Artificial intelligence-based marketing system optimization method, apparatus, and device

Abstract

The application belongs to the field of financial technology, and relates to a marketing system optimization method and device based on artificial intelligence, computer equipment and a storage medium. The method comprises the following steps: performing data analysis on standard sample data to generate a user portrait; constructing and training a feature engineering model according to a training data set and a feature engineering strategy; inputting a test data set into the feature engineering model to extract key features, and training an algorithm model according to the key features to obtain a marketing prediction model; performing prediction accuracy evaluation on the marketing prediction model and comparison, adjusting the marketing prediction model according to a comparison result to obtain an optimized marketing prediction model; performing marketing pre-rehearsal, comparing a pre-rehearsal result with a marketing index, adjusting a marketing strategy according to a comparison result, and obtaining an optimized marketing system. The application can automatically optimize the marketing strategy and the marketing prediction model, and effectively improve the prediction accuracy of the marketing system for marketing activities.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, specifically to the field of financial technology, and particularly relates to a marketing system optimization method and device based on artificial intelligence, a computer device and a storage medium. BACKGROUND

[0002] A marketing campaign is a promotional method for extracting marketable customers from a target user group. In order to effectively implement a marketing campaign, a corresponding marketing strategy needs to be specified for the marketing campaign to ensure that the effect of the marketing campaign can reach the expected effect.

[0003] A marketing strategy simulation model system is a model for simulating and predicting a marketing strategy. By inputting a specified marketing strategy into the marketing strategy simulation model, the marketing effect that can be achieved by the marketing strategy can be determined, so as to select the optimal marketing strategy that meets the actual situation. The marketing strategy simulation model mainly uses a machine learning algorithm to train and learn a large amount of user historical data, and then analyzes the consumption habits, shopping preferences, search history, etc. of the user, accurately predicts the possible actions of the user, such as purchase, abandonment of purchase, attention, etc., improves the user activity, retention rate, and conversion rate, and also improves the prediction accuracy and efficiency, further promoting the user growth and business improvement of the enterprise.

[0004] However, the above marketing strategy simulation model still has some problems in actual application. For example, when formulating a marketing strategy, the target user is not fully and accurately understood, and the behavior habits and preferences of the user are not captured, resulting in the inability to grasp the real needs of the user and the difficulty in reaching the potential real users who can be converted. The marketing strategy simulation model needs a large amount of reliable data for support, but the data in reality is often not complete, and the data set may have problems such as difficult-to-handle noise, thereby affecting the accuracy and robustness of the model and leading to the final output prediction result being insufficient. At the same time, after predicting the marketing strategy, the marketing strategy needs to be manually optimized and simulated again to obtain the result, which is troublesome and cannot meet the demand for formulating a marketing strategy under big data. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a marketing system optimization method and device based on artificial intelligence, a computer device and a storage medium, to solve the problem of poor accuracy of marketing strategy simulation results and the inability to automatically optimize and adjust according to user needs.

[0006] To solve the above technical problems, the embodiments of the present application provide a marketing system optimization method based on artificial intelligence, which adopts the following technical solutions:

[0007] obtaining pre-processed standard sample data, performing data analysis on the standard sample data to obtain an analysis result, and generating a user portrait according to the analysis result;

[0008] obtaining a training data set, constructing and training a feature engineering model according to the training data set and a preset feature engineering strategy, and obtaining a trained feature engineering model;

[0009] obtaining a test data set, inputting the test data set into the trained feature engineering model for feature extraction to obtain key features, and training a preset algorithm model according to the key features to obtain a marketing prediction model;

[0010] performing prediction accuracy evaluation on the marketing prediction model to obtain a prediction accuracy score, and comparing the prediction accuracy score with a preset standard score;

[0011] if the prediction accuracy score is greater than or equal to the standard score, the marketing prediction model is used as an optimized marketing prediction model;

[0012] if the prediction accuracy score is less than the standard score, the model parameters of the marketing prediction model and the feature engineering strategy are iteratively adjusted until the prediction accuracy score is greater than or equal to the standard score, and after the prediction accuracy score is greater than or equal to the standard score, the adjusted marketing prediction model is used as the optimized marketing prediction model;

[0013] performing marketing pre-rehearsal according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy to obtain marketing prediction data; and

[0014] comparing the marketing prediction data with a preset marketing index to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system.

[0015] Further, the step of obtaining pre-processed standard sample data, performing data analysis on the standard sample data to obtain an analysis result, and generating a user portrait according to the analysis result, specifically comprises:

[0016] obtaining historical sample data, performing data cleaning and data screening on the historical sample data to obtain the standard sample data;

[0017] performing user data extraction on the standard sample data to obtain user attributes and user behavior data, and performing data association on the user attributes and the user behavior data to obtain user association information; and

[0018] The user association information is arranged, and a depiction operation is performed according to the arranged user association information, so that a corresponding user portrait is obtained.

