Marketing activity generation method and device, electronic equipment and storage medium
By performing semantic analysis and automated processing on the marketing campaign configuration text, the problem of low efficiency in marketing campaign generation is solved, and efficient and automated marketing campaign generation and precise user push are achieved.
Patent Information
- Application Number
- CN202510860120.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-24
- Publication Date
- 2025-10-17
AI Technical Summary
In existing technologies, the efficiency of marketing campaign generation is low, and operators need to invest a lot of time in manual configuration, resulting in low efficiency.
By obtaining the marketing campaign configuration text, performing semantic analysis, filtering, sorting and configuring activity modules, and combining artificial intelligence technology to automatically generate marketing campaigns, including semantic analysis, activity module combination and user push optimization.
It improves the automation and efficiency of marketing campaign generation, reduces the impact of human factors, and enhances the efficiency and effectiveness of marketing campaign generation and promotion.
Smart Images
Figure CN120806995A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of text processing, and is suitable for the financial field, and particularly relates to a marketing activity generation method and device, an electronic device and a storage medium. BACKGROUND
[0002] A marketing activity is an activity for product promotion. For example, in the insurance field, an operator usually designs a marketing activity around a specific insurance product to attract potential customers and promote product sales. At present, when generating a marketing activity, the operator usually manually configures an activity node based on a standard operation procedure (SOP) document. This method requires the operator to spend a lot of time learning and operating, and reduces the efficiency of marketing activity generation. Therefore, how to improve the efficiency of marketing activity generation has become a problem to be solved. SUMMARY
[0003] The main purpose of the embodiments of the present application is to provide a marketing activity generation method and device, an electronic device and a storage medium, which aims to improve the efficiency of marketing activity generation.
[0004] To achieve the above purpose, a first aspect of the embodiments of the present application provides a marketing activity generation method, which comprises:
[0005] Obtaining a marketing activity configuration text;
[0006] Performing semantic analysis on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data has an activity sequence identifier, an activity type identifier and activity configuration data;
[0007] Filtering a preset activity module according to the activity type identifier to obtain a candidate activity module;
[0008] According to the activity sequence identifier, the candidate activity module is sorted to obtain a sorted activity module;
[0009] According to the activity configuration data, the sorted activity module is configured to obtain a target activity module;
[0010] Combining the target activity module to obtain a target marketing activity.
[0011] In some embodiments, after the activity configuration data is configured to obtain the target activity module, the method further comprises:
[0012] Sending the target activity module to a preset operation end to obtain a configuration optimization text returned by the operation end based on the target activity module;
[0013] According to the configuration optimization text, the target activity module is configured and optimized to obtain an optimized activity module, and the optimized activity module is taken as the target activity module.
[0014] In some embodiments, after the target activity module is combined to obtain the target marketing activity, the method further includes:
[0015] Obtaining a reference marketing activity;
[0016] According to the marketing activity type, the reference marketing activity and the target marketing activity are clustered to obtain a target clustering area; wherein the target clustering area includes the target marketing activity and the remaining marketing activities;
[0017] According to the remaining marketing activities, a preset user is screened to obtain a target user;
[0018] According to the remaining marketing activities, a push time of the target user is predicted to obtain an activity push time;
[0019] According to the activity push time, the target marketing activity is pushed to the target user.
[0020] In some embodiments, according to the remaining marketing activities, the preset user is screened to obtain a target user, including:
[0021] Obtaining an activity push record of the preset user, and matching the activity push record based on the remaining marketing activities to screen out a record corresponding to the remaining marketing activities to obtain an activity matching record;
[0022] According to the activity matching record, the preset user is screened to obtain a selected user;
[0023] Obtaining a click rate, a conversion rate and a user retention rate of the selected user on the remaining marketing activities;
[0024] According to the click rate, the conversion rate and the user retention rate, the selected user is screened to obtain the target user.
[0025] In some embodiments, according to the remaining marketing activities, the target user is pushed to obtain an activity push time, including:
[0026] Obtaining an activity click timestamp and an activity conversion timestamp of the target user on the remaining marketing activities, and obtaining target user information of the target user;
[0027] According to the activity click timestamp, the activity conversion timestamp, the target user information and the reference marketing activity, a preset push time prediction model is trained to obtain a target push time prediction model;
[0028] Information extraction is performed on the target marketing activity to obtain target activity information.
[0029] According to the target push time prediction model, the target activity information and the target user information, a push time is predicted to obtain the activity push time.
[0030] In some embodiments, the training of the preset push time prediction model according to the activity click timestamp, the activity conversion timestamp, the target user information and the reference marketing activity to obtain the target push time prediction model comprises:
[0031] The difference between the activity conversion timestamp and the activity click timestamp is calculated to obtain a reference push conversion time.
[0032] Information extraction is performed on the reference marketing activity to obtain reference activity information.
[0033] According to the reference activity information, the reference push conversion time and the target user information, the preset push time prediction model is trained to obtain the target push time prediction model.
[0034] In some embodiments, the pushing of the target marketing activity to the target user according to the activity push time comprises:
[0035] The behavior data of the target user is monitored to determine a trigger timestamp when the target user occurs a preset trigger event.
[0036] According to the trigger timestamp and the activity push time, a target push time is determined.
[0037] When the target push time is reached, the target marketing activity is pushed to the target user.
[0038] To achieve the above-mentioned purpose, a second aspect of the embodiment of the present application proposes a marketing activity generation device, which comprises:
[0039] An acquisition data module is configured to acquire a marketing activity configuration text.
