Marketing scene intelligent training management method and device and storage medium

Abstract

The invention provides a marketing scene intelligent training management method and device and a storage medium, and relates to the technical field of marketing scenes, and the method comprises the steps: determining a consumption scene of a user according to a consumption portrait and a corresponding consumption parameter; the corresponding marketing scene is determined according to the consumption scene of the user, the residence place of the user and the corresponding store distribution map, so that the accuracy of the marketing scene is ensured; and determining a consumption route of the user based on the current position of the user, the corresponding demand information and the current time, matching corresponding marketing information according to a plurality of consumption nodes and marketing scenes in the consumption route of the user, and sequentially presenting each piece of marketing information to the user. Therefore, the training set is determined according to the content of each piece of marketing information, the grade coefficient of each piece of marketing information and the consumption portrait of the user in the city, and the intelligent training of the marketing scene is triggered according to the training set and the marketing scene, so that the applicability of the marketing scene in each city is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of marketing scenarios, and particularly relates to a marketing scenario intelligent training management method, device, and storage medium. Background Art

[0002] With the development of technology, marketing scenarios have been gradually applied to life and are applicable to user consumption in various cities at various promotion levels. At this time, the user's place of residence and the corresponding store distribution map are introduced, and the user's place of residence and the corresponding store distribution map are interacted to output the corresponding marketing scenario. However, the user's consumption scenario is not compatible, which affects the accuracy of the marketing scenario and reduces the applicability of the marketing scenario in each city. Summary of the Invention

[0003] Based on this, it is necessary to provide a marketing scenario intelligent training management method, device, and storage medium for the above technical problems.

[0004] A marketing scenario intelligent training management method includes: collecting a set of consumption data of a user in different cities; determining the consumption portrait of the user in the city according to the set of consumption data, the user's life information, and the corresponding city; determining the consumption parameters of the user in the city based on the user's income level, the corresponding consumption time, and the user's preferences in the city; determining the consumption scenario of the user according to the consumption portrait and the corresponding consumption parameters, and determining the corresponding marketing scenario according to the user's consumption scenario, the user's place of residence, and the corresponding store distribution map; in the marketing scenario, determining the consumption route of the user based on the user's current location, the corresponding demand information, and the current time, and matching the corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and presenting each marketing information to the user in turn; determining the grade coefficient of each marketing information according to each marketing information, the user's viewing duration, and the user's click times, determining a training set according to the content of each marketing information, the grade coefficient of each marketing information, and the user's consumption portrait in the city, and triggering the intelligent training of the marketing scenario according to the training set and the marketing scenario.

[0005] A marketing scenario intelligent training management device is applied to the above marketing scenario intelligent training management method. The marketing scenario intelligent training management device includes: A collection module for collecting a set of consumption data of a user in different cities; A consumption portrait module for determining the consumption portrait of the user in the city according to the set of consumption data, the user's life information, and the corresponding city; A consumption parameter module for determining the consumption parameters of the user in the city based on the user's income level, the corresponding consumption time, and the user's preferences in the city; A marketing scenario module, which is used to determine the consumption scenario of a user according to the consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario according to the user's consumption scenario, the user's place of residence, and the corresponding store distribution map; A marketing information module, which is used to determine the user's consumption route based on the user's current location, the corresponding demand information, and the current time in the marketing scenario, and match the corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and each marketing information is presented to the user in turn; An intelligent training module, which is used to determine the grade coefficient of each marketing information according to each marketing information, the user's viewing duration, and the user's click times, determine a training set according to the content of each marketing information, the grade coefficient of each marketing information, and the user's consumption portrait in the city, and trigger the intelligent training of the marketing scenario according to the training set and the marketing scenario.

[0006] A storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned marketing scenario intelligent training management method are realized.

[0007] For the above-mentioned marketing scenario intelligent training management method, device and storage medium, a consumption data set of the user in different cities is collected; the consumption portrait of the user in the city is determined according to the consumption data set, the user's living information, and the corresponding city; the consumption parameters of the user in the city are determined based on the user's income level, the corresponding consumption time, and the user's preferences in the city; the consumption scenario of the user is determined according to the consumption portrait and the corresponding consumption parameters, and the corresponding marketing scenario is determined according to the user's consumption scenario, the user's place of residence, and the corresponding store distribution map, which comprehensively considers the user's consumption scenario, the user's place of residence, and the corresponding store distribution map, and ensures the accuracy of the marketing scenario.