[0019] Further, the step of constructing and training the feature engineering model according to the training data set and the preset feature engineering strategy specifically comprises:

[0020] The preset feature engineering strategy is acquired, and the preset feature engineering strategy is parsed to obtain data processing information;

[0021] The training data set is subjected to feature extraction to obtain data features, and the data features are subjected to variable selection to obtain effective features; and

[0022] The target processing step corresponding to the effective features is acquired in the data processing information, and the feature engineering model is constructed and trained according to the effective features and the target processing step.

[0023] Further, the step of evaluating the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score and comparing the prediction accuracy score with a preset standard score specifically comprises:

[0024] The marketing prediction result predicted by the marketing prediction model is acquired, and a prediction accuracy rate is calculated according to the marketing prediction result;

[0025] The prediction accuracy is evaluated according to the prediction accuracy rate to obtain the prediction accuracy score; and

[0026] The preset standard score is acquired, and the prediction accuracy score is compared with the preset standard score.

[0027] Further, the step of performing marketing pre-rehearsal according to the standard sample data, the user portrait, the optimized marketing prediction model and a preset marketing strategy to obtain marketing prediction data specifically comprises:

[0028] The standard sample data is input into the optimized marketing prediction model for prediction to obtain an optimized prediction result;

[0029] A user identifier corresponding to the optimized prediction result is acquired;

[0030] The user data is classified into a user class group corresponding to the user portrait according to the user identifier;

[0031] The preset marketing strategy is parsed to obtain pre-rehearsal data, and a pre-rehearsal step is extracted from the pre-rehearsal data; and

[0032] The pre-acting step is performed on the user group to obtain the marketing prediction data.

[0033] Further, the step of comparing the marketing prediction data with the preset marketing indicators to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system, specifically includes:

[0034] detecting whether the marketing prediction data meets the preset marketing indicators;

[0035] If the marketing prediction data meets the preset marketing indicators, the marketing strategy corresponding to the marketing prediction data is taken as the optimized marketing strategy, and the optimized marketing system is generated according to the optimized marketing strategy and the optimized marketing prediction model; and

[0036] If the marketing prediction data does not meet the marketing indicators, the marketing strategy is iteratively adjusted until the marketing prediction data meets the preset marketing indicators, and the adjusted marketing strategy is taken as the optimized marketing strategy, and the optimized marketing system is generated according to the optimized marketing strategy and the optimized marketing prediction model.

[0037] Further, after the step of comparing the marketing prediction data with the preset marketing indicators to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system, the following steps are further included:

[0038] detecting whether the number of the optimized marketing strategies is greater than one or equal to one;

[0039] If the number of the optimized marketing strategies is greater than one, the optimized marketing strategies are subjected to AB testing, the optimal marketing strategy is selected according to the test result, and the test result is recorded to generate a visual log report; and

[0040] If the number of the optimized marketing strategies is equal to one, the optimized marketing strategy is subjected to a strategy delivery test, and the test result is recorded to generate a visual log report.

[0041] To solve the above technical problems, the embodiment of the present application further provides an artificial intelligence-based marketing system optimization device, which adopts the following technical solutions:

[0042] A data analysis module is configured to obtain pre-processed standard sample data, perform data analysis on the standard sample data to obtain an analysis result, and generate a user portrait according to the analysis result.

[0043] The model construction module is configured to obtain a training data set, construct and train a feature engineering model according to the training data set and a preset feature engineering strategy, and obtain a trained feature engineering model.

[0044] The model training module is configured to obtain a test data set, input the test data set into the trained feature engineering model for feature extraction, obtain key features, and train a preset algorithm model according to the key features to obtain a marketing prediction model.

[0045] The model evaluation module is configured to evaluate the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score, and compare the prediction accuracy score with a preset standard score.

[0046] The first model output module is configured to, if the prediction accuracy score is greater than or equal to the standard score, take the marketing prediction model as an optimized marketing prediction model.

[0047] The second model output module is configured to, if the prediction accuracy score is less than the standard score, iteratively adjust the model parameters of the marketing prediction model and the feature engineering strategy until the prediction accuracy score is greater than or equal to the standard score, and take the corresponding marketing prediction model as the optimized marketing prediction model after the prediction accuracy score is greater than or equal to the standard score.

[0048] The strategy pre-performance module is configured to perform marketing pre-performance according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy to obtain marketing prediction data.

[0049] The strategy optimization module is configured to compare the marketing prediction data with a preset marketing index to obtain a comparison result, and adjust the marketing strategy according to the comparison result to obtain an optimized marketing system.

[0050] To solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the technical scheme as follows:

[0051] A computer device includes a memory and a processor, the memory stores computer readable instructions, and the processor executes the computer readable instructions to realize the steps of the marketing system optimization method based on artificial intelligence as described in any one of the above.

[0052] To solve the above technical problems, the embodiment of the present application further provides a computer readable storage medium, which adopts the technical scheme as follows:

[0053] A computer-readable storage medium having stored thereon computer-readable instructions which, when executed by a processor, implement the steps of the artificial intelligence-based marketing system optimization method of any of the above.