[0040] A semantic analysis module is configured to perform semantic analysis on the marketing activity configuration text to obtain marketing activity configuration data, wherein the marketing activity configuration data has an activity sequence identifier, an activity type identifier and activity configuration data.
[0041] an activity filtering module configured to filter preset activity modules according to the activity type identifier to obtain candidate activity modules;
[0042] an activity sorting module configured to sort the candidate activity modules according to the activity sequence identifier to obtain sorted activity modules;
[0043] an activity configuration module configured to configure the sorted activity modules according to the activity configuration data to obtain target activity modules;
[0044] an activity combination module configured to combine the target activity modules to obtain a target marketing activity.
[0045] To achieve the above object, a third aspect of the embodiments of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the method of the first aspect when executing the computer program.
[0046] To achieve the above object, a fourth aspect of the embodiments of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method of the first aspect.
[0047] The marketing activity generation method and device, electronic device and storage medium provided by the present application obtain a marketing activity configuration text, perform semantic analysis on the marketing activity configuration text to obtain marketing activity configuration data, wherein the marketing activity configuration data has an activity sequence identifier, an activity type identifier and activity configuration data, filter preset activity modules according to the activity type identifier to obtain candidate activity modules, sort the candidate activity modules according to the activity sequence identifier to obtain sorted activity modules, configure the sorted activity modules according to the activity configuration data to obtain target activity modules, and combine the target activity modules to obtain a target marketing activity. In this way, the embodiments of the present application determine the activity sequence, type and specific configuration content by performing semantic analysis on the marketing activity configuration text, and then accurately filter, reasonably sort and specifically configure the preset activity modules to finally combine to form a complete target marketing activity. The embodiments of the present application do not need an operator to manually configure and generate a marketing activity by relying on a standard operation process document, which not only eliminates the tedious manual operation steps, effectively reduces the influence of human factors on the activity generation process, but also significantly improves the automation degree of the marketing activity generation and the generation efficiency. BRIEF DESCRIPTION OF DRAWINGS
[0048] Figure 1 is a flowchart of the marketing activity generation method provided by the embodiments of the present application;
[0049] Figure 2 is a flowchart of the marketing activity generation method provided by another embodiment of the present application;
[0050] Figure 3 is a flowchart of the marketing activity generation method provided by another embodiment of the present application;
[0051] Figure 4 is a flowchart of step S303 in Figure 3
[0052] Figure 5 is a flowchart of step S304 in Figure 3
[0053] Figure 6 is a flowchart of step S502 in Figure 5
[0054] Figure 7 is a flowchart of step S105 in Figure 1
[0055] Figure 8 is a structural schematic diagram of the marketing activity generation apparatus provided by an embodiment of the present application;
[0056] Figure 9 is a hardware structural schematic diagram of the electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION
[0057] In order to make the objects, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not intended to limit the present application.
[0058] It should be noted that although the functional modules are divided in the apparatus schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a manner different from the module division in the apparatus or the order in the flowchart. The terms "first", "second", etc. in the specification and claims and the above-described drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.
[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application, and are not intended to limit the present application.
[0060] First, several terms involved in the present application are analyzed:
[0061] Long Short-Term Memory Neural Network (LSTM): Long Short-Term Memory Neural Network is a recurrent neural network structure used for processing and predicting time series data, which is an important branch of deep learning and sequence modeling techniques in artificial intelligence. This model introduces a gating mechanism to explicitly control the memory and forgetting process of information in the neural network, effectively alleviating the gradient vanishing and gradient explosion problems that traditional recurrent neural networks are prone to in long sequence training. The core structure of LSTM network includes input gate, forget gate and output gate, each of which is responsible for managing the fusion and update of current information and historical information, so that the model can both retain long-term dependency information and respond sensitively to short-term dynamics. Long Short-Term Memory Neural Network is widely used in speech recognition, machine translation, text generation, sentiment analysis, financial prediction, handwriting recognition and other tasks that require capturing context dependence and time series structure, and is one of the basic modules for building natural language processing, time series signal modeling and generation models. LSTM shows strong generalization ability and stability in processing unstructured time series data, and promotes the development of various language and behavior intelligent systems.
[0062] Decision Tree: Decision Tree is a classification and regression model based on tree structure, used for splitting and decision path modeling of data, and is one of the important algorithms in machine learning and data mining in artificial intelligence. Decision Tree constructs a structure composed of nodes and branches by layer-by-layer partitioning of feature space, each internal node represents a feature judgment condition, each branch represents a value path of the condition, and leaf nodes correspond to the final prediction output category or value. During model training, information gain, Gini coefficient or variance minimization are usually used to select the optimal partition feature, so as to realize the step-by-step classification or fitting of samples. Decision Tree model has clear structure and strong logical interpretability, and is widely used in risk assessment, credit scoring, medical diagnosis, marketing recommendation, text classification and other fields, and is also one of the basic components of ensemble learning methods such as random forest and gradient boosting tree. As a non-parametric model, Decision Tree is suitable for processing data sets with mixed numerical and categorical features, and is one of the key algorithms for building efficient, interpretable and low preprocessing requirement intelligent systems.