[0008] Furthermore, in the marketing scenario, the user's consumption route is determined based on the user's current location, the corresponding demand information, and the current time, and the corresponding marketing information is matched according to multiple consumption nodes in the user's consumption route and the marketing scenario, and each marketing information is presented to the user in turn, so as to present various aspects of marketing information according to the consumption route, facilitating subsequent control of each marketing information.

[0009] Therefore, the grade coefficient of each marketing information is determined according to each marketing information, the user's viewing duration, and the user's click times, a training set is determined according to the content of each marketing information, the grade coefficient of each marketing information, and the user's consumption portrait in the city, and the intelligent training of the marketing scenario is triggered according to the training set and the marketing scenario, so as to comprehensively consider the user in each city and ensure the applicability of the marketing scenario in each city. Brief Description of the Drawings

[0010] Figure 1 It is a schematic diagram of the application scenario of the intelligent training management method for the marketing scenario in an embodiment; Figure 2 It is a schematic flowchart of the intelligent training management method for the marketing scenario in an embodiment; Figure 3 It is a structural block diagram of the intelligent training management device for the marketing scenario in an embodiment; Figure 4 It is the internal structure diagram of the intelligent training management device for the marketing scenario in an embodiment. Detailed Description of the Embodiments

[0011] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. Embodiment 1

[0012] The intelligent training management method for the marketing scenario provided by the present application can be applied to the application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through the network. Among them, the terminal 102 can be, but is not limited to, various personal computers, servers, and marketing scenarios, and the server 104 can be implemented by an independent server or a server cluster composed of multiple servers. Embodiment 2

[0013] In this embodiment, please refer to Figures 2 to 4 , an intelligent training management method for the marketing scenario, which is applied to the intelligent training management scenario of the marketing scenario; the intelligent training management method for the marketing scenario includes: Step S11: Collect the consumption data set of users in different cities; Step S12: Determine the consumption portrait of the user in the city according to the consumption data set, the user's living information and the corresponding city; Step S13: Determine the consumption parameters of the user in the city based on the user's income level, the corresponding consumption time and the user's preferences in the city; Step S14: Determine the consumption scenario of the user according to the consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario according to the consumption scenario of the user, the place of residence of the user and the corresponding store distribution map; Step S15: In this marketing scenario, determine the user's consumption route based on the user's current location, corresponding demand information, and current time, and match corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and present each piece of marketing information to the user in turn; Step S16: Determine the grade coefficients of each piece of marketing information according to each piece of marketing information, the user's viewing duration, and the user's click times. Determine the training set according to the content of each piece of marketing information, the grade coefficients of each piece of marketing information, and the user's consumption portrait in this city, and trigger the intelligent training of the marketing scenario according to this training set and this marketing scenario.

[0014] In step S11, collect the consumption data sets of the user in different cities; In the specific implementation process of the present invention, the specific steps may be: S111: Mark the user's payment platform; S112: Export the user's past payment data according to the user's payment platform; S113: Determine the data sets of each city according to the matching of the user's past payment data and the corresponding payment location; S114: Determine the user's consumption data in different cities based on the data sets of each city and the user's consumption types.

[0015] In the embodiment of the present application, mark the user's payment platform; export the user's past payment data according to the user's payment platform, and further manage and control based on the user's payment platform, which ensures the export of the user's past payment data for subsequent management and control of the user's past payment data.

[0016] At this time, marking the user's payment platform is to distinguish which payment platform the user uses (such as Alipay, WeChat Pay, UnionPay, etc.) for subsequent data export and processing; usually, when the user registers or uses its service, the payment platform will require the user to perform identity verification and account binding; through this process, the payment platform can obtain the user's payment information and store it in the database; in order to mark the user's payment platform, a specific identifier or label can be assigned to the user in the database, and this identifier is associated with the user's payment platform.