[0054] Compared with the prior art, the embodiment of the present application has the following beneficial effects: the embodiment can obtain the behavior habits and preferences of the user according to the user portrait, so as to formulate corresponding marketing strategies for the user groups formed after the classification of the user data, by performing data analysis on the preprocessed standard sample data and generating a user portrait according to the analysis result. The key features that can better train the algorithm model are obtained through the trained feature engineering model, by constructing and training the feature engineering model according to the training data set and the preset feature engineering strategy. The marketing prediction model that can accurately predict the marketing result of the marketing strategy is obtained by training the preset algorithm model after obtaining the key features of the test data set. The model parameters of the marketing prediction model and the feature engineering strategy corresponding to the feature engineering model are adjusted according to the comparison result of the evaluation score and the standard score, so as to obtain the optimized marketing prediction model that meets the accurate prediction requirement after adjustment and optimization. The marketing pre-play is performed according to the standard sample data, the user portrait, the optimized marketing prediction model and the preset marketing strategy, the marketing prediction data is obtained, the marketing prediction data and the preset marketing indicators are compared, the marketing strategy is adjusted according to the comparison result, and the optimized marketing system that can effectively improve the marketing prediction effect is generated according to the adjusted optimized marketing strategy and the optimized marketing prediction model. The embodiment can be applied to the optimization of the marketing system of the financial marketing activity, and can effectively improve the marketing prediction effect of the marketing system, so as to improve the conversion rate of the user. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the schemes in the present application, the drawings needed in the description of the embodiments of the present application will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0056] Figure 1 is an exemplary system architecture diagram to which the present application can be applied;

[0057] Figure 2 Flow chart of an embodiment of the artificial intelligence-based marketing system optimization method according to the present application;

[0058] Figure 3 is Figure 2 Flow chart of a specific implementation of step S10 in

[0059] Figure 4 is Figure 2 a flowchart of one specific implementation of step S20 in

[0060] Figure 5 is Figure 2 a flowchart of one specific implementation of step S40 in

[0061] Figure 6 is Figure 2 a flowchart of one specific implementation of step S70 in

[0062] Figure 7 is Figure 2 a flowchart of one specific implementation of step S80 in

[0063] Figure 8 is a structural schematic diagram of one embodiment of the marketing system optimization device based on artificial intelligence according to the present application;

[0064] Figure 9 is a structural schematic diagram of one embodiment of the computer device according to the present application. DETAILED DESCRIPTION

[0065] 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 this application belongs; the terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application; the description and claims of this application as well as the above abstract are intended to be construed in accordance with the scope of the appended claims and the legal principles governing the interpretation of patent claims, including the legal principles established by the courts of competent jurisdiction in the United States and the courts of other jurisdictions that have jurisdiction over patent law matters. The use herein of terms such as "comprise", "comprises", "comprising", "containing", "contain", "contains", "including", "includes", "include", "have" and "has" and any variations thereof are intended to be construed as non-exclusive. The use herein of terms such as "first", "second", and the like are intended to distinguish between different objects, and not to describe a particular order.

[0066] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase that invarious places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is expressly understood that the embodiments described herein are merely examples from a

[0067] In order to make the technical personnel in the art better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings.

[0068] As Figure 1As shown, the system architecture 100 can include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links, or fiber optic cables, and the like.

[0069] A user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, and the like. Various communication client applications can be installed on the terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, and the like.

[0070] The terminal devices 101, 102, 103 can be various electronic devices with display screens and support for web browsing, including but not limited to smartphones, tablet computers, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptop computers, desktop computers, and the like.

[0071] The server 105 can be a server that provides various services, such as a background server that provides support for pages displayed on the terminal devices 101, 102, 103.

[0072] It should be noted that the marketing system optimization method based on artificial intelligence provided by the embodiments of the present application is generally executed by a server, and accordingly, the marketing system optimization device based on artificial intelligence is generally provided in a server.

[0073] It should be understood that Figure 1 The number of terminal devices, networks, and servers in

[0074] With reference to Figure 2 , a flowchart of one embodiment of a method for system security monitoring calculation according to the present application is shown. The marketing system optimization method based on artificial intelligence includes the following steps:

[0075] In step S10, pre-processed standard sample data is obtained, data analysis is performed on the standard sample data, an analysis result is obtained, and a user portrait is generated according to the analysis result;

[0076] In the embodiment, the standard sample data is obtained from the historical sample data after preprocessing, and the user portrait refers to classifying the users in the standard sample data according to their behaviors and preference habits, and adding the behavior characteristics of the users to the data.

[0077] In step S20, a training data set is obtained, a feature engineering model is constructed and trained according to the training data set and a preset feature engineering strategy, and a trained feature engineering model is obtained.

[0078] In the embodiment, the training data set is obtained from the historical sample data extracted from the system database, the feature engineering strategy is the data information for constructing the feature engineering model pre-stored in the system, and the feature engineering strategy at least includes some of the sum operation, the normalization operation, the missing value filling operation, the type conversion operation, and the judgment of whether it is empty operation.