[0063] Random Forest Model: Random Forest Model is a supervised learning algorithm based on the idea of ensemble learning, which classifies or regresses input data by integrating multiple decision tree models. It is one of the important technologies in machine learning and data modeling in artificial intelligence. During training, the model uses random sampling and feature subset selection mechanisms, uses the Bagging strategy to sample the original data set with replacement, trains multiple structurally differentiated decision trees, and in the prediction stage, uses majority voting or average output to improve the stability and generalization ability of the overall model. Random Forest Model has strong anti-overfitting ability and robustness, and is suitable for processing high-dimensional data, nonlinear relationship modeling, and feature importance evaluation, etc. It is widely used in financial risk control, biological information, image recognition, speech analysis, medical diagnosis and behavior prediction, etc. Random Forest Model is one of the important methods for building high-performance, scalable and easy-to-deploy intelligent analysis systems due to its strong parallelism in training process, good interpretability of results, and certain tolerance to outliers and missing values.
[0064] K-means Clustering Algorithm: K-means clustering algorithm is a commonly used unsupervised learning method, which is used to divide data sets into several non-overlapping clusters. Each cluster has high similarity in feature space and belongs to one of the important algorithms in clustering analysis and data mining in artificial intelligence. The algorithm presets the number of clusters K, randomly initializes K cluster centers, and iteratively performs two steps: one is to assign each data point to the cluster represented by the nearest cluster center; the second is to update the cluster center position according to the mean of all samples in the current cluster, until the cluster assignment result is stable or the maximum iteration number is reached. K-means clustering has the advantages of high computational efficiency, simple implementation, strong adaptability, etc., and is widely used in image segmentation, market segmentation, behavior analysis, anomaly detection and recommendation system, etc. The algorithm is suitable for processing large-scale numerical data, but is sensitive to initial centers and prone to local optimum, so in practical application, it is often combined with multiple initialization or improved algorithm (such as K-means++) to improve the clustering effect. As a basic clustering method, K-means clustering is one of the key tools for building intelligent analysis models and pattern discovery systems.
[0065] Marketing activities are activities for product promotion, for example, in the insurance field, operators usually design marketing activities around specific insurance products to attract potential customers and promote product sales. Currently, operators usually use manual methods based on standard operating procedure (SOP) documents to configure activity nodes when generating marketing activities. This method requires operators to spend a lot of time learning and operating, reducing the efficiency of marketing activity generation. Therefore, how to improve the efficiency of marketing activity generation has become a problem to be solved.
[0066] Based on this, the embodiment of the application provides a marketing activity generation method and device, an electronic device and a storage medium, aiming to improve the efficiency of marketing activity generation.
[0067] The marketing activity generation method and device, the electronic device and the storage medium provided by the embodiment of the application are specifically described through the following embodiments. First, the marketing activity generation method in the embodiment of the application is described.
[0068] The embodiment of the application can acquire and process related data based on artificial intelligence technology. The artificial intelligence (AI) is a theory, method, technology and application system for simulating, extending and expanding human intelligence by using a digital computer or a machine controlled by a digital computer, perceiving an environment, acquiring knowledge and using the knowledge to obtain the best results.
[0069] The artificial intelligence basic technology generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The artificial intelligence software technology mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology and machine learning / deep learning, etc.
[0070] The marketing activity generation method provided by the embodiment of the application relates to the technical field of text processing and is suitable for the financial field. The marketing activity generation method provided by the embodiment of the application can be applied in a terminal, can be applied in a server end, and can also be software running in a terminal or a server end. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc.; the server end can be configured as an independent physical server, can be configured as a server cluster or a distributed system composed of multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN and big data and artificial intelligence platform, etc.; and the software can be an application for implementing the marketing activity generation method, but is not limited to the above forms.
[0071] The application is operable in a variety of general purpose or special purpose computer systems environments or configurations. Examples of well-known computing systems, environments, and / or configurations that can be suitable for use with the application include personal computers, server computers, handheld or laptop devices, tablet devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like. The application can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. The application can also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules can be located in both local and remote computer storage media including memory storage devices.
[0072] It should be noted that in each specific embodiment of the present application, when it is necessary to process relevant data related to the identity or characteristics of the user according to user information, user behavior data, user history data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiments of the present application need to obtain sensitive personal information of the user, the separate permission or separate consent of the user will be obtained through a pop-up window or a jump to a confirmation page, and after obtaining the separate permission or separate consent of the user, the necessary user-related data for enabling the embodiments of the present application to normally operate will be obtained.
[0073] Figure 1 is an optional flowchart of the marketing campaign generation method provided by the embodiments of the present application, Figure 1 The method in the above step S100 can include but is not limited to steps S101 to S106.
[0074] In step S101, a marketing campaign configuration text is obtained.
[0075] In step S102, the marketing campaign configuration text is subjected to semantic analysis to obtain marketing campaign configuration data; wherein the marketing campaign configuration data has an activity order identifier, an activity type identifier, and activity configuration data.
[0076] In step S103, the pre-set activity modules are filtered according to the activity type identifier to obtain candidate activity modules.
[0077] In step S104, the candidate activity modules are sorted according to the activity order identifier to obtain sorted activity modules.
[0078] Step S105, configuring the sorting activity module according to the activity configuration data, to obtain a target activity module;
[0079] Step S106, combining the target activity module to obtain a target marketing activity.