[0017] Furthermore, exporting the user's past payment data is for analyzing the user's payment behavior, consumption habits, etc., providing a basis for subsequent decision-making or optimization; according to the payment platforms marked by the user in step S111, it is possible to connect to the corresponding payment platform databases, query and export the user's payment data; payment data usually includes transaction time, transaction amount, transaction type (such as transfer, payment, refund, etc.), information of the other party in the transaction, etc.; optionally, according to the user ID and payment platform name, locate the corresponding payment platform database; query all the payment records of this user in the database; export the query results into a specified file format (such as Excel, CSV, PDF, etc.) for subsequent analysis and processing.

[0018] Specifically, suppose there is a user A who uses both Alipay and WeChat Pay; when user A registers or uses the services of these two payment platforms, Alipay and WeChat Pay will respectively assign a unique user ID to user A in their respective databases and associate this ID with the user's payment information; in order to mark user A's payment platforms, a record can be added for user A in a unified database, and this record contains user A's name, the user IDs of the two payment platforms, and the corresponding payment platform names (such as "Alipay" and "WeChat Pay").

[0019] Suppose we need to export all the payment data of user A on Alipay and WeChat Pay; first, according to the payment platform names marked by user A in step S111 (i.e., "Alipay" and "WeChat Pay"), we connect to the databases of these two payment platforms respectively; then, query all the payment records of user A in the databases; finally, export the query results into an Excel file; in the Excel file, we can clearly see all the transaction records of user A on Alipay and WeChat Pay, including transaction time, transaction amount, transaction type, and information of the other party in the transaction, etc.

[0020] Therefore, determine the data sets of each city according to the matching of the user's past payment data and the corresponding payment locations; based on the data sets of each city and the user's consumption categories, determine the user's consumption data in different cities, taking into account the overall consideration of the data sets of each city and the user's consumption categories, ensuring the accuracy of the user's consumption data in different cities.

[0021] At this time, based on the user's payment data and payment location information, the payment data is classified into each city, thereby obtaining the data sets of each city. First, it is necessary to integrate the user's past payment data, which usually includes fields such as transaction time, transaction amount, transaction type, and transaction location. Then, the transaction location information in the payment data is parsed, which usually involves geocoding technology, that is, converting the address into longitude and latitude coordinates, or through address matching technology, matching the address with a preset list of cities. Finally, according to the parsed location information, the payment data is classified into the corresponding city to form the data sets of each city.

[0022] Meanwhile, based on the data sets of each city and the user's consumption category information, analyze the consumption habits and preferences of users in different cities. First, it is necessary to divide the user's consumption categories, which can be carried out according to dimensions such as the category of goods and the type of service. Then, filter out the payment data corresponding to the consumption category in the data sets of each city. Finally, analyze the filtered payment data, including calculating indicators such as consumption amount and consumption frequency, to reveal the consumption habits and preferences of users in different cities.

[0023] In step S12, determine the user's consumption portrait in the city according to the consumption data set, the user's life information, and the corresponding city. In the specific implementation process of the present invention, the specific steps can be as follows: S121: Obtain the consumption data set; S122: Collect the user's life information based on the traversal of the user's life database; S123: Associate the consumption data set, the user's life information, and the corresponding city; S124: Determine the first consumption feature according to the consumption data set and the user's life information; S125: Determine the second consumption feature according to the consumption data set and the corresponding city; S126: Determine the user's consumption portrait in the city according to the first consumption feature, the second consumption feature, and the duration of the user's stay in the city.

[0024] In the embodiment of the present application, obtaining the consumption data set and collecting the user's life information based on the traversal of the user's life database realizes the traversal of the user's life database and ensures the accuracy of the user's life information.

[0025] At this time, collect the consumption data of users at different times and locations, including but not limited to transaction amount, transaction time, transaction commodity / service type, transaction merchant, etc.; optionally, the purchase records of users on the payment platform, including commodity name, purchase time, price, payment method, etc.; the payment records of users, which record each transaction of users, including transaction time, transaction amount, transaction object, etc.; the consumption records of users in physical stores, which can be obtained through channels such as membership systems and POS machines; provide a consumption data set integrating multiple data sources to facilitate enterprises to quickly obtain comprehensive user consumption data.

