candidate activity plans based on the task content and incentive content; and generate a plan feature vector for each candidate activity plan based on the determined candidate activity plans. Each candidate activity plan includes at least one task and at least one incentive.

[0075] In this way, task content and incentive content can be generated specifically based on the second service information. For example, new task content and incentive content can be generated based on public hot information, such as initiating voting tasks and scenario simulation tasks, or generating new incentives based on best-selling lists. Based on the task content and incentive content, candidate activity plans are combined and generated, and the plan feature vector for each candidate activity plan is generated by combining the calculated weights. This makes the generated candidate activity plans more flexible and diverse, and closer to the actual activity situation.

[0076] Method 2: For each task and each incentive that can be provided by the service provider, generate a task feature vector for each task and an incentive feature vector for each incentive based on the second service information corresponding to at least one second service-related dimension; combine the task feature vector for each task and the incentive feature vector for each incentive to obtain a solution feature vector for each candidate activity solution; wherein, each candidate activity solution includes at least one task and at least one incentive.

[0077] In this way, the task content and the incentive content are preset contents that can be provided by the service provider. Here, in an example, the application provides ten services such as login service, chat service, and creation service, which can be set as ten tasks accordingly; and ten incentives such as cash, tokens, points, and number of uses are provided. Then, when activities and incentives are combined one by one, 100 activity plans can be generated. When multiple tasks and multiple incentive combinations are further considered (such as one incentive for multiple tasks, or multiple incentives for one task), more activity plans can be generated. In this way, the embodiment of the present application can generate a large number of candidate activity plans suitable for different users in a variety of scenarios. Indirectly, the best activity plan can be found from the massive candidate activity plans for recommendation.

[0078] Multi-dimensional calculation weights are set for each task and / or each incentive. Based on the second service information, a task feature vector for each task and an incentive feature vector for each incentive are generated. Each task feature vector and each incentive feature vector are then combined to obtain a solution feature vector for each candidate activity solution. The solution feature vector is a vector feature related to the task and incentive, such as the combination of task 1 and incentive 1 (1, 1), the combination of task 1 and incentive 2 (1, 2), ..., the combination of task 5 and incentive 3 (5, 3), and so on.

[0079] Method three: Combine each task and each incentive that can be provided by the service provider to generate each candidate activity plan; wherein each candidate activity plan includes at least one task and at least one incentive; and generate a plan feature vector for each candidate activity plan based on the second service information corresponding to the at least one second service-related dimension.

[0080] In this way, the task content and incentive content are preset contents that can be provided by the service provider, and candidate activity plans can be generated by combining them; multi-dimensional calculation weights are set for each task and / or each incentive that can be provided by the service provider respectively, and according to the combination of each task and each incentive, the corresponding multi-dimensional calculation weights are spliced, added, multiplied, or exponentiation operation, etc., and combined with the second service information corresponding to at least one second service-related dimension to generate a plan feature vector.

[0081] In specific implementation, when generating a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension in the above three methods of step S102, a solution feature vector for each candidate activity solution can be generated based on the second service information and multi-dimensional calculation weights. For example, a pre-trained model can be used to implement it, and the second service information and candidate activity solution are input into the model, and the parameters in the model include multi-dimensional calculation weights.

[0082] Furthermore, generating a solution feature vector for each candidate activity solution according to the second service information corresponding to at least one second service-related dimension in step S102 further includes:

[0083] When any second service information and / or candidate activity plan meets predetermined conditions, a dynamic weight adjustment factor corresponding to the second service information and / or candidate activity plan is generated. The dynamic weight adjustment factor can increase or decrease the weight of the second service information and / or candidate activity plan in the multi-dimensional weight calculation. Both the predetermined conditions and the dynamic weight adjustment factor can be customized based on different business scenarios.