[0080] The steps S101 to S106 shown in the embodiments of the present application are as follows: obtaining a marketing activity configuration text; performing semantic analysis on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data has activity order identification, activity type identification and activity configuration data; filtering a preset activity module according to the activity type identification to obtain a candidate activity module; sorting the candidate activity module according to the activity order identification to obtain a sorting activity module; configuring the sorting activity module according to the activity configuration data to obtain a target activity module; and combining the target activity module to obtain a target marketing activity. In this way, the embodiments of the present application determine the activity order, type and specific configuration content by performing semantic analysis on the marketing activity configuration text, and then realize accurate filtering, reasonable sorting and targeted configuration of the preset activity module, and finally form a complete target marketing activity. The embodiments of the present application do not need an operator to manually configure and generate a marketing activity according to a standard operation process document, which not only eliminates the tedious manual operation steps, effectively reduces the influence of human factors on the activity generation process, but also significantly improves the automation degree of marketing activity generation and the generation efficiency.
[0081] In some embodiments, in step S101, the marketing activity configuration text is a text input by an operator for designing a marketing activity. For example, the operator generates a marketing activity configuration text in the form of keyboard input or voice input according to the demand of promoting an insurance product, for example, the operator inputs by keyboard: "first send an insurance product discount short message, then follow up the customers with response intention through telephone, and finally send detailed product information through email." Thus, the marketing activity configuration text is obtained.
[0082] In some embodiments, in step S102, semantic analysis refers to a process of performing semantic analysis on the marketing activity configuration text to obtain structured data. The implementation of semantic analysis is as follows: based on a preset natural language processing model, the marketing activity configuration text is analyzed to extract keywords and semantic relationships, and the unstructured text is converted into structured data. For example, the marketing activity configuration text is "send a short message first, then follow up by telephone", and through semantic analysis, the above text is converted into structured data with the activity order "short message-telephone" and the activity types "short message activity" and "telephone follow-up activity" respectively.
[0083] The marketing activity configuration data is structured data obtained by performing semantic analysis on the marketing activity configuration text, and specifically includes activity sequence identifiers, activity type identifiers, and activity configuration data. The activity sequence identifiers represent the execution order of the modules of the marketing activity. For example, in the above example, "SMS" is the first order, and "telephone follow-up" is the second order. The activity type identifiers represent the specific execution type of the modules of the marketing activity, such as "SMS activity", "telephone follow-up activity", or "mail activity". The activity configuration data represents the detailed parameters of the specific marketing activity, such as the SMS template content and the sending time corresponding to the "SMS activity", and the target customer group and the call script corresponding to the "telephone follow-up activity".
[0084] In some embodiments, in step S103, the activity type identifier is used to indicate the type to which each activity in the marketing activity configuration data belongs, and the preset activity module is a plurality of different types of marketing activity templates stored in advance. According to the activity type identifier, the preset activity module is filtered to retain only the marketing activity templates matching the activity type identifier, thereby forming the candidate activity module. For example, when the activity type identifier is "SMS push", the SMS template is filtered from the preset activity module, and when the activity type identifier is "telephone follow-up", the telephone communication script module is filtered from the preset activity module.
[0085] In some embodiments, in step S104, the candidate activity module is arranged in the order specified by the marketing activity configuration data according to the activity sequence identifier, thereby forming the sorted activity module. For example, the activity sequence identifier in the marketing activity configuration data is "SMS push first and telephone follow-up later", and the SMS push module is sorted first and the telephone follow-up module is sorted later.
[0086] In some embodiments, in step S105, the sorted activity module is configured with specific parameters according to the activity configuration data contained in the marketing activity configuration data, thereby obtaining the target activity module. The activity configuration data includes specific parameters and content required for executing the marketing activity, such as sending time, push content template, and customer selection conditions. For example, there is an SMS push module in the sorted activity module, and the activity configuration data specifies that the sending time of the SMS push module is 9:00 am and the SMS content is "insurance product discount offer". The SMS push module is configured according to the above specified parameters to form the target activity module that can be executed.
[0087] In some embodiments, in step S106, the target activity modules are combined in the sorted order to form a complete target marketing activity, i.e., an executable marketing activity is automatically generated. For example, a configured short message push module, a telephone follow-up module, and an email sending module are combined in sequence to form a complete marketing activity of sequentially pushing short messages, then calling customers for their intentions, and finally sending detailed product information through emails.
[0088] Referring to Figure 2 In some embodiments, after step S106, the marketing activity generation method can further include, but is not limited to, steps S201 to S202:
[0089] In step S201, the target activity module is sent to a preset operation end to obtain a configuration optimization text returned by the operation end based on the target activity module;
[0090] In step S202, the target activity module is configured and optimized based on the configuration optimization text to obtain an optimized activity module, and the optimized activity module is used as the target activity module.
[0091] The steps S201 to S202 shown in the embodiments of the present application send the target activity module to the preset operation end to obtain a configuration optimization text returned by the operation end based on the target activity module, configure and optimize the target activity module based on the configuration optimization text to obtain an optimized activity module, and use the optimized activity module as the target activity module.
[0092] Therefore, the embodiments of the present application send the target activity module to the preset operation end, use the operation personnel of the operation end to manually confirm and feed back the configuration optimization text, further automatically configure and optimize the target activity module according to the configuration optimization text, so as to obtain the optimized activity module, so that the operation personnel do not need to manually configure but only need to review, thereby improving the generation efficiency of the marketing activity.
[0093] In step S201 of some embodiments, the preset operation end is a platform for the operation personnel to check the activity module, and after the target activity module is sent to the preset operation end, the operation end manually reviews and optimizes the target activity module. For example, the target activity module is a short message push activity module, and after the operation end receives the module, the operation end can adjust the sending time or content of the short message based on actual promotion needs, and returns a configuration optimization text containing the optimized sending time and short message content.