[0079] In step S30, a test data set is obtained, the test data set is input into the trained feature engineering model for feature extraction, key features are obtained, and a preset algorithm model is trained according to the key features, and a marketing prediction model is obtained.

[0080] In the embodiment, the test data set is obtained from the historical sample data extracted from the system database, and the preset algorithm model is a neural network model or a random forest model composed of multiple decision tree models.

[0081] In step S40, the prediction accuracy of the marketing prediction model is evaluated, a prediction accuracy score is obtained, and the prediction accuracy score is compared with a preset standard score.

[0082] In step S50, if the prediction accuracy score is greater than or equal to the standard score, the marketing prediction model is used as an optimized marketing prediction model.

[0083] In step S60, if the prediction accuracy score is less than the standard score, the model parameters of the marketing prediction model and the feature engineering strategy are iteratively adjusted until the prediction accuracy score is greater than or equal to the standard score, and after the prediction accuracy score is greater than or equal to the standard score, the adjusted marketing prediction model is used as the optimized marketing prediction model.

[0084] In step S70, a marketing prediction data is obtained by performing a marketing pre-rehearsal according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy.

[0085] In the embodiment, the marketing strategy is used for placement in various financial marketing scenarios, such as product marketing, advertising marketing, business marketing, etc., and the preset standard score and marketing indicators can be set according to actual conditions.

[0086] Step S80, comparing the marketing prediction data with the preset marketing index to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system.

[0087] In this embodiment, the standard sample data after preprocessing is analyzed, and a user portrait is generated according to the analysis result, so that the behavior habits and preferences of the user can be obtained according to the user portrait, and the corresponding marketing strategy can be formulated for the user class group formed after the user data is classified. The feature engineering model is constructed and trained according to the training data set and the preset feature engineering strategy, so that the key features that can better train the algorithm model can be obtained through the trained feature engineering model. After obtaining the key features of the test data set, the preset algorithm model is trained, so that a marketing prediction model that can accurately predict the marketing result of the marketing strategy is obtained. The prediction accuracy of the marketing prediction model is evaluated, and the evaluation score is compared with the standard score. According to the comparison result, the model parameters of the marketing prediction model and the feature engineering strategy corresponding to the feature engineering model are adjusted, so that an optimized marketing prediction model that meets the accurate prediction requirement after adjustment and optimization is obtained. Then, according to the standard sample data, the user portrait, the optimized marketing prediction model and the preset marketing strategy, a marketing pre-rehearsal is performed to obtain marketing prediction data, and the marketing prediction data and the preset marketing index are compared. According to the comparison result, the marketing strategy is adjusted, and the optimized marketing system that can effectively improve the marketing prediction effect is generated according to the adjusted optimized marketing strategy and the optimized marketing prediction model. This embodiment can be applied to the optimization of the marketing system of the financial marketing activity, and can effectively improve the marketing prediction effect of the marketing system, so as to improve the conversion rate of the user.

[0088] With reference to Figure 3 In some optional implementations of this embodiment, step S10 includes the following steps:

[0089] Step S101, obtaining historical sample data, and performing data cleaning and data screening on the historical sample data to obtain the standard sample data;

[0090] In the embodiment, the historical sample data refers to user data obtained after a marketing activity has been performed on the user, which can be user data collected in an offline marketing activity, for example, the number of users participating, the time period of user participation, user feedback experience, user favorite items, etc., or can be user data collected in an online marketing activity, for example, the number of user clicks, the time of user staying on an activity page, the depth of user activity page, the number of user clicks, etc. Data cleaning includes data deduplication, data sampling, data normalization processing, etc., and data screening includes data completeness judgment, data anomaly processing, etc. By performing data cleaning and data screening on the historical sample data, standard sample data that can be conveniently processed by the system is obtained.

[0091] Step S102, user data extraction is performed on the standard sample data to obtain user attribute and user behavior data, and the user attribute and the user behavior data are associated to obtain user association information; and

[0092] In the embodiment, the user attribute refers to basic information of the user, for example, the gender of the user, the age of the user, the marital status of the user, the consumption level of the user, etc., and the user behavior data refers to the behavior of the user in the face of a marketing item in a marketing activity, for example, the participation intention of the user, the participation completion degree of the user, the experience feedback of the user, etc. By associating the user attribute and the user behavior data, the user group can be distinguished according to the behavior habit of the user, so that the system can face diversified user data.

[0093] Step S103, the user association information is arranged and sorted, and a drawing operation is performed according to the user association information after the arrangement and sorting to obtain corresponding user portraits.

[0094] In the embodiment, the user association information is displayed on the user portrait, including the user attribute and the user behavior data, wherein the user data and the user behavior data can be displayed in a hierarchical distribution form, for example, the user age is 20-25 years old, and the associated user behavior data is online marketing-user attention item-project depth-project completion degree; offline marketing-user participation project-user participation degree-user conversion degree, etc. By drawing the user portrait, the user can be effectively clustered and distinguished according to the user portrait, so as to facilitate the formulation of corresponding marketing strategies according to different user groups.