[0084] For example, if a task has no effect (or has a poor effect) regardless of the incentive used, or if an incentive has no effect (or has a poor effect) regardless of the task information used, then the task information and / or incentive information can be excluded when generating the task information and / or incentive information (the corresponding weight in the candidate activity plan is 0), or the weight of the corresponding dimension can be set when generating the task information and / or incentive information (the corresponding weight in the candidate activity plan is reduced). For another example, if the user is "not interested" in the recommended hot information as a whole, then the weight of the second service information corresponding to the timeliness dimension should be reduced.

[0085] Afterwards, a solution feature vector for each candidate activity solution is generated based on the second service information corresponding to each second service-related dimension, the multi-dimensional calculation weight, and the dynamic weight adjustment factor corresponding to the second service information and / or the candidate activity solution.

[0086] S103 : Match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, and select a target activity solution from the multiple candidate activity solutions.

[0087] In a specific implementation, step S103 may include: determining the vector matching degree between the demand feature vector and the solution feature vector of each candidate activity solution respectively; wherein the vector matching degree is used to measure the similarity between two vectors, which can be represented by distance, angle, etc., such as cosine similarity, Euclidean distance, Manhattan distance, etc.

[0088] Based on the vector matching degree, the target activity plan that best matches the service requirement corresponding to the target user is selected from the multiple candidate activity plans. Generally speaking, the first one or several candidate activity plans with the highest vector matching degree can be used as the target activity plan that best matches the service requirement corresponding to the target user.

[0089] Furthermore, before step S103 , if the dimensions of the demand feature vector and the solution feature vector are different, the recommendation method may further include: unifying the dimensions of the demand feature vector and the solution feature vector of each candidate activity solution.

[0090] In specific implementation, the dimensions of a vector can be unified to the other side through linear transformation, filling, interpolation, projection and other methods, or the demand feature vector and the solution feature vector can be mapped to the same multidimensional vector space through linear transformation respectively, and then the vector matching degree between the transformed demand feature vector and the multiple transformed solution feature vectors can be calculated respectively.

[0091] An embodiment of the present application provides a method for recommending an activity plan, in which a service provider provides at least one service to a user through an application; the recommendation method includes: generating a demand feature vector for characterizing the service demand corresponding to a target user based on first service information corresponding to at least one first service-related dimension; generating a solution feature vector for each candidate activity plan based on second service information corresponding to at least one second service-related dimension; matching the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity plan, respectively, to screen out a target activity plan from multiple candidate activity plans.

[0092] The above method can be executed each time a user logs in to determine a target activity plan, which is then presented to the user after login for selection and interaction. The above method can also be executed periodically to determine a target activity plan, which is then presented to the user after login for selection and interaction. In this way, activity plans that meet the personalized service needs of target users can be screened. Indirectly, the targeted activity plans can increase user participation, reduce idle service resources, and improve service resource utilization.

[0093] See also Figure 2 , Figure 2 A schematic diagram of the structure of a device for recommending an activity plan provided in an embodiment of the present application. A service provider provides at least one service to a user through an application program; Figure 2 As shown in , the recommendation device 200 includes:

[0094] A first generating module 210 is configured to generate a demand feature vector for representing a service demand corresponding to a target user based on first service information corresponding to at least one first service-related dimension;

[0095] A second generating module 220 is configured to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension;

[0096] The matching module 230 is configured to match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, and select a target activity solution from the multiple candidate activity solutions.

[0097] Furthermore, the first service-related dimension includes a user dimension, and the second service-related dimension includes a service provider dimension; the recommendation device further includes an acquisition module; the acquisition module is configured to:

[0098] Obtaining historical operation data of the target user for the application; determining first service information corresponding to the user dimension based on data related to service usage in the historical operation data; wherein the data related to service usage is used to characterize the target user's willingness to use the service;

[0099] According to the resource usage of each service, the service guidance demand of the service provider is determined; the service guidance demand is determined as the second service information corresponding to the service provider dimension; the service guidance demand refers to the demand for guiding users to use specific services.