[0094] In step S202 of some embodiments, according to the configuration optimization text received from the operation end, the specific configuration parameters of the target activity module are automatically adjusted and optimized, an optimized activity module is obtained, and the optimized activity module is updated as the target activity module for subsequent marketing activity execution. For example, the configuration optimization text returned by the operation end indicates that the short message push time should be modified from 9 am to 10 am, then the short message sending time of the target activity module is updated according to the text, and the optimized target activity module is formed to improve the implementation effect of the marketing activity.
[0095] Please refer to Figure 3 In some embodiments, after step S106, the marketing activity generation method can further include but is not limited to steps S301 to S305:
[0096] Step S301, obtaining a reference marketing activity;
[0097] Step S302, clustering the reference marketing activity and the target marketing activity according to the marketing activity type to obtain a target clustering area; wherein the target clustering area includes the target marketing activity and the remaining marketing activities;
[0098] Step S303, filtering a preset user according to the remaining marketing activities to obtain a target user;
[0099] Step S304, predicting a push time of the target marketing activity to the target user according to the remaining marketing activities;
[0100] Step S305, pushing the target marketing activity to the target user according to the push time.
[0101] The steps S301 to S305 shown in the embodiments of the present application obtain a reference marketing activity, cluster the reference marketing activity and the target marketing activity according to the marketing activity type to obtain a target clustering area, wherein the target clustering area includes the target marketing activity and the remaining marketing activities, filter a preset user according to the remaining marketing activities to obtain a target user, predict a push time of the target marketing activity to the target user according to the remaining marketing activities, and push the target marketing activity to the target user according to the push time. Therefore, the embodiments of the present application obtain a reference marketing activity, cluster the reference marketing activity and the target marketing activity according to the marketing activity type, construct a target clustering area containing the target marketing activity, and accurately determine the target user and predict the best push time based on the remaining marketing activities in the target clustering area, so as to accurately push the target marketing activity to the target user at the appropriate push time. The embodiments of the present application effectively filter the target user and improve the accuracy of the marketing push time, thereby improving the promotion effect of the marketing activity.
[0102] In step S301 of some embodiments, the reference marketing activity is a historical marketing activity that has been executed, for example, the reference marketing activity can be a previous short message push insurance discount promotion activity for a 30-45 year-old group.
[0103] In step S302 of some embodiments, the marketing activity type is a specific classification of the promotion form corresponding to the marketing activity, such as a short message promotion type, a telephone call back type, or an email promotion type. Clustering refers to classifying the reference marketing activity and the target marketing activity according to similar characteristics to obtain a group of marketing activities with common attributes. The implementation of clustering is to use clustering analysis algorithms, such as K-means clustering algorithm or hierarchical clustering algorithm, to analyze the characteristics such as marketing activity type and activity effect data, to divide into multiple similar marketing activity groups.
[0104] Please refer to Figure 4 In some embodiments, step S303 can include but is not limited to steps S401 to S404:
[0105] Step S401, obtaining the activity push record of the preset user, and matching the activity push record based on the remaining marketing activities to filter out the records corresponding to the remaining marketing activities, and obtaining the activity matching record;
[0106] Step S402, filtering the preset user according to the activity matching record to obtain the selected user;
[0107] Step S403, obtaining the click rate, conversion rate, and user retention rate of the selected user on the remaining marketing activities;
[0108] Step S404, filtering the selected user according to the click rate, conversion rate, and user retention rate to obtain the target user.
[0109] The steps S401 to S404 shown in the embodiments of the present application are as follows: the activity push record of the preset user is obtained, the activity push record is matched based on the remaining marketing activities, the record corresponding to the remaining marketing activities is filtered out to obtain an activity matching record; the preset user is filtered according to the activity matching record to obtain a selected user; the click rate, conversion rate and user retention rate of the selected user to the remaining marketing activities are obtained; and the selected user is filtered according to the click rate, conversion rate and user retention rate to obtain a target user. In this way, the embodiments of the present application analyze the historical activity push record of the preset user, filter out the user to which the remaining marketing activities are pushed, and obtain the selected user; and then the selected user is further filtered based on the click rate, conversion rate and user retention rate of the selected user to the remaining marketing activities, so as to effectively determine the target user. That is, the embodiments of the present application first filter out the user to which the other marketing activities are pushed, and then filter out the user with high click rate, high conversion rate and high user retention rate from the users, to obtain the target user.
[0110] In step S401 of some embodiments, the activity push record is historical push information, which records the specific situation of the preset user receiving the marketing activities in the past. The activity push record is matched based on the remaining marketing activities, that is, the marketing activities that have been pushed are found from the activity push record, which are one of the remaining marketing activities. That is to say, the marketing activities that a certain user has been pushed are one of the remaining marketing activities. For example: the remaining marketing activities include: A, B, C and D. The activity push record of the preset user includes: A, E and G. Then the activity matching record is A.
[0111] In step S402 of some embodiments, the preset user is filtered according to the activity matching record, that is, the user whose activity push record matches the remaining marketing activities is selected from the preset user as a selected user. In other words, only those users with corresponding records of the remaining marketing activities in the historical push information will be selected. For example: the remaining marketing activities include: A, B, C and D. The activity push record of the preset user includes: E and G. Then the activity matching record is null. Therefore, the preset user will not be filtered as a selected user.
[0112] In step S403 of some embodiments, the click rate represents the ratio between the number of times of clicking to view the related activity content and the number of times of push after the selected user receives the push of the remaining marketing activities; the conversion rate represents the proportion of the users who actually generate a purchase behavior or other target behavior after clicking to view the content of the marketing activities; and the user retention rate represents the proportion of the selected users who continue to use or pay attention to the corresponding products or services after participating in the marketing activities.