[0095] With reference to Figure 4 In some optional implementation manners of the embodiment, step S20 includes the following steps:

[0096] Step S201, obtaining the preset feature engineering strategy and analyzing the preset feature engineering strategy to obtain data processing information;

[0097] In the embodiment, the data processing information refers to step information on how the data should be processed, which corresponds to the number and content of the feature engineering items involved in the feature engineering strategy. The feature engineering strategy includes summation operation, normalization operation, type conversion operation, and judgment of whether to be null operation, and the data processing information corresponds to the summation step, normalization step, type conversion step, and null judgment step. The above feature engineering strategy content and data processing information content can be adjusted according to actual conditions.

[0098] In step S202, feature extraction is performed on the training data set to obtain data features, and variable selection is performed on the data features to obtain effective features.

[0099] In the embodiment, the variable selection includes statistical feature processing, time series feature processing, and user behavior feature processing. Variable selection on data features can effectively obtain more accurate features.

[0100] In step S203, a target processing step corresponding to the effective features is obtained from the data processing information, and the feature engineering model is constructed and trained according to the effective features and the target processing step.

[0101] The embodiment can effectively extract the feature that has the greatest impact on the marketing strategy from the data by constructing and training the feature engineering model, and train the marketing prediction model with the feature to improve the prediction accuracy of the marketing prediction model.

[0102] Continuing to refer to Figure 5 In some optional implementations of the embodiment, step S40 includes the following steps:

[0103] In step S401, a marketing prediction result predicted by the marketing prediction model is obtained, and a prediction accuracy is calculated according to the marketing prediction result.

[0104] In the embodiment, the prediction accuracy is calculated from the correct prediction data in the marketing prediction result and the input test data set. For example, if the input test data set is 200 and the correct prediction data in the marketing prediction result is 170, the prediction accuracy is 85%. By calculating the prediction accuracy, an effective basis for determining the prediction accuracy of the model can be provided.

[0105] In step S402, the prediction accuracy is evaluated according to the prediction accuracy to obtain a prediction accuracy score.

[0106] In the embodiment, the evaluation of the prediction accuracy is obtained by scoring the prediction accuracy of the model according to the prediction accuracy rate. The scoring criteria can be set for scoring, or a relationship mapping table of the prediction accuracy rate and the prediction accuracy can be set for fast and effective scoring. In order to ensure the reliability of the model prediction accuracy score, as many as possible prediction accuracy rates of several sets of marketing prediction results should be referred to, for example, three sets of prediction accuracy rates are 80%, 90% and 90% respectively, and the average prediction accuracy rate 86.7% can be taken, so that the corresponding prediction accuracy score can be taken as 86.7. According to the prediction rate, the number of test data sets can be scored and the score weight can be allocated, for example, the first set of prediction accuracy rate sample quantity is the most, so that the score weight of the first set can be set as 0.35, the second set of prediction accuracy rate sample quantity is moderate, so that the score weight of the second set can be set as 0.33, and the third set of prediction accuracy rate sample quantity is the least, so that the score weight of the third set can be set as 0.32, and the corresponding prediction accuracy score is 80*0.35+90*0.33+90*0.32=86.5.

[0107] In step S403, the preset standard score is obtained, and the prediction accuracy score is compared with the preset standard score.

[0108] In the embodiment, the standard score is preset as 90 points. When the prediction accuracy score of the model is greater than or equal to 90 points, it indicates that the prediction accuracy of the model has reached the standard, and the corresponding marketing prediction model can be used as the optimized marketing prediction model. When the prediction accuracy score of the model is less than 90 points, the model parameters of the marketing prediction model and the feature engineering strategy corresponding to the feature engineering model are iteratively adjusted until the iteration condition that the prediction accuracy score of the marketing prediction model is greater than or equal to 90 points is met. The above standard score can be adjusted according to actual conditions.

[0109] With reference to Figure 6 In some optional implementation of the embodiment, step S70 includes the following steps:

[0110] In step S701, the standard sample data is input into the optimized marketing prediction model for prediction to obtain an optimized prediction result.

[0111] In step S702, a user identifier corresponding to the optimized prediction result is obtained.

[0112] In the embodiment, the user identifier is identification information saved in the historical sample data for distinguishing users.

[0113] In step S703, the user data is distinguished according to the user identifier, and the user data is classified into a user group corresponding to the user portrait.

[0114] In the embodiment, the user portrait contains a plurality of user groups to represent a certain type of user. When the user data is distinguished, the user data can be distinguished according to the coincidence and similarity of the user data and the user groups, so as to classify the user data well.

[0115] In step S704, the preset marketing strategy is analyzed to obtain rehearsal data, and a rehearsal step is extracted from the rehearsal data.

[0116] In the embodiment, the marketing strategy includes not only the rehearsal data but also marketing basic information, wherein the marketing basic information includes a strategy delivery platform, a strategy delivery time, strategy content, and the like.

[0117] In step S705, the user groups are executed with the rehearsal step to obtain the marketing prediction data.

[0118] With reference to Figure 7 In some optional implementations of the embodiment, step S80 includes the following steps.