[0026] Specifically, dock with third-party service providers such as payment platforms and payment platforms to access the API interface to obtain the consumption data of users in real time; scrape the consumption information of users from public websites; purchase the integrated consumption data set from third-party data providers.

[0027] At the same time, collect the life information of users such as basic information, hobbies, and social behaviors to understand the consumption motivation and behavior patterns of users more deeply; at this time, the personal information filled in by users when registering on the platform, such as name, age, gender, occupation, etc.; the behavior records of users on social media, including posts, comments, likes, people followed, etc.; the behavior records of users on the platform, such as browsing records, search records, click records, etc.; optionally, obtain relevant information by authorizing users to access their social media accounts or behavior logs; design questionnaires to collect the basic information and hobbies of users; use data mining technology to extract useful life information from user behavior logs.

[0028] Furthermore, associate the consumption data set, the life information of users, and the corresponding cities; determine the first consumption feature according to the consumption data set and the life information of users; determine the second consumption feature according to the consumption data set and the corresponding cities, which is compatible with the multi-dimensional control of the consumption data set, the life information of users, and the corresponding cities, and ensures the accuracy of the first consumption feature and the second consumption feature.

[0029] At this time, associate the consumption data and life information of users with specific cities to analyze the consumption habits and lifestyles of users in different cities; merge the consumption data set and the user life information database to ensure that the data of each user is complete; add the corresponding city labels to each user according to the geographical location information in the consumption records or life information of users; clean the integrated data to remove duplicate, incorrect, or invalid data to ensure the accuracy and reliability of the data.

[0030] Analyze the user's consumption habits, preferences and needs to form the first consumption characteristics; at this time, analyze the user's consumption data, including consumption amount, consumption frequency, consumption type (such as catering, shopping, entertainment, etc.), consumption brand, etc.; associate the user's consumption data with his or her life information, and analyze the impact of life information on consumption habits; extract the user's consumption characteristics based on the analysis results, such as consumption capacity, consumption preferences, brand loyalty, etc.

[0031] Analyze the differences and characteristics of users' consumption in different cities to form the second consumption characteristics. At this time, analyze the consumption data of users in different cities, including consumption amount, consumption type, consumption frequency, etc. Compare the consumption data of users in different cities to find out consumption trends and differences. Extract the consumption characteristics of users in specific cities based on the analysis results, such as urban consumption hotspots, differences in consumption habits, etc.

[0032] Specifically, suppose a payment platform wants to analyze the consumption differences of users in different cities; first, the platform integrates the user's consumption data set and the user's life information database; then, based on the delivery address information in the user's purchase record, the corresponding city label is added to each user; for example, user A's purchase record shows that his delivery address is mainly in Beijing, so the city label "Beijing" is added to user A; after data cleaning, the payment platform obtains a complete data set containing user consumption data, life information and city labels.

[0033] By analyzing the consumption data of user A, the payment platform found that user A spent a lot of money on shopping and mainly bought high-end brand clothing and electronic products. At the same time, combined with user A's life information, the payment platform learned that user A is a 35-year-old professional who pays attention to personal image and quality of life. Therefore, the payment platform extracted the first consumption characteristic of user A: high consumption ability, focus on quality, and preference for high-end brands.

[0034] The payment platform further analyzed the consumption data of user A in Beijing and Shanghai. Through comparison, it was found that user A's consumption in Beijing was mainly concentrated in high-end shopping malls and brand stores, and he purchased clothing, electronic products and luxury goods, etc.; while in Shanghai, user A preferred to visit fashionable and trendy streets and purchased fashionable clothing and accessories. Therefore, the payment platform extracted the second consumption characteristic of user A: in Beijing, he pays attention to quality and brand, and the amount of consumption is relatively high; in Shanghai, he pays more attention to fashion and trends, with a higher consumption frequency but a relatively low single consumption amount.