[0113] In step S404 of some embodiments, the selected users are filtered according to the click rate, conversion rate and user retention rate, and possible filtering methods include but are not limited to one or more of the following: setting a specific threshold, for example, only selecting users with a click rate higher than a certain percentage (such as 20%), a conversion rate higher than a certain value (such as 10%) or a user retention rate higher than a certain value (such as 30%); or by comprehensive score, i.e. assigning a certain weight to the click rate, conversion rate and user retention rate, calculating the total score, and selecting users with a comprehensive score reaching a preset standard as target users. For example, the selected users can be set to have a click rate higher than 25%, a conversion rate higher than 15% and a user retention rate higher than 40% as the filtering standard.
[0114] Please refer to Figure 5 In some embodiments, step S304 includes but is not limited to steps S501 to S504:
[0115] Step S501, obtaining the activity click timestamp and activity conversion timestamp of the target user on the remaining marketing activities, and obtaining the target user information of the target user;
[0116] Step S502, training the preset push timing prediction model according to the activity click timestamp, the activity conversion timestamp, the target user information and the reference marketing activity, to obtain a target push timing prediction model;
[0117] Step S503, information extraction is performed on the target marketing activity to obtain target activity information;
[0118] Step S504, push timing prediction is performed according to the target push timing prediction model, the target activity information and the target user information, to obtain the activity push time.
[0119] The steps S501 to S504 shown in the embodiments of the present application obtain the activity click timestamp and activity conversion timestamp of the target user on the remaining marketing activities, and obtain the target user information of the target user; train the preset push timing prediction model according to the activity click timestamp, the activity conversion timestamp, the target user information and the reference marketing activity, to obtain a target push timing prediction model; perform information extraction on the target marketing activity to obtain target activity information; and perform push timing prediction according to the target push timing prediction model, the target activity information and the target user information, to obtain the activity push time. In this way, the embodiments of the present application train a target push timing prediction model capable of accurately predicting the best push time by combining the timestamps of the clicks and conversions of the target user on historical marketing activities and the relevant information of the user itself, and historical reference marketing activities; and then accurately predict the best push timing of the target marketing activity based on the prediction model, thereby improving the push accuracy of the marketing activity and further improving the push effect.
[0120] In step S501 of some embodiments, the activity click timestamp is the specific time point recorded when the target user clicks the push content of the remaining marketing activities, for example, the specific time when the target user clicks a certain insurance marketing short message at 10:30 am on June 15, 2024. The activity conversion timestamp is the specific time point recorded when the target user completes a specific expected conversion behavior (such as purchasing insurance or submitting personal information) after clicking the marketing activity content; the target user information is the basic attributes and historical interaction information of the target user, such as the age, gender, income level, geographic location of the target user, and historical activity response behavior records.
[0121] Please refer to Figure 6 In some embodiments, step S502 includes but is not limited to steps S601 to S603:
[0122] Step S601, calculate the difference between the activity conversion timestamp and the activity click timestamp to obtain the reference push conversion time;
[0123] Step S602, information extraction is performed on the reference marketing activity to obtain reference activity information;
[0124] Step S603, train the preset push timing prediction model according to the reference activity information, the reference push conversion time, and the target user information to obtain a target push timing prediction model.
[0125] The steps S601 to S603 shown in the embodiments of the present application calculate the difference between the activity conversion timestamp and the activity click timestamp to obtain the reference push conversion time; perform information extraction on the reference marketing activity to obtain reference activity information; and train the preset push timing prediction model according to the reference activity information, the reference push conversion time, and the target user information to obtain a target push timing prediction model. In this way, the embodiments of the present application calculate the reference push conversion time required by the user from clicking to conversion in the marketing activity, and combine the activity characteristic information of the reference marketing activity and the target user information to train the target push timing prediction model, so as to accurately predict the best push timing of subsequent marketing activities.
[0126] In step S602 of some embodiments, the reference marketing activity is a historical marketing activity that has been executed, including but not limited to activity type, activity content, etc. Information extraction refers to the structured data extracted from the reference marketing activity. The specific content of information extraction can include but is not limited to the type of marketing activity, marketing content keywords, push channel type, etc. For example, through information extraction, the original marketing activity data "short message channel, insurance promotion discount content" can be converted into structured reference activity information for training of the subsequent push timing prediction model.
[0127] In step S603 of some embodiments, the preset push timing prediction model is used to analyze the click and conversion time regularity of the target user on the remaining marketing activities and the related user information, and then predict the best response time of the target user on the future marketing activities. The preset push timing prediction model may be, for example, a decision tree model, a random forest model, or a long short-term memory neural network (LSTM), etc., wherein the long short-term memory neural network model is suitable for capturing long-term regularity and short-term fluctuations in user historical activity timestamps. For example, by analyzing and training the click and conversion timestamp sequence of the target user in the past three months through the long short-term memory neural network, a target push timing prediction model is established to accurately predict the best response time of the target user when receiving the next insurance preferential push.
[0128] In steps S503 to S504 and steps S602 to S603 of some embodiments, the principles are similar, and details are not repeated here.
[0129] Please refer to Figure 7 In some embodiments, step S106 may include, but is not limited to, steps S701 to S703:
[0130] Step S701, monitoring the behavior data of the target user to determine the trigger timestamp when the target user occurs a preset trigger event;
[0131] Step S702, calculating according to the trigger timestamp and the activity push time to determine the target push time;
[0132] Step S703, when the target push time is reached, pushing the target marketing activity to the target user.