[0100] Furthermore, the first service-related dimension and / or the second service-related dimension further includes: a timeliness dimension and / or a prediction dimension; and the acquisition module is further configured to:

[0101] For the timeliness dimension, determining the current time information and / or the public hot spot information at the current time as the first service information and / or the second service information corresponding to the timeliness dimension;

[0102] For the prediction dimension, predict at least one of the predicted resource usage of the service provider for each service, the predicted service usage of the target user, and the predicted hotspot information within a predetermined time period in the future; determine the first service information and / or second service information corresponding to the prediction dimension based on at least one of the predicted resource usage, the predicted service usage, and the predicted hotspot information.

[0103] Furthermore, when the second generating module 220 is used to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension, the second generating module 220 is used to:

[0104] Determining task content and incentive content based on second service information corresponding to at least one second service-related dimension;

[0105] Determine candidate activity plans based on task content and incentive content;

[0106] Based on the determined candidate activity plans, a plan feature vector is generated for each candidate activity plan.

[0107] Furthermore, when the second generating module 220 is used to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension, the second generating module 220 is used to:

[0108] For each task and each incentive that can be provided by the service provider, generate a task feature vector for each task and an incentive feature vector for each incentive according to the second service information corresponding to at least one second service-related dimension;

[0109] The task feature vector of each task and the incentive feature vector of each incentive are combined to obtain a scheme feature vector of each candidate activity scheme; wherein each candidate activity scheme includes at least one task and at least one incentive.

[0110] Furthermore, when the second generating module 220 is used to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension, the second generating module 220 is used to:

[0111] Combining each task and each incentive that can be provided by the service provider to generate each candidate activity plan; wherein each candidate activity plan includes at least one task and at least one incentive;

[0112] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to at least one second service-related dimension.

[0113] Furthermore, when the second generating module 220 is used to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to at least one second service-related dimension, the second generating module 220 is used to:

[0114] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to the at least one second service-related dimension and the multi-dimensional calculation weight.

[0115] Furthermore, when the second generating module 220 is used to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension, the second generating module 220 is used to:

[0116] When any second service information and / or candidate activity plan meets a predetermined condition, a dynamic weight adjustment factor corresponding to the second service information and / or candidate activity plan is generated;

[0117] A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to each second service-related dimension, the multi-dimensional calculation weight, and the dynamic weight adjustment factor corresponding to the second service information and / or the candidate activity solution.

[0118] Furthermore, when the matching module 230 is used to match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution and screen out the target activity solution from multiple candidate activity solutions, the matching module 230 is used to:

[0119] Determining respectively the vector matching degree between the demand feature vector and the solution feature vector of each candidate activity solution;

[0120] The target activity plan that best matches the service demand corresponding to the target user is selected from the multiple candidate activity plans according to the vector matching degree.

[0121] Furthermore, when the matching module 230 is used to match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, the matching module 230 is further used to:

[0122] The demand feature vector and the solution feature vector of each candidate activity solution are dimensionally unified.

[0123] See also Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 3 As shown in FIG, the electronic device 300 includes a processor 310 , a memory 320 and a bus 330 .

[0124] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 communicates with the memory 320 via the bus 330. When the machine-readable instructions are executed by the processor 310, the above-mentioned Figure 1 The specific implementation of the steps of the method for recommending an activity plan in the illustrated method embodiment can be found in the method embodiment and will not be described in detail here.

[0125] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The specific implementation of the steps of the method for recommending an activity plan in the illustrated method embodiment can be found in the method embodiment and will not be described in detail here.

[0126] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. There may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interface, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0128] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0129] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0130] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0131] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The scope of protection of the present application is not limited thereto. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-mentioned embodiments within the technical scope disclosed in the present application, or perform equivalent replacements for some of the technical features thereof. These modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for recommending an activity plan, characterized in that: The service provider provides at least one service to the user through the application; the recommendation method includes: generating, based on the first service information corresponding to the at least one first service-related dimension, a demand feature vector for characterizing the service demand corresponding to the target user; generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension; The demand feature vector corresponding to the target user is matched with the solution feature vector of each candidate activity solution, and a target activity solution is selected from the multiple candidate activity solutions.

