A Marketing Strategy Generation Method and System Based on Artificial Intelligence

Abstract

This application relates to the field of artificial intelligence technology, disclosing a marketing strategy generation method and system based on artificial intelligence. The method first obtains current and historical return information of the target user, along with information on the same products. Then, based on the return information, it determines the target user's expected demand index for pre-ordered goods. Next, it combines this with a purchase impulse model corresponding to the target user's profile to correct the expected demand index and obtain the actual demand intensity. When the actual demand intensity exceeds a threshold, it obtains candidate product information associated with the pre-ordered goods. Finally, based on the return reason and the purchase impulse model, it determines multiple target products and forms marketing strategy information including a marketing display order. Thus, this technical solution can accurately analyze user needs and correct demand assessments by combining purchase impulse factors, improving the accuracy of judging users' actual needs. This allows for the development of marketing strategies based on users' real needs, precise product recommendations, and enhanced marketing effectiveness.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, specifically to a marketing strategy generation method and system based on artificial intelligence. Background Technology

[0002] With the rapid development of e-commerce, precision marketing strategies have become crucial for improving user experience and business efficiency. Traditional marketing methods primarily rely on user behavior data or historical purchase records for product recommendations, but they overlook the demand signals hidden in user return behavior. Existing technologies suffer from the following problems: return data is only used for quality monitoring and is not transformed into demand insights; furthermore, marketing strategy formulation relies on empirical rules, failing to achieve personalized recommendations and resulting in low marketing efficiency. Summary of the Invention

[0003] The main objective of this invention is to provide a marketing strategy generation method and system based on artificial intelligence, aiming to solve the technical problem of low marketing efficiency in the prior art.

[0004] To achieve the above objectives, in a first aspect, this application provides an artificial intelligence-based marketing strategy generation method, applied to a marketing management system, the method comprising:

[0005] Obtain the target user's current return information and the historical return information associated with the current return information. The historical return information is the same as the product information in the current return information. Both the current return information and the historical return information carry return reason information.

[0006] The expected demand index of the target user is determined based on the target user's current return information and historical return information. The expected demand index is used to characterize the intensity of the target user's expected demand for pre-ordered goods.

[0007] Obtain a purchase impulse model corresponding to the user profile of the target user, and correct the expected demand index based on the purchase impulse model to obtain a target demand index, which is used to characterize the actual demand intensity of the target user for pre-ordered goods.

[0008] If the actual demand intensity is greater than the intensity threshold, obtain the product information of the candidate products associated with the pre-ordered products;

[0009] Based on the return reason information and the purchase impulse model, multiple target products are identified from the candidate products, and marketing strategy information is formed.

[0010] The marketing strategy information is used to market and display the multiple target products to the target users, wherein the marketing strategy information includes at least the marketing display order of the multiple target products.

[0011] In one possible implementation, determining the target user's expected demand index based on the target user's current return information and historical return information includes:

[0012] Based on the target user's current return information and historical return information, determine the target user's order frequency for pre-ordered goods and the similarity of return reasons;

[0013] The target user's order frequency for pre-ordered goods and the similarity of return reasons are input into the demand index prediction model to obtain the target user's expected demand index. The order frequency and the similarity of return reasons are positively correlated with the expected demand index.

[0014] In one possible implementation, determining the similarity of return reasons for pre-ordered goods by the target user based on the target user's current return information and historical return information includes:

[0015] The return reason information is processed into a text vector to obtain a blurred vector of the return reason;

[0016] Calculate the cosine value between every two virtual vectors of return reasons and sum them to obtain the cosine similarity representation value;

[0017] The cosine similarity value is input into a similarity regression model to obtain the similarity of return reasons, wherein the similarity regression model satisfies the following expression:

[0018] ;

[0019] in, The cosine similarity representation value, Let a be the virtual vector of the reason for return, and b be the similarity control values ​​of the reasons for return.

[0020] In one possible implementation, before obtaining the purchase impulse model corresponding to the user profile of the target user, the method further includes:

[0021] The user profile of the target user is obtained by inputting the user identity information and historical purchase information of the target user into the profile generation model.

[0022] Based on the user profile of the target user, a purchase impulse model of the target user is constructed. The purchase impulse model includes purchasing ability information and purchase impulse information. The purchasing ability information is used to characterize the target user's ability to pay, and the purchase impulse information is used to characterize the target user's immediate purchase intention.