[0133] The steps S701 to S703 shown in the embodiments of the present application determine the trigger timestamp when the target user occurs a preset trigger event by monitoring the behavior data of the target user; calculate the target push time according to the trigger timestamp and the activity push time; and when the target push time is reached, push the target marketing activity to the target user. In this way, the embodiments of the present application determine the timestamp when the target user triggers a specific event in real time by monitoring the behavior data of the target user, further calculate the accurate target push time in combination with the activity push time, and dynamically push the target marketing activity to the target user when the target push time is reached, thereby improving the accuracy of the push timing of the marketing activity.
[0134] In step S701 of some embodiments, the target user's behavioral data includes real-time operational data on the platform or system, such as browsing insurance product pages, inquiring about products, adding products to favorites or shopping carts, and so on. Preset trigger events are specific behaviors that can indicate the target user's potential interest or purchasing intent, such as browsing an insurance product page for a duration exceeding a set threshold. When the target user executes one of these preset trigger events, the time at which the event occurred is recorded, which serves as the trigger timestamp.
[0135] In step S702 of some embodiments, the target push time is calculated based on the trigger timestamp and combined with the previously predicted activity push time. For example, when the trigger timestamp is 9 a.m. and the predicted activity push time is 1 hour after the trigger event occurs, the determined target push time is 10 a.m.
[0136] In step S703 of some embodiments, when the time reaches the target push time, the target marketing activity is pushed to the target user terminal device, for example, sent to the target user's mobile terminal in the form of text message, email or platform push message, to ensure that the push timing of the marketing activity is accurate and effective.
[0137] See also Figure 8 The present application also provides a marketing activity generating device that can implement the above-mentioned marketing activity generating method. The device includes:
[0138] The data acquisition module 801 is used to acquire the marketing campaign configuration text;
[0139] Semantic parsing module 802, configured to perform semantic parsing on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data includes an activity sequence identifier, an activity type identifier, and activity configuration data;
[0140] An activity screening module 803 is used to screen preset activity modules according to activity type identifiers to obtain candidate activity modules;
[0141] An activity sorting module 804 is used to sort the candidate activity modules according to the activity sequence identifiers to obtain sorted activity modules;
[0142] The activity configuration module 805 is used to configure the sorting activity module according to the activity configuration data to obtain the target activity module;
[0143] The activity combination module 806 is used to combine the target activity modules to obtain the target marketing activity.
[0144] The specific implementation of the marketing activity generating device is substantially the same as the specific embodiment of the above-mentioned marketing activity generating method, and will not be described in detail here.
[0145] The embodiment of the present application further provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the marketing activity generation method. The electronic device can be any intelligent terminal, such as a tablet computer or a vehicle-mounted computer.
[0146] Please refer to Figure 9 , Figure 9 The hardware structure of the electronic device of another embodiment is shown, which comprises:
[0147] The processor 901 can be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is used to execute related programs to realize the technical solutions provided by the embodiments of the present application.
[0148] The memory 902 can be implemented in the form of a ROM (ReadOnly Memory), a static storage device, a dynamic storage device, or a RAM (Random Access Memory). The memory 902 can store an operating system and other application programs. When the technical solutions provided by the embodiments of the present application are implemented by software or firmware, the related program codes are stored in the memory 902 and are called and executed by the processor 901 to realize the marketing activity generation method of the embodiments of the present application.
[0149] The input / output interface 903 is used to realize information input and output.
[0150] The communication interface 904 is used to realize the communication interaction between the device and other devices, and can realize communication through a wired manner (such as a USB, a network cable, etc.) or a wireless manner (such as a mobile network, WIFI, Bluetooth, etc.).
[0151] The bus 905 is used to transmit information between various components (such as the processor 901, the memory 902, the input / output interface 903, and the communication interface 904) of the device.
[0152] The processor 901, the memory 902, the input / output interface 903, and the communication interface 904 are connected to each other through the bus 905 to realize communication connection between them in the device.
[0153] The embodiment of the application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the marketing activity generation method.
[0154] The memory, as a non-transitory computer readable storage medium, can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.
[0155] The embodiment of the application provides the marketing activity generation method, the marketing activity generation device, the electronic equipment and the storage medium, which obtain a marketing activity configuration text; perform semantic analysis on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data has an activity order identifier, an activity type identifier and activity configuration data; filter a preset activity module according to the activity type identifier to obtain a candidate activity module; sort the candidate activity module according to the activity order identifier to obtain a sorted activity module; configure the sorted activity module according to the activity configuration data to obtain a target activity module; and combine the target activity module to obtain a target marketing activity. Therefore, the embodiment of the application determines the activity order, the type and the specific configuration content by performing semantic analysis on the marketing activity configuration text, and then realizes accurate filtering, reasonable sorting and targeted configuration of the preset activity module, and finally combines to form a complete target marketing activity. The embodiment of the application does not need an operator to manually configure and generate a marketing activity by relying on a standard operation process document, which not only eliminates the tedious operation steps of manual operation, effectively reduces the influence of human factors on the activity generation process, but also significantly improves the automation degree of the marketing activity generation and the generation efficiency.
[0156] The embodiments described in the embodiments of the application are used to more clearly illustrate the technical solutions of the embodiments of the application, and do not constitute a limitation on the technical solutions provided by the embodiments of the application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the embodiments of the application are also applicable to similar technical problems.