[0035] Therefore, the consumption portrait of the user in this city is determined based on the first consumption characteristic, the second consumption characteristic, and the length of stay of the user in this city, which incorporates the overall consideration of the first consumption characteristic, the second consumption characteristic, and the length of stay of the user in this city, realizes the multi-dimensional control of the first consumption characteristic, the second consumption characteristic, and the length of stay of the user in this city, and ensures the accuracy of the consumption portrait of the user in this city.

[0036] At this time, the first consumption characteristic (characteristics based on the user's personal life information and general consumption habits), the second consumption characteristic (the user's specific consumption habits in different cities), and the length of stay of the user in this city are integrated to comprehensively depict the user's consumption portrait in this city; this portrait helps enterprises to better understand the user's consumption behavior in this city, so as to formulate more accurate marketing strategies.

[0037] Integrate the first consumption characteristic, the second consumption characteristic, and the length of stay of the user in this city to form a comprehensive data set; assign appropriate weights to each characteristic according to the importance and relevance of the characteristics; for example, for users with a long stay, their second consumption characteristic has a higher weight; while for users with a short stay, their length of stay and the first consumption characteristic are more important; based on the integrated data set and weight assignment, use data analysis tools or algorithms to construct the user's consumption portrait in this city; the consumption portrait includes information such as the user's consumption preferences, consumption frequency, consumption amount, consumption hotspots, etc.; verify the accuracy of the consumption portrait by comparing it with the actual consumption data, and optimize it as needed.

[0038] Specifically, assume that a certain travel service platform hopes to build a consumption portrait for user B in the city he is about to visit - Chengdu; the first consumption characteristic of user B shows that he is a 30-year-old young man who loves food and culture and has a high pursuit of high-quality life experiences; the second consumption characteristic shows that when user B is in Beijing, he tends to dine in high-end restaurants, visit cultural attractions, and buy special souvenirs; while in Shanghai, he prefers to try street snacks and visit trendy neighborhoods; user B plans to stay in Chengdu for 5 days.

[0039] Based on this information, the travel service platform has constructed a consumption portrait of User B in Chengdu: Consumption Preferences: Loves food, especially Sichuan cuisine and local specialties; Has a strong interest in cultural attractions and museums; Prefers high-quality life experiences, such as high-end accommodation and customized travel services; Consumption Frequency: Due to the relatively long stay (5 days), it is expected that User B will taste local food multiple times, visit multiple cultural attractions, and purchase some special souvenirs; Consumption Amount: Given User B's pursuit of a high-quality life, it is expected that their consumption amount in Chengdu will be relatively high, especially in terms of dining and accommodation; Consumption Hotspots: User B will go to well-known food streets in Chengdu (such as Jinli, Kuanzhai Alleys, etc.) to taste local snacks; Visit cultural attractions such as Wuhou Shrine and Du Fu Thatched Cottage; And choose to stay in high-end hotels located in the city center or scenic areas.

[0040] Through this consumption portrait, the travel service platform can provide more personalized service recommendations for User B, such as recommending local restaurants that match their taste, customizing travel routes that match their interests, etc., thereby enhancing user satisfaction and the platform's competitiveness.

[0041] In another embodiment of the present application, an example of a consumption portrait matching table is used for illustration. Example of the consumption portrait matching table:

[0042] In this consumption portrait matching table, we have determined a description of the user's consumption portrait in the city based on the user's first consumption characteristics (such as food lover, shopping enthusiast, etc.), second consumption characteristics (such as likes to try local specialties, prefers high-end mall shopping, etc.), and the length of stay of the user in the city (such as 3 days, 7 days, etc.).

[0043] In addition, First Consumption Characteristic: 0.4 (because the first consumption characteristic reflects the user's long-term consumption habits and has a greater impact on the user's consumption behavior in the city); Second Consumption Characteristic: 0.3 (because the second consumption characteristic reflects the user's specific consumption habits in the city and has a certain impact on the user's consumption behavior in the city); Length of Stay: 0.3 (because the length of stay determines the nature and time of the user's consumption in the city); Suppose a user's first consumption characteristic is "food lover" (score: 8 points, full score: 10 points), the second consumption characteristic is "likes to try local specialties" (score: 7 points, full score: 10 points), and the length of stay is 5 days (score: 5 points, full score: 10 points, scored according to the length of the stay).