[0023] In one possible implementation, the step of correcting the expected demand index based on the purchase impulse model to obtain the target demand index includes:

[0024] Obtain a first representation value corresponding to the purchasing power information and a second representation value corresponding to the purchasing impulse information. The first representation value is used to represent the size of the target user's purchasing power, and the second representation value is used to represent the strength of the target user's immediate purchasing intention.

[0025] A correction index for the expected demand index is determined based on the first and second characterization values.

[0026] The target demand index is obtained based on the expected demand index and the correction index.

[0027] In one possible implementation, determining the correction index for the expected demand index based on the first characterization value and the second characterization value includes:

[0028] The influence weights of the first and second characterization values ​​on the expected demand index are obtained respectively.

[0029] The first and second characterization values ​​are weighted and summed to obtain a correction index for the expected demand index;

[0030] The step of obtaining the target demand index based on the expected demand index and the correction index includes:

[0031] If the correction index is greater than or equal to the index threshold, the expected demand index is positively corrected.

[0032] If the correction index is less than the index threshold, the expected demand index is reversed.

[0033] In one possible implementation, the candidate products include similar alternatives to the pre-ordered product, and / or, the candidate products include dissimilar alternatives to the pre-ordered product, and the step of obtaining product information of the candidate products associated with the pre-ordered product includes:

[0034] Obtain first candidate products by acquiring similar alternative products to the pre-ordered products, and obtain second candidate products by acquiring non-similar alternative products to the pre-ordered products, wherein the number of first candidate products is greater than or equal to the number of second candidate products;

[0035] Extract the product information of the first candidate product and the second candidate product to obtain the product information of the candidate product associated with the pre-ordered product.

[0036] In one possible implementation, the step of determining multiple target products from candidate products and forming marketing strategy information based on the return reason information and the purchase impulse model includes:

[0037] Based on the return reason information, select the candidate target products from the candidate products that match the return reason;

[0038] Based on the purchase impulse model, the target product is obtained by selecting the product that matches the purchase impulse model from the candidate target products;

[0039] The marketing display order of the target product is determined based on the matching degree of the return reason and the matching degree of the purchase impulse model;

[0040] Marketing strategy information is generated based on the marketing display order of the target products, and the marketing strategy information is used to market and display the multiple target products to the target users.

[0041] In one possible implementation, the step of selecting a product from the candidate products that matches the return reason based on the return reason information to obtain the candidate target product includes:

[0042] The historical return reasons for candidate products are categorized into structured tags;

[0043] For each candidate product, the similarity ratio value is obtained by calculating the proportion of its historical returns that are the same as the current return reason tag of the target user;

[0044] Select candidate products with a similarity ratio value less than a preset ratio threshold as candidate target products;

[0045] The step of selecting a target product from the candidate target products that matches the purchase impulse model based on the purchase impulse model includes:

[0046] Extract the purchasing power information of target users from the purchasing power model to determine the price range of goods that users can afford;

[0047] Iterate through the candidate target products, obtain the price of each product, and include products that fall within the price range that the user can afford into the target product set.

[0048] Secondly, this application also provides a marketing management system, including: a memory and a processor, wherein the memory is used to store program code; and the processor is used to call the program code to execute the method described in the first aspect.

[0049] Unlike existing technologies, this application provides an AI-based marketing strategy generation method. First, it obtains historical return information for the target user and related products, both including return reasons. Then, based on this return information, it determines the target user's expected demand index for pre-ordered products. Next, it adjusts the expected demand index using a purchase impulse model corresponding to the target user's profile, resulting in a target demand index reflecting the actual demand intensity. When the actual demand intensity exceeds a threshold, it obtains candidate product information associated with the pre-ordered products. Then, based on the return reasons and the purchase impulse model, it identifies multiple target products and forms marketing strategy information including a marketing display order. Finally, it displays the target products to the target user according to this strategy. Thus, this technical solution can accurately analyze user needs and adjust demand assessments based on purchase impulse factors, improving the accuracy of judging actual user needs. This allows for the development of marketing strategies based on real user needs, precise product recommendations, enhanced marketing effectiveness, and increased user satisfaction and purchase conversion rates. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating the AI-based marketing strategy generation method in some embodiments of this application;

[0052] Figure 2 This is a flowchart illustrating step S200 of the marketing strategy generation method based on artificial intelligence in some other embodiments of this application;

[0053] Figure 3 This is a flowchart illustrating step S500 of the marketing strategy generation method based on artificial intelligence in some other embodiments of this application;

[0054] Figure 4 This is a schematic diagram of the hardware structure of the marketing management system in some embodiments of this application.