[0119] In step S801, it is detected whether the marketing prediction data meets the preset marketing index.

[0120] In step S802, if the marketing prediction data meets the preset marketing index, the marketing strategy corresponding to the marketing prediction data is taken as the optimized marketing strategy, and the optimized marketing system is generated according to the optimized marketing strategy and the optimized marketing prediction model.

[0121] In step S803, if the marketing prediction data does not meet the marketing index, the marketing strategy is iteratively adjusted until the marketing prediction data meets the preset marketing index, and the adjusted marketing strategy is taken as the optimized marketing strategy. The optimized marketing system is generated according to the optimized marketing strategy and the optimized marketing prediction model.

[0122] In the embodiment, the marketing index is set to determine whether the marketing strategy needs to be optimized. The marketing index refers to the expected effect that the marketing strategy should finally achieve. For example, the marketing index is that at least 50 customers out of 200 customers should finally participate in the marketing activity, and 30 customers should participate in the marketing promotion project. By setting the marketing index, the marketing strategy can be flexibly controlled to be optimized to meet different marketing strategy requirements. Since the marketing strategy is distinguished according to the user portrait and there can be multiple marketing strategies, the marketing index corresponding to each marketing strategy is different. The above marketing index can be set and adjusted according to actual conditions.

[0123] In some optional implementations of the embodiment, after step S80, the following steps are further included.

[0124] detecting whether the number of the optimized marketing strategies is greater than one or equal to one;

[0125] In the embodiment, when the gap between the marketing prediction data of the current optimized marketing strategy and the marketing prediction data of the last optimized marketing strategy is too small, in order to ensure the optimal result of the marketing strategy, the two optimized marketing strategies are saved and compared through AB testing to determine one of the optimized marketing strategies as the optimal marketing strategy. Alternatively, a standard marketing strategy can be preset, and when the model determines the optimized marketing strategy, the optimized marketing strategy and the standard marketing strategy can be tested and compared to determine whether the optimized marketing strategy is the current optimal marketing strategy. When the above situation occurs, the standard marketing strategy can be counted as an optimized marketing strategy, and the number of the optimized marketing strategies is greater than one.

[0126] If the number of the optimized marketing strategies is greater than one, the optimized marketing strategies are subjected to AB testing, the optimal marketing strategy is selected according to the test result, and the test result is recorded to generate a visual log report; and

[0127] If the number of the optimized marketing strategies is equal to one, the optimized marketing strategy is subjected to a strategy delivery test, and the test result is recorded to generate a visual log report.

[0128] In the embodiment, by recording the test result and the test result, and generating a visual log report, the strategy delivery effect and the strategy return on investment can be evaluated and analyzed to adjust the corresponding indicators.

[0129] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by computer readable instructions instructing related hardware, and the computer readable instructions can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments. The storage medium can be a non-volatile storage medium such as a magnetic disc, an optical disc, a read-only memory (ROM), or a random access memory (RAM).

[0130] It should be understood that although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless explicitly stated herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be alternately executed with at least part of other steps or sub-steps or stages of other steps.

[0131] Further referring to Figure 8 , as an implementation of the method shown in the above Figure 1 , the present application provides an embodiment of a business artificial intelligence-based marketing system optimization device. The device embodiment corresponds to the method embodiment shown in Figure 1 , and the device can be specifically applied to various electronic devices.

[0132] As shown in Figure 8 , the artificial intelligence-based marketing system optimization device 900 described in the embodiment includes a data analysis module 901, a model construction module 902, a model training module 903, a model evaluation module 904, a first model output module 905, a second model output module 906, a strategy pre-rehearsal module 907, and a strategy optimization module 908. Among them:

[0133] The data analysis module 901 is configured to obtain preprocessed standard sample data, perform data analysis on the standard sample data to obtain an analysis result, and generate a user portrait according to the analysis result.

[0134] The model construction module 902 is configured to obtain a training data set, construct and train a feature engineering model according to the training data set and a preset feature engineering strategy, and obtain a trained feature engineering model.

[0135] The model training module 903 is configured to obtain a test data set, input the test data set into the trained feature engineering model for feature extraction to obtain key features, and train a preset algorithm model according to the key features to obtain a marketing prediction model.

[0136] The model evaluation module 904 is configured to evaluate the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score, and compare the prediction accuracy score with a preset standard score.

[0137] The first model output module 905 is configured to, if the prediction accuracy score is greater than or equal to the standard score, take the marketing prediction model as an optimized marketing prediction model.

[0138] The second model output module 906 is configured to iteratively adjust the model parameters of the marketing prediction model and the feature engineering strategy until the prediction accuracy score is greater than or equal to the standard score, and to take the corresponding marketing prediction model as the optimized marketing prediction model after the prediction accuracy score is greater than or equal to the standard score.

[0139] The strategy pre-performance module 907 is configured to perform marketing pre-performance according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy, to obtain marketing prediction data.

[0140] The strategy optimization module 908 is configured to compare the marketing prediction data with a preset marketing index to obtain a comparison result, and to adjust the marketing strategy according to the comparison result to obtain an optimized marketing system.