[0044] The consumption profile score of this user in Chengdu is: 0.48 + 0.37 + 0.3 * 5 = 3.2 + 2.1 + 1.5 = 6.8 points; According to the score, we can describe the user's consumption profile as: During the stay in Chengdu, this user tends to taste local special snacks and enjoy the pleasure brought by food. However, due to the low scores of the stay duration and some consumption characteristics, the user will not consume overly frequently or try too many different foods.

[0045] In step S13, determine the consumption parameters of the user in this city based on the user's income level, corresponding consumption time, and the user's preferences in this city. In the specific implementation process of the present invention, the specific steps can be: S131: Collect the user's income data by traversing the user's payment platforms. S132: Determine the user's income level based on the user's income data, the user's job title in the enterprise, and the user's place of residence. S133: Determine multiple consumption categories and corresponding consumption frequencies by traversing the user's consumption data in this city. S134: Determine the user's preferences in this city based on multiple consumption categories, corresponding consumption frequencies, and the city. S135: Interact the user's income level, corresponding consumption time, and the user's preferences in this city. S136: Determine the user's consumption parameters in this city based on the interaction of the user's income level, corresponding consumption time, and the user's preferences in this city.

[0046] In the embodiment of the present application, collect the user's income data by traversing the user's payment platforms; determine the user's income level based on the user's income data, the user's job title in the enterprise, and the user's place of residence, which incorporates the overall consideration of the user's income data, the user's job title in the enterprise, and the user's place of residence, ensuring the accuracy of the user's income level.

[0047] At this time, traverse the user's payment platforms. This step involves checking all payment platforms used by the user, including but not limited to Alipay, WeChat Pay, bank transfer applications, etc.; these platforms record all transaction activities of the user, including income; extract transaction records from these payment platforms, especially records related to the user's income; this includes salary income, bonuses, investment income, etc.; the extracted data needs to be integrated to form a comprehensive view of the user's income; this helps to analyze the user's income level and income sources.

[0048] Based on the income data collected from the payment platform, analyze the user's total income, income sources, and income stability; combined with the user's job title in the enterprise, the income level can be more accurately evaluated; different job titles often correspond to different salary ranges; the cost of living and the overall salary level in the place of residence are also important factors in determining the user's income level; the cost of living in first-tier cities is relatively high, and the salary level is also correspondingly high; while it is lower in second- or third-tier cities.

[0049] Suppose there is a user, Mr. Zhang, who uses Alipay and WeChat Pay as the main payment tools; by traversing these two platforms, we can collect the following income data of Mr. Zhang: Alipay: fixed monthly salary income, year-end bonus, investment income, etc.; WeChat Pay: occasional part-time income, red envelope income, etc. After integrating these data, we can obtain Mr. Zhang's total income situation. For example, his fixed monthly salary is 10,000 yuan, the year-end bonus is 30,000 yuan, the average monthly investment income is 2,000 yuan, and the part-time income and red envelope income are relatively random.

[0050] Suppose he is a project manager in a technology company and lives in Beijing; based on his income data and job title, we can conduct the following analysis: Analysis of income data: Mr. Zhang's total income includes fixed salary, year-end bonus, and investment income, with an average monthly total income of approximately 14,000 yuan (fixed salary + investment income, and the year-end bonus is amortized annually); Consideration of job title: As a project manager, Mr. Zhang's salary level is usually higher than that of ordinary employees; in a first-tier city like Beijing, the salary range for project managers is relatively wide, but combined with his total income situation, we can infer that he is at an above-average income level; Factor of place of residence: Beijing, as a first-tier city, has a relatively high cost of living; however, Mr. Zhang's total income is sufficient to support his life in Beijing and he has a certain savings ability.

[0051] Taking all the above factors into consideration, we can determine that Mr. Zhang's income level is above average; this kind of analysis helps the enterprise to more accurately understand the user's financial situation and consumption ability, so as to formulate more appropriate marketing strategies; At the same time, by traversing the consumption data of the user in the city, multiple consumption categories and the corresponding consumption frequencies are determined; based on multiple consumption categories, the corresponding consumption frequencies, and the city, the user's preferences in the city are determined. By introducing multiple consumption categories, the corresponding consumption frequencies, and the city, multi-dimensional control of the user's preferences in the city is achieved.