[0055] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0057] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0058] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the term "and / or" throughout the text includes three solutions; taking A and / or B as an example, it includes technical solution A, technical solution B, and a technical solution that simultaneously satisfies A and B. Furthermore, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of a person skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0059] With the rapid development of e-commerce, precision marketing strategies have become crucial for improving user experience and business efficiency. Traditional marketing methods primarily rely on user behavior data or historical purchase records for product recommendations, but they overlook the demand signals hidden in user return behavior. Existing technologies suffer from the following problems: return data is only used for quality monitoring and is not transformed into demand insights; furthermore, marketing strategy formulation relies on empirical rules, failing to achieve personalized recommendations and resulting in low marketing efficiency.

[0060] It is understandable that when a user makes a purchase and returns an item due to dissatisfaction, the platform system needs to accurately grasp the user's true needs to precisely recommend products, improve marketing effectiveness, and enhance user satisfaction and purchase conversion rates in order to attract the user to shop on the platform again. Based on this, this application provides an artificial intelligence-based marketing strategy generation method. Figures 1-3 As shown, the following explanation uses a marketing management system as an example to illustrate this AI-based marketing strategy generation method. It should be noted that although the flowchart shows a logical order, in some cases, the steps shown or described may be performed in a different order. Please refer to the appendix. Figure 1The method includes the following steps S100-S600:

[0061] Step S100: Obtain the target user's current return information and the historical return information associated with the current return information. The historical return information is the same as the product information in the current return information. Both the current return information and the historical return information carry return reason information.

[0062] Specifically, after a target user completes a current return, the system obtains the target user's current return information and historical return information associated with the current return information. The association can refer to the association of return orders with the same product name or the same product purpose, and both the current return information and the historical return information carry return reason information.

[0063] For example, when a user returns a purchased "water cup", the system iterates through all return information related to water cups in the historical return information, especially all return information related to water cups in the historical return information within the most recent period.

[0064] For example, if a user returns a "water cup" as the reason for the return, and the system can find information on three other water cup returns within the past month based on the current return information, with reasons such as "poor quality" and "size mismatch".

[0065] Step S200: Determine the expected demand index of the target user based on the target user's current return information and historical return information. The expected demand index is used to characterize the intensity of the target user's expected demand for pre-ordered goods.

[0066] It is understandable that a user's willingness to buy a certain product or the strength of their expected demand can be seen from return information. For example, when a user buys a certain product a lot and returns it frequently, it means that the user really wants to buy the product, but for various reasons, they are unable to find a product that meets their needs.

[0067] Therefore, in one embodiment, step S200: determining the expected demand index of the target user based on the target user's current return information and historical return information includes:

[0068] S210. Determine the target user's order frequency for pre-ordered goods and the similarity of return reasons based on the target user's current return information and historical return information;

[0069] S220. Input the target user's order frequency for pre-ordered goods and the similarity of return reasons into the demand index prediction model to obtain the target user's expected demand index, wherein the order frequency is positively correlated with the expected demand index, and the similarity of return reasons is positively correlated with the expected demand index.

[0070] Specifically, the system first counts the number of orders placed by target users for pre-ordered goods (or similar goods). Pre-ordered goods refer to items that users have currently or historically attempted to purchase but may return due to dissatisfaction. Order frequency reflects the user's level of interest or demand for these types of goods. If a user frequently orders the same type of goods, it indicates a strong willingness to purchase them. The system then analyzes the reasons for each return and calculates the similarity between these reasons. This can be achieved, for example, through text analysis techniques such as keyword matching, topic modeling (e.g., LDA), or semantic similarity calculation. The similarity of return reasons suggests that users have persistent dissatisfaction with certain aspects of the goods, which also reflects a continued demand for these types of goods, but a lack of a completely satisfactory option yet.

[0071] The demand index prediction model is a pre-trained machine learning model used to predict a user's expected demand index based on order frequency and the similarity of return reasons. The model can be trained on historical data, including a large amount of user purchasing behavior, return records, and corresponding demand intensity labels. In this embodiment, the demand index prediction model outputs the target user's expected demand index, a quantitative value representing the intensity of the user's expected demand for pre-ordered goods. In this embodiment, the model design ensures that order frequency and the similarity of return reasons are positively correlated with the expected demand index. This indicates that if a user places orders frequently and returns for similar reasons, the system will consider the user to have a high and sustained demand for that type of product.