[0141] The above device can automatically optimize the marketing strategy and the marketing prediction model, and effectively improve the prediction accuracy of the marketing system for the marketing activities.

[0142] To solve the above technical problem, the embodiment of the present application further provides a computer device. For details, please refer to Figure 9 , Figure 9 The basic structure block diagram of the computer device of the embodiment is shown in the figure.

[0143] The computer device 2 includes a memory 21, a processor 22, and a network interface 23 which are connected to each other through a system bus. It should be noted that only the computer device 2 with components 21-23 is shown in the figure, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented. Among them, those skilled in the art can understand that the computer device here is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0144] The computer device can be a desktop computer, a notebook computer, a palm computer, a cloud server, and other computing devices. The computer device can interact with the user through a keyboard, a mouse, a remote controller, a touchpad, a voice control device, and other ways.

[0145] The memory 21 includes at least one type of readable storage medium, such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 21 can be an internal storage unit of the computer device 2, such as a hard disk or a memory of the computer device 2. In other embodiments, the memory 21 can also be an external storage device of the computer device 2, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 2. Of course, the memory 21 can also include both the internal storage unit and the external storage device of the computer device 2. In this embodiment, the memory 21 is generally used to store an operating system and various application software installed on the computer device 2, such as computer readable instructions of the marketing system optimization method based on artificial intelligence, etc. In addition, the memory 21 can also be used to temporarily store various data that have been output or will be output.

[0146] The processor 22 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 22 is generally used to control the overall operation of the computer device 2. In this embodiment, the processor 22 is used to run computer readable instructions or process data stored in the memory 21, such as running computer readable instructions of the marketing system optimization method based on artificial intelligence.

[0147] The network interface 23 can include a wireless network interface or a wired network interface, and is generally used to establish a communication connection between the computer device 2 and other electronic devices.

[0148] By using the above computer device, the embodiment can automatically optimize the marketing strategy and the marketing prediction model, and effectively improve the prediction accuracy of the marketing system for the marketing activities.

[0149] The present application also provides another embodiment, i.e., to provide a computer readable storage medium storing computer readable instructions, which can be executed by at least one processor to make the at least one processor execute the steps of the marketing system optimization method based on artificial intelligence as described above.

[0150] The embodiment can automatically optimize the marketing strategy and the marketing prediction model, and effectively improve the prediction accuracy of the marketing system on the marketing activities by using the computer readable storage medium.

[0151] Through the description of the above implementation, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better implementation. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the method described in each embodiment of the present application.

[0152] Obviously, the above-described embodiments are only some of the embodiments of the present application, not all the embodiments, and the preferred embodiments of the present application are given in the drawings, but do not limit the patent scope of the present application. The present application can be realized in many different forms, and conversely, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing specific embodiments, or make equivalent replacements to some technical features. Any equivalent structure made by using the content of the specification and drawings, directly or indirectly applied to other related technical fields, is also within the scope of the patent protection of the present application.

Claims

1. An artificial intelligence-based marketing system optimization method, characterized by, The method comprises the following steps: obtaining preprocessed standard sample data, performing data analysis on the standard sample data to obtain an analysis result, and generating a user portrait according to the analysis result; obtaining a training data set, constructing and training a feature engineering model according to the training data set and a preset feature engineering strategy, and obtaining a trained feature engineering model; obtaining a test data set, inputting the test data set into the trained feature engineering model for feature extraction to obtain key features, training a preset algorithm model according to the key features, and obtaining a marketing prediction model; evaluating the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score, and comparing the prediction accuracy score with a preset standard score; if the prediction accuracy score is greater than or equal to the standard score, the marketing prediction model is used as an optimized marketing prediction model; if the prediction accuracy score is less than the standard score, the model parameters of the marketing prediction model and the feature engineering strategy are iteratively adjusted until the prediction accuracy score is greater than or equal to the standard score, and after the prediction accuracy score is greater than or equal to the standard score, the adjusted marketing prediction model is used as the optimized marketing prediction model; performing marketing pre-rehearsal according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy to obtain marketing prediction data; and comparing the marketing prediction data with a preset marketing index to obtain a comparison result, adjusting the marketing strategy according to the comparison result, and obtaining an optimized marketing system; The step of performing marketing pre-rehearsal according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy to obtain marketing prediction data specifically comprises: inputting the standard sample data into the optimized marketing prediction model for prediction to obtain an optimized prediction result; obtaining a user identifier corresponding to the optimized prediction result; distinguishing user data according to the user identifier, and classifying the user data into a user group corresponding to the user portrait; analyzing the preset marketing strategy to obtain pre-rehearsal data, and extracting a pre-rehearsal step from the pre-rehearsal data; and executing the pre-rehearsal step on the user group to obtain the marketing prediction data.