[0157] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the application, and can include more or fewer steps than the figures, or combine certain steps, or different steps.
[0158] The apparatus embodiments described above are merely exemplary, and the units described as separate units can or can not be physically separate, i.e., can be located in one place, or can be distributed over multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purposes of the embodiments.
[0159] Those skilled in the art can understand that all or some of the steps in the method disclosed above, the functional modules / units in the system and the device can be implemented as software, firmware, hardware and appropriate combinations thereof.
[0160] The terms "first", "second", "third", "fourth" and the like in the description of the application and in the claims of the foregoing drawings, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the use of the terms so
[0161] It should be understood that in the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" is used to describe the relationship between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that there are three cases: only A, only B, and A and B at the same time, where A and B can be singular or plural. The character " / " generally represents that the associated objects before and after are in an "or" relationship. "At least one of the following" or the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can mean a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0162] In several embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented by other manners. For example, the apparatus embodiments described above are merely illustrative, for example, the division of the above units is merely a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or components shown or discussed can be indirect coupling or communication connection through some interfaces, apparatuses or units, and can be electrical, mechanical or other forms.
[0163] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they can be located in one place or distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.
[0164] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.
[0165] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part of the prior art that makes a contribution or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing 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 of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.
[0166] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, but this does not limit the scope of the embodiments of the present application. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.
Claims
1. A marketing campaign generation method, characterized in that: The method comprises: Get the campaign configuration text; Performing semantic parsing on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data includes an activity sequence identifier, an activity type identifier, and activity configuration data; Filtering the preset activity modules according to the activity type identifier to obtain candidate activity modules; Sorting the candidate activity modules according to the activity sequence identifiers to obtain sorted activity modules; Configuring the sorting activity module according to the activity configuration data to obtain a target activity module; The target activity modules are combined to obtain a target marketing activity.
2. The method according to claim 1, characterized in that After configuring the sorting activity modules according to the activity configuration data to obtain the target activity modules, the method further includes: Sending the target activity module to a preset operation terminal to obtain a configuration optimization text returned by the operation terminal based on the target activity module; The target activity module is configured and optimized based on the configuration optimization text to obtain an optimized activity module, and the optimized activity module is used as the target activity module.
3. The method according to claim 1, characterized in that After combining the target activity modules to obtain the target marketing activity, the method further includes: Get reference marketing campaigns; Clustering the reference marketing campaign and the target marketing campaign according to the marketing campaign type to obtain a target clustering area; wherein the target clustering area includes the target marketing campaign and other marketing campaigns; Filtering the preset users according to the remaining marketing activities to obtain target users; Predicting the push timing for the target user based on the remaining marketing activities to obtain the activity push time; The target marketing activity is pushed to the target user according to the activity push time.
4. The method according to claim 3, characterized in that The screening of preset users according to the remaining marketing activities to obtain target users includes: Obtaining the activity push records of the preset user, and matching the activity push records based on the remaining marketing activities, filtering out records corresponding to the remaining marketing activities, and obtaining activity matching records; Filtering the preset users according to the activity matching records to obtain selected users; Obtaining the click-through rate, conversion rate, and user retention rate of the selected users for the remaining marketing activities; The selected users are screened according to the click-through rate, the conversion rate, and the user retention rate to obtain the target users.
5. The method according to claim 3, characterized in that The predicting of the push timing for the target user based on the remaining marketing activities to obtain the activity push time includes: Obtain the target user's activity click timestamps and activity conversion timestamps for the remaining marketing activities, and obtain the target user information of the target user; Training a preset push timing prediction model according to the activity click timestamp, the activity conversion timestamp, the target user information, and the reference marketing activity to obtain a target push timing prediction model; Extracting information from the target marketing activity to obtain target activity information; The push timing is predicted based on the target push timing prediction model, the target activity information and the target user information to obtain the activity push time.
6. The method according to claim 5, characterized in that The step of training a preset push timing prediction model based on the activity click timestamp, the activity conversion timestamp, the target user information, and the reference marketing activity to obtain a target push timing prediction model includes: Calculate the difference between the activity conversion timestamp and the activity click timestamp to obtain a reference push conversion time; Extracting information from the reference marketing activity to obtain reference activity information; The preset push timing prediction model is trained according to the reference activity information, the reference push conversion time and the target user information to obtain a target push timing prediction model.
7. The method according to claim 3, characterized in that Pushing the target marketing activity to the target user according to the activity push time includes: Monitor the target user's behavior data and determine a trigger timestamp when a preset trigger event occurs to the target user; Calculate the target push time based on the trigger timestamp and the activity push time; When the target push time is reached, the target marketing activity is pushed to the target user.
8. A marketing activity generating device, characterized in that: The device comprises: Get data module, used to obtain marketing campaign configuration text; A semantic parsing module, configured to perform semantic parsing on the marketing activity configuration text to obtain marketing activity configuration data; wherein the marketing activity configuration data comprises an activity sequence identifier, an activity type identifier, and activity configuration data; An activity screening module, configured to screen preset activity modules according to the activity type identifier to obtain candidate activity modules; An activity sorting module, configured to sort the candidate activity modules according to the activity sequence identifiers to obtain sorted activity modules; an activity configuration module, configured to configure the sorting activity module according to the activity configuration data to obtain a target activity module; The activity combination module is used to combine the target activity modules to obtain a target marketing activity.
9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the marketing activity generation method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the marketing campaign generating method according to any one of claims 1 to 7 is implemented.