[0072] In one embodiment, determining the similarity of return reasons for pre-ordered goods by the target user based on the target user's current return information and historical return information includes: performing text vectorization processing on the return reason information to obtain a blurred vector of return reasons; calculating the cosine value between every two blurred vectors of return reasons and summing them to obtain a cosine similarity representation value; inputting the cosine similarity representation value into a similarity regression model to obtain the similarity of return reasons, wherein the similarity regression model satisfies the following expression:

[0073] ;

[0074] in, The cosine similarity representation value, Let a be the virtual vector of the reason for return, and b be the similarity control values ​​of the reasons for return.

[0075] Specifically, one can first utilize word embedding techniques in Natural Language Processing (NLP), such as Word2Vec, GloVe, or Transformer-based BERT models, to convert the return reason information into vector form, obtaining a blurred vector of the return reason. After obtaining the blurred vector of the return reason, for every two vectors... and Using the cosine similarity formula Calculate the cosine value between them, where It is the vector dot product. and These are the magnitudes of the vectors. The cosine similarity values ​​are obtained by summing the cosines between all pairwise vectors. Finally, the cosine similarity value is input into the similarity regression model to obtain the similarity of the return reasons. Here, a and b are control values ​​for the magnitude of the similarity of the return reasons, which can be set according to actual needs.

[0076] For example, if all return reason information is vectorized to obtain a vector... , and At this time and cosine value, and cosine value and and The cosine values ​​are summed to obtain the cosine similarity representation value. Finally, the cosine similarity representation value is substituted into the similarity regression model to obtain the similarity of the reasons for return.

[0077] It is understandable that the text vectorization of the return reason information is processed to obtain the return reason dummy vector, and the cosine value between every two return reason dummy vectors is calculated and summed to obtain the cosine similarity representation value. If the cosine similarity representation value is closer to 0, it means that the return reasons are more similar. If the cosine similarity representation value is closer to 1, it means that the return reasons are less similar, that is, the greater the difference between the return reasons.

[0078] Step S300: Obtain the purchase impulse model corresponding to the user profile of the target user, and correct the expected demand index based on the purchase impulse model to obtain the target demand index. The target demand index is used to characterize the actual demand intensity of the target user for the pre-purchased goods.

[0079] It is understandable that the expected demand index obtained above only considers users' return information. Although the expected demand index can reflect users' willingness to purchase pre-ordered goods to a certain extent, it cannot truly reflect users' active and passive needs.

[0080] Therefore, after obtaining the expected demand index of the target users, it is necessary to acquire a purchase impulse model corresponding to the user profile of the target users, and then correct the expected demand index based on the purchase impulse model to obtain the target demand index. In this way, the target demand index integrates the return information of the target users and the purchase impulse model information, and can more accurately reflect the actual demand intensity of the target users for pre-ordered goods. This is very important for merchants because it can help them more accurately predict user demand, thereby formulating more effective sales strategies and inventory plans.

[0081] In one embodiment, before obtaining the purchase impulse model corresponding to the user profile of the target user, the method further includes: inputting the user identity information and historical purchase information of the target user into a profile generation model to obtain the user profile of the target user; constructing the purchase impulse model of the target user based on the user profile of the target user, wherein the purchase impulse model includes purchasing power information and purchase impulse information, the purchasing power information is used to characterize the target user's payment ability, and the purchase impulse information is used to characterize the target user's immediate purchase intention.

[0082] Specifically, the target user's identity information can include their age, gender, occupation, and geographical location. Historical purchase information can include the user's past purchase records, such as the type, quantity, price, purchase time, and time spent on each purchase. A pre-trained profile generation model is used to construct the target user's profile, which can generate the profile based on the user's identity information and historical purchase information. The profile generation model can employ machine learning or deep learning techniques, learning user characteristics and preferences by analyzing large amounts of user data. After obtaining the target user's profile, machine learning algorithms such as logistic regression, decision trees, random forests, and neural networks can be used to construct a purchase impulse model. During model training, a large amount of data containing user profiles and corresponding purchase behavior labels is required for supervised learning.

[0083] The purchase impulse model includes purchasing power information and purchase impulse information. Purchasing power information represents the target user's ability to pay, while purchase impulse information represents the target user's immediate willingness to purchase. For ease of calculation, both purchasing power and purchase impulse information can be quantified numerically. For example, purchasing power information can be quantified using a first representation value, and purchase impulse information can be quantified using a second representation value.