2. The recommendation method according to claim 1, characterized in that: The first service-related dimension includes a user dimension, and the second service-related dimension includes a service provider dimension; The recommended method further includes: Obtaining historical operation data of the target user for the application; determining first service information corresponding to the user dimension based on data related to service usage in the historical operation data; wherein the data related to service usage is used to characterize the target user's willingness to use the service; According to the resource usage of each service, the service guidance demand of the service provider is determined; the service guidance demand is determined as the second service information corresponding to the service provider dimension; the service guidance demand refers to the demand for guiding users to use specific services.

3. The recommendation method according to claim 1 or 2, characterized in that: The first service-related dimension and / or the second service-related dimension further include: a timeliness dimension and / or a prediction dimension; and the recommendation method further includes: For the timeliness dimension, determining the current time information and / or the hot spot information at the current time as the first service information and / or the second service information corresponding to the timeliness dimension; For the prediction dimension, predict at least one of the predicted resource usage of the service provider for each service, the predicted service usage of the target user, and the predicted hotspot information within a predetermined time period in the future; determine the first service information and / or second service information corresponding to the prediction dimension based on at least one of the predicted resource usage, the predicted service usage, and the predicted hotspot information.

4. The recommendation method according to claim 1, characterized in that: Generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes: Determining task content and incentive content based on second service information corresponding to at least one second service-related dimension; Determine candidate activity plans based on task content and incentive content; Based on the determined candidate activity plans, a plan feature vector is generated for each candidate activity plan.

5. The recommendation method according to claim 4, characterized in that: Generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes: For each task and each incentive that can be provided by the service provider, generate a task feature vector for each task and an incentive feature vector for each incentive according to the second service information corresponding to at least one second service-related dimension; The task feature vector of each task and the incentive feature vector of each incentive are combined to obtain a scheme feature vector of each candidate activity scheme; wherein each candidate activity scheme includes at least one task and at least one incentive.

6. The recommendation method according to claim 4, characterized in that: Generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes: Combining each task and each incentive that can be provided by the service provider to generate each candidate activity plan; wherein each candidate activity plan includes at least one task and at least one incentive; A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to at least one second service-related dimension.

7. The recommendation method according to claim 1, characterized in that: Generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes: A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to the at least one second service-related dimension and the multi-dimensional calculation weight.

8. The recommendation method according to claim 7, characterized in that: Generating a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension includes: When any second service information and / or candidate activity plan meets a predetermined condition, a dynamic weight adjustment factor corresponding to the second service information and / or candidate activity plan is generated; A solution feature vector for each candidate activity solution is generated according to the second service information corresponding to each second service-related dimension, the multi-dimensional calculation weight, and the dynamic weight adjustment factor corresponding to the second service information and / or the candidate activity solution.

9. The recommendation method according to claim 1, characterized in that: Match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, and select the target activity solution from the multiple candidate activity solutions, including: Determining respectively the vector matching degree between the demand feature vector and the solution feature vector of each candidate activity solution; The target activity plan that best matches the service demand corresponding to the target user is selected from the multiple candidate activity plans according to the vector matching degree.

10. The recommendation method according to claim 1, characterized in that: Before matching the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, the recommendation method further includes: The demand feature vector and the solution feature vector of each candidate activity solution are dimensionally unified.

11. A device for recommending an activity plan, characterized in that: The service provider provides at least one service to the user through the application program; the recommendation device includes: A first generating module, configured to generate a demand feature vector for representing a service demand corresponding to a target user based on first service information corresponding to at least one first service-related dimension; A second generating module, configured to generate a solution feature vector for each candidate activity solution based on the second service information corresponding to the at least one second service-related dimension; The matching module is used to match the demand feature vector corresponding to the target user with the solution feature vector of each candidate activity solution, and select the target activity solution from the multiple candidate activity solutions.

Images