2. The artificial intelligence-based marketing system optimization method of claim 1, wherein, The step of obtaining preprocessed standard sample data, performing data analysis on the standard sample data to obtain an analysis result, and generating a user portrait according to the analysis result specifically comprises: obtaining historical sample data, performing data cleaning and data screening on the historical sample data to obtain the standard sample data; extracting user data from the standard sample data to obtain user attributes and user behavior data, and performing data association on the user attributes and the user behavior data to obtain user association information; and arranging and organizing the user association information, and performing a drawing operation according to the arranged and organized user association information to obtain a corresponding user portrait. 3.The AI-based marketing system optimization method of claim 1, wherein, The step of constructing and training a feature engineering model according to the training data set and a preset feature engineering strategy specifically comprises: Obtaining the preset feature engineering strategy, and analyzing the preset feature engineering strategy to obtain data processing information; Performing feature extraction on the training data set to obtain data features, and performing variable selection on the data features to obtain effective features; and Obtaining a target processing step corresponding to the effective features in the data processing information, and constructing and training the feature engineering model according to the effective features and the target processing step. 4.The AI-based marketing system optimization method of claim 1, wherein, The step of evaluating the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score, and comparing the prediction accuracy score with a preset standard score, specifically includes: Obtaining a marketing prediction result predicted by the marketing prediction model, and calculating a prediction accuracy rate according to the marketing prediction result; Evaluating the prediction accuracy according to the prediction accuracy rate to obtain the prediction accuracy score; and Obtaining the preset standard score, and comparing the prediction accuracy score with the preset standard score. 5.The AI-based marketing system optimization method of claim 4, wherein, The step of comparing the marketing prediction data with a preset marketing indicator to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system, specifically includes: Detecting whether the marketing prediction data meets the preset marketing indicator; If the marketing prediction data meets the preset marketing indicator, taking the marketing strategy corresponding to the marketing prediction data as an optimized marketing strategy, and generating the optimized marketing system according to the optimized marketing strategy and the optimized marketing prediction model; and If the marketing prediction data does not meet the marketing indicator, iteratively adjusting the marketing strategy until the marketing prediction data meets the preset marketing indicator, taking the adjusted marketing strategy as the optimized marketing strategy, and generating the optimized marketing system according to the optimized marketing strategy and the optimized marketing prediction model. 6.The AI-based marketing system optimization method of claim 1, wherein, After the step of comparing the marketing prediction data with a preset marketing indicator to obtain a comparison result, and adjusting the marketing strategy according to the comparison result to obtain an optimized marketing system, the following steps are further included: Detecting whether the number of optimized marketing strategies is greater than one or equal to one; If the number of optimized marketing strategies is greater than one, performing AB testing on the optimized marketing strategies, selecting an optimal marketing strategy according to a test result, and recording the test result to generate a visual log report; and If the number of optimized marketing strategies is equal to one, performing a strategy delivery test on the optimized marketing strategy, and recording a test result to generate a visual log report. 7.A marketing system optimization apparatus based on artificial intelligence, characterized by, The data analysis module is configured to obtain preprocessed standard sample data, perform data analysis on the standard sample data to obtain an analysis result, and generate a user portrait according to the analysis result. The model construction module is configured to obtain a training data set, construct and train a feature engineering model according to the training data set and a preset feature engineering strategy, and obtain a trained feature engineering model. ​ The model training module is configured to obtain a test data set, input the test data set into the trained feature engineering model to perform feature extraction, obtain key features, and train a preset algorithm model according to the key features to obtain a marketing prediction model. The model evaluation module is configured to evaluate the prediction accuracy of the marketing prediction model to obtain a prediction accuracy score, and compare the prediction accuracy score with a preset standard score. The first model output module is configured to, if the prediction accuracy score is greater than or equal to the standard score, take the marketing prediction model as an optimized marketing prediction model. The second model output module is configured to, if the prediction accuracy score is less than the standard score, iteratively adjust the model parameters of the marketing prediction model and the feature engineering strategy until the prediction accuracy score is greater than or equal to the standard score, and take the corresponding marketing prediction model as the optimized marketing prediction model after the prediction accuracy score is greater than or equal to the standard score. The strategy pre-performance module is configured to perform marketing pre-performance according to the standard sample data, the user portrait, the optimized marketing prediction model, and a preset marketing strategy to obtain marketing prediction data. The strategy optimization module is configured to compare the marketing prediction data with a preset marketing index to obtain a comparison result, and adjust the marketing strategy according to the comparison result to obtain an optimized marketing system. The strategy pre-performance module is specifically configured to input the standard sample data into the optimized marketing prediction model to perform prediction and obtain an optimized prediction result, obtain a user identifier corresponding to the optimized prediction result, distinguish user data according to the user identifier, classify the user data into a user class group corresponding to the user portrait, analyze the preset marketing strategy to obtain pre-performance data, extract a pre-performance step from the pre-performance data, and perform the pre-performance step on the user class group to obtain the marketing prediction data. The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by the processor to implement the steps of the marketing system optimization method based on artificial intelligence.

8. A computer device, comprising: The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by the processor to implement the steps of the marketing system optimization method based on artificial intelligence.

9. A computer-readable storage medium, characterized in that, ​

Images