[0084] In one embodiment, the step of correcting the expected demand index based on the purchase impulse model to obtain the target demand index includes: obtaining a first representation value corresponding to the purchasing power information and a second representation value corresponding to the purchase impulse information, wherein the first representation value is used to represent the target user's purchasing power and the second representation value is used to represent the target user's immediate purchase intention intensity; determining a correction index for the expected demand index based on the first representation value and the second representation value; and obtaining the target demand index based on the expected demand index and the correction index.

[0085] Specifically, the process begins by obtaining a first representative value, a second representative value, and their respective weights on the expected demand index. Then, a weighted sum of the first and second representative values ​​is calculated to obtain a correction index for the expected demand index. This correction index integrates the user's purchasing power and immediate purchase intention, and is used to adjust the expected demand index. If the correction index is greater than or equal to the index threshold, it indicates that the user's purchasing power and immediate purchase intention are both strong; therefore, a positive correction is needed to the expected demand index, i.e., increasing the value of the expected demand index to reflect stronger actual demand. If the correction index is less than the index threshold, it indicates that the user's purchasing power or immediate purchase intention is weak; therefore, a negative correction is needed to the expected demand index, i.e., decreasing the value of the expected demand index to reflect weaker actual demand.

[0086] Step S400: If the actual demand intensity is greater than the intensity threshold, obtain the product information of the candidate products associated with the pre-ordered products;

[0087] After obtaining the target demand index, which represents the actual demand intensity of the target user for pre-ordered goods, if the actual demand intensity exceeds a threshold, it indicates that the user has a strong willingness to purchase the pre-ordered goods. Based on this information, the system can take a series of follow-up actions to optimize the user experience, improve sales conversion rates, or recommend related products. For example, it can first obtain product information of candidate products associated with the pre-ordered goods, and then select the product that best matches the target user from the candidate products as the target product, thereby accurately recommending the target product to the user.

[0088] In one embodiment, candidate goods may include similar alternatives to the pre-ordered goods. For example, if the pre-ordered goods are a single bed, candidate goods may include a single bed, a double bed, etc. Candidate goods may also include dissimilar alternatives to the pre-ordered goods. For example, if the pre-ordered goods are a single bed, candidate goods may include a folding sofa, etc., because a sofa can be used as a sleeping space in some cases.

[0089] The step of obtaining product information of candidate products associated with the pre-ordered product includes: obtaining first candidate products by obtaining similar alternative products of the pre-ordered product, and obtaining second candidate products by obtaining non-similar alternative products of the pre-ordered product, wherein the number of first candidate products is greater than or equal to the number of second candidate products; and extracting product information of the first candidate products and the second candidate products to obtain product information of candidate products associated with the pre-ordered product.

[0090] Specifically, the system uses search algorithms or database queries to find multiple alternative products similar to the pre-ordered product. The number of these alternatives is typically set to be greater than or equal to the number of dissimilar alternatives, as similar products are more likely to meet the user's direct needs. The system also uses search algorithms or database queries to find products that are dissimilar to the pre-ordered product but may be substitutes. The number of these alternatives may be relatively small because their substitutability may be weaker or more dependent on the user's personal preferences and specific needs. For each category of candidate products (whether similar or dissimilar), the system extracts its product information, which may include product name, price, images, specifications, user reviews, etc. This information is then integrated to form a set of product information for candidate products associated with the pre-ordered product, thus providing sufficient data support for the selection of the target product.

[0091] Step S500: Based on the return reason information and the purchase impulse model, identify multiple target products from the candidate products and form marketing strategy information;

[0092] In one embodiment, step S500: determining multiple target products from candidate products based on the return reason information and the purchase impulse model, and forming marketing strategy information, includes:

[0093] S510. Based on the return reason information, select the product that matches the return reason from the candidate products to obtain the candidate target product;

[0094] S520. Based on the purchase impulse model, select the product that matches the purchase impulse model from the candidate target products to obtain the target product;

[0095] S530. Determine the marketing display order of the target product based on the matching degree of the return reason and the matching degree of the purchase impulse model;

[0096] S540. Marketing strategy information is generated according to the marketing display order of the target products, and the marketing strategy information is used to market and display the multiple target products to the target users.

[0097] Specifically, the system first filters candidate products to identify those that do not conflict with the user's return reasons or can resolve the user's return issues, thus obtaining candidate target products. Then, from these candidate target products, further filtering is performed to select those that align with the user's purchase impulse model, resulting in target products. After obtaining the target products, the system calculates the return reason matching degree and purchase impulse model matching degree for each target product, and ranks the target products based on these matching degrees. Ranking can also consider other factors, such as the product's sales history, inventory status, and price competitiveness. In this way, the system can more accurately identify products that meet user needs and purchase impulses, and formulate corresponding marketing strategies, thereby improving marketing effectiveness and user satisfaction.

[0098] For example, if the target user's reason for returning a product is poor quality, then products with a low return rate due to quality issues can be selected from the candidate products as candidate target products. Then, from the candidate target products, products with comparable prices (within the target user's acceptable range) can be further selected as the final target products.

[0099] In one embodiment, the step of selecting a product that matches the return reason from the candidate products to obtain a candidate target product based on the return reason information includes: dividing the historical return reasons of the candidate products into structured tags; for each candidate product, calculating the proportion of its historical returns that are the same as the current return reason tag of the target user to obtain a similarity ratio value; and selecting candidate products with similarity ratio values ​​less than a preset ratio threshold as candidate target products.

[0100] Specifically, the system first collects historical return reasons for candidate products, which may come from multiple users and time periods. Then, the system performs text analysis on these return reasons, identifying common return causes and categorizing them into structured tags. These tags can include "size not right," "quality problem," "color mismatch," "dislike of style," etc. For each candidate product, the system iterates through its historical return records, counting the number of records where the return reason tags match the target user's current return reason tags. The system then calculates the proportion of this number to the total number of returns for that product, obtaining a similarity ratio value. This value reflects the degree of match between the product and the target user's current return reason. Finally, products with similarity ratio values ​​below a preset threshold are selected as candidate target products. These products are more likely to be accepted by users because they are unrelated to or have minimal conflict with the target user's return reason. Thus, by recommending products that do not conflict with the target return reason, the system can increase user interest and purchase intention for these products, thereby improving sales conversion rates.

[0101] In one embodiment, the step of selecting a product that matches the purchase impulse model from the candidate target products to obtain the target product includes: extracting the target user's purchasing power information from the purchasing power model to determine the price range of the product that the user can afford; traversing the candidate target products, obtaining the price of each product, and including products that fall within the price range that the user can afford into the target product set.

[0102] Specifically, the system first accesses the purchasing power model and extracts the target user's purchasing power information, typically represented by the price range of goods the user can afford. This price range reflects the user's economic constraints and budget limitations when purchasing goods. Then, the system iterates through the previously filtered list of candidate target goods, obtaining the current selling price for each product. The system then compares the price of each candidate target product with the user's affordable price range. If a product's price falls within the user's price range, the system includes that product in the target product set. This ensures that the selected target products not only do not conflict with the user's reason for returning the product but also align with the user's purchasing power and budget limitations.

[0103] Thus, by comprehensively considering the return reasons and purchasing power information of target users, this application can determine the target products, provide users with more personalized product recommendations, improve the accuracy of recommendations and user satisfaction, and thereby increase sales conversion rates.

[0104] The matching degree of return reasons and the matching degree of purchase impulse models can be the degree of matching between return reasons and purchasing power. For example, the less conflict there is between the historical returns of the target product and the user's return reasons, the higher the matching degree. The closer the price of the target product is to the user's purchasing power, the higher the matching degree. After obtaining the target products and their corresponding matching degrees, marketing display ranking of the target products can be obtained based on the corresponding matching degrees to obtain marketing strategy information.

[0105] Step S600: Present the multiple target products to the target user according to the marketing strategy information, wherein the marketing strategy information includes at least the marketing presentation order of the multiple target products.

[0106] Specifically, after obtaining marketing strategy information, target products can be displayed according to the marketing display ranking information of the target products in the marketing strategy information. That is, the target products with the highest relevance to return reasons and purchase impulse models are displayed first, thereby improving sales conversion rates. In addition, technologies such as payment systems, short video viral marketing, and NFC interaction can be integrated across three platforms and embedded in various payment hardware devices to reduce customer acquisition costs and improve sales conversion rates.

[0107] Based on this, this application provides an AI-based marketing strategy generation method. First, it obtains the target user's current and historical return information (including return reasons) for the same products. Then, it determines the target user's expected demand index for pre-ordered goods based on this return information. Next, it adjusts the expected demand index using a purchase impulse model corresponding to the target user's profile, resulting in a target demand index reflecting the actual demand intensity. When the actual demand intensity exceeds a threshold, it obtains candidate product information associated with the pre-ordered goods. Then, it determines multiple target products based on the return reasons and the purchase impulse model, forming marketing strategy information including a marketing display order. Finally, it displays the target products to the target user according to this strategy. Thus, this technical solution can accurately analyze user needs and adjust demand assessments based on purchase impulse factors, improving the accuracy of judging actual user needs. This allows for the development of marketing strategies based on real user needs, precise product recommendations, enhanced marketing effectiveness, and increased user satisfaction and purchase conversion rates.

[0108] like Figure 4 As shown, Figure 4 The diagram below illustrates the hardware structure of a marketing management system in some embodiments of this application. The marketing management system provided in this application includes a memory 1000 and a processor 2000. The memory 1000 is used to store computer-readable instructions, and the processor 2000 is used to call the computer-readable instructions to execute the marketing strategy generation method based on artificial intelligence as described above.

[0109] The processor 2000 provides computing and control capabilities to control the marketing management system to perform corresponding tasks. For example, it controls the marketing management system to execute the AI-based marketing strategy generation method in any of the above method embodiments. The method includes: acquiring current return information of a target user and historical return information associated with the current return information, wherein the historical return information is identical to the product information in the current return information, and both the current return information and the historical return information carry return reason information; determining the target user's expected demand index based on the target user's current return information and historical return information, wherein the expected demand index characterizes the target user's expected demand for pre-ordered goods. Intensity; Obtain a purchase impulse model corresponding to the user profile of the target user, and correct the expected demand index based on the purchase impulse model to obtain a target demand index, which is used to characterize the actual demand intensity of the target user for the pre-ordered goods; If the actual demand intensity is greater than the intensity threshold, obtain product information of candidate products associated with the pre-ordered goods; Determine multiple target products from the candidate products according to the return reason information and the purchase impulse model and form marketing strategy information; Market and display the multiple target products to the target user according to the marketing strategy information, wherein the marketing strategy information includes at least the marketing display order of the multiple target products.

[0110] The processor 2000 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The aforementioned PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0111] The memory 1000, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the AI-based marketing strategy generation method in the embodiments of this application. The processor 2000 can implement the AI-based marketing strategy generation method in any of the above method embodiments by running the non-transitory software programs, instructions, and modules stored in the memory 1000.

[0112] Specifically, memory 1000 may include volatile memory (VM), such as random access memory (RAM); memory 1000 may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), or other non-transitory solid-state storage devices; memory 1000 may also include combinations of the above types of memory.

[0113] In summary, the marketing management system of this application adopts the technical solution of any of the above-mentioned embodiments of the marketing strategy generation method based on artificial intelligence. Therefore, it has at least the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated here.

[0114] This application also provides a computer-readable storage medium, such as a memory including program code, which can be executed by a processor to complete the AI-based marketing strategy generation method described in the above embodiments. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CDROM), magnetic tape, floppy disk, or optical data storage device, etc.

[0115] This application also provides a computer program product comprising one or more lines of program code stored in a computer-readable storage medium. The processor of the early warning system reads the program code from the computer-readable storage medium and executes the program code to complete the steps of the artificial intelligence-based marketing strategy generation method provided in the above embodiments.

[0116] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program or program code related to hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0117] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software and a general-purpose hardware platform, or of course, using hardware. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0119] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A marketing strategy generation method based on artificial intelligence, applied to a marketing management system, characterized in that, The method includes: Obtain the target user's current return information and the historical return information associated with the current return information. The historical return information is the same as the product information in the current return information. Both the current return information and the historical return information carry return reason information. The expected demand index of the target user is determined based on the target user's current return information and historical return information. The expected demand index is used to characterize the intensity of the target user's expected demand for pre-ordered goods. Obtain a purchase impulse model corresponding to the user profile of the target user, and correct the expected demand index based on the purchase impulse model to obtain a target demand index, which is used to characterize the actual demand intensity of the target user for pre-ordered goods. If the actual demand intensity is greater than the intensity threshold, obtain the product information of the candidate products associated with the pre-ordered products; Based on the return reason information and the purchase impulse model, multiple target products are identified from the candidate products, and marketing strategy information is formed. The marketing strategy information is used to market and display the multiple target products to the target users, wherein the marketing strategy information includes at least the marketing display order of the multiple target products.

2. The marketing strategy generation method based on artificial intelligence as described in claim 1, characterized in that, The step of determining the expected demand index of the target user based on the target user's current return information and historical return information includes: Based on the target user's current return information and historical return information, determine the target user's order frequency for pre-ordered goods and the similarity of return reasons; The target user's order frequency for pre-ordered goods and the similarity of return reasons are input into the demand index prediction model to obtain the target user's expected demand index. The order frequency and the similarity of return reasons are positively correlated with the expected demand index.

3. The marketing strategy generation method based on artificial intelligence as described in claim 2, characterized in that, The step of determining the similarity of return reasons for pre-ordered goods by the target user based on the target user's current return information and historical return information includes: The return reason information is processed into a text vector to obtain a blurred vector of the return reason; Calculate the cosine value between every two virtual vectors of return reasons and sum them to obtain the cosine similarity representation value; The cosine similarity value is input into a similarity regression model to obtain the similarity of return reasons, wherein the similarity regression model satisfies the following expression: ； in, The cosine similarity representation value, Let a be the virtual vector of the reason for return, and b be the similarity control values ​​of the reasons for return.

4. The marketing strategy generation method based on artificial intelligence as described in claim 1, characterized in that, Before obtaining the purchase impulse model corresponding to the user profile of the target user, the method further includes: The user profile of the target user is obtained by inputting the user identity information and historical purchase information of the target user into the profile generation model. Based on the user profile of the target user, a purchase impulse model of the target user is constructed. The purchase impulse model includes purchasing ability information and purchase impulse information. The purchasing ability information is used to characterize the target user's ability to pay, and the purchase impulse information is used to characterize the target user's immediate purchase intention.

5. The marketing strategy generation method based on artificial intelligence as described in claim 4, characterized in that, The process of adjusting the expected demand index based on the purchase impulse model to obtain the target demand index includes: Obtain a first representation value corresponding to the purchasing power information and a second representation value corresponding to the purchasing impulse information. The first representation value is used to represent the size of the target user's purchasing power, and the second representation value is used to represent the strength of the target user's immediate purchasing intention. A correction index for the expected demand index is determined based on the first and second characterization values. The target demand index is obtained based on the expected demand index and the correction index.

6. The marketing strategy generation method based on artificial intelligence as described in claim 5, characterized in that, The step of determining the correction index for the expected demand index based on the first characterization value and the second characterization value includes: The influence weights of the first and second characterization values ​​on the expected demand index are obtained respectively. The first and second characterization values ​​are weighted and summed to obtain a correction index for the expected demand index; The step of obtaining the target demand index based on the expected demand index and the correction index includes: If the correction index is greater than or equal to the index threshold, the expected demand index is positively corrected. If the correction index is less than the index threshold, the expected demand index is reversed.

7. The marketing strategy generation method based on artificial intelligence as described in claim 1, characterized in that, The candidate products include similar alternatives to the pre-ordered products, and / or, the candidate products include dissimilar alternatives to the pre-ordered products. Obtaining product information for the candidate products associated with the pre-ordered products includes: Obtain first candidate products by acquiring similar alternative products to the pre-ordered products, and obtain second candidate products by acquiring non-similar alternative products to the pre-ordered products, wherein the number of first candidate products is greater than or equal to the number of second candidate products; Extract the product information of the first candidate product and the second candidate product to obtain the product information of the candidate product associated with the pre-ordered product.

8. The marketing strategy generation method based on artificial intelligence as described in claim 7, characterized in that, The process of identifying multiple target products from candidate products and forming marketing strategy information based on the return reason information and the purchase impulse model includes: Based on the return reason information, select the candidate target products from the candidate products that match the return reason; Based on the purchase impulse model, the target product is obtained by selecting the product that matches the purchase impulse model from the candidate target products; The marketing display order of the target product is determined based on the matching degree of the return reason and the matching degree of the purchase impulse model; Marketing strategy information is generated based on the marketing display order of the target products, and the marketing strategy information is used to market and display the multiple target products to the target users.

9. The marketing strategy generation method based on artificial intelligence as described in claim 8, characterized in that, The step of selecting a product from the candidate products that matches the return reason information to obtain the candidate target product includes: The historical return reasons for candidate products are categorized into structured tags; For each candidate product, the similarity ratio value is obtained by calculating the proportion of its historical returns that are the same as the current return reason tag of the target user; Select candidate products with a similarity ratio value less than a preset ratio threshold as candidate target products; The step of selecting a target product from the candidate target products that matches the purchase impulse model based on the purchase impulse model includes: Extract the purchasing power information of target users from the purchasing power model to determine the price range of goods that users can afford; Iterate through the candidate target products, obtain the price of each product, and include products that fall within the price range that the user can afford into the target product set.

10. A marketing management system, characterized in that, include: Memory and processor, wherein the memory is used to store program code; The processor is used to call the program code to perform the method as described in any one of claims 1 to 9.

Images