[0052] At this time, conduct a comprehensive inspection of all the user's consumption data in this city; this includes the user's transaction records on various payment platforms (such as Alipay and WeChat Pay), bank card transaction records, and transaction records on online and offline shopping platforms involved; identify different types of consumption from the consumption data, such as dining, transportation, entertainment, shopping, housing (rent or mortgage), education, medical care, etc.; for each type of consumption, calculate the user's consumption frequency; this can be calculated on a daily, weekly, monthly, or annual basis, depending on the availability of the data and the purpose of the analysis.

[0053] Based on the types of consumption identified in the previous step and the calculated consumption frequencies, analyze the user's spending and preferences in various types of consumption; combined with the consumption environment and cultural characteristics of the city where the user is located, further determine the user's consumption preferences; for example, some cities are famous for their cuisine, and users consume at restaurants in this city particularly frequently; while other cities are known for their rich shopping or entertainment activities; comprehensively considering the types of consumption, frequencies, and city characteristics, determine the user's preferences in this city; this helps enterprises understand the user's consumption tendencies and provides a basis for offering personalized services and products.

[0054] Specifically, suppose there is a user, Mr. Zhang, who has recently moved to Shanghai to live; in order to understand her consumption habits in Shanghai, we have traversed all her consumption data in Shanghai.

[0055] Dining consumption: Mr. Zhang dines at a restaurant near his company 5 times a week on weekdays, and on weekends he will try different restaurants or order takeout. The average weekly dining consumption frequency is 10 times; Transportation consumption: She takes the subway to and from work every day, and the frequency of recharging the transportation card is 4 times a month (once a week), and occasionally she also uses Didi Chuxing, but the frequency is relatively low; Entertainment consumption: Mr. Zhang goes to the movies 1 - 2 times a month and participates in friend gatherings or social activities 2 - 3 times a month; Shopping consumption: She consumes relatively frequently on online shopping platforms, and there are records of buying daily necessities or clothes several times a week, but the frequency of large - scale shopping (such as electronic products and furniture) is relatively low.

[0056] Combined with her types of consumption and frequencies in Shanghai, as well as the consumption environment and cultural characteristics of Shanghai, we can draw the following conclusions: Dining preference: Mr. Zhang has a relatively high consumption frequency for dining and a strong willingness to try different restaurants; considering the rich food culture in Shanghai, we can infer that she has a strong interest in the food culture in Shanghai.

[0057] Entertainment preference: Although Mr. Zhang's frequency of watching movies and participating in social activities is not particularly high, there is a fixed entertainment consumption every month, indicating that she enjoys social and leisure activities; as an international metropolis, Shanghai offers rich entertainment options, and Mr. Zhang will try different entertainment methods.

[0058] Shopping Preferences: Mr. Zhang has a high frequency of online shopping, indicating that she likes convenient shopping methods. At the same time, in Shanghai, a fashion capital with a dazzling array of shopping venues, although her frequency of large - value shopping is not high, she pays attention to fashion trends and new product releases.

[0059] Therefore, the income level of the user, the corresponding consumption time, and the user's preferences in the city are interacted; the consumption parameters of the user in the city are determined based on the interaction of the income level of the user, the corresponding consumption time, and the user's preferences in the city, realizing the interaction of the income level of the user, the corresponding consumption time, and the user's preferences in the city, ensuring the overall consideration of the income level of the user, the corresponding consumption time, and the user's preferences in the city, and achieving the precise control of the consumption parameters of the user in the city.

[0060] In step S14, the consumption scenario of the user is determined according to the consumption portrait and the corresponding consumption parameters, and the corresponding marketing scenario is determined according to the consumption scenario of the user, the place of residence of the user, and the corresponding store distribution map; In the specific implementation process of the present invention, the specific steps can be: S141: Obtain the consumption portrait and the corresponding consumption parameters; S142: Match the corresponding scenario weights to the consumption portrait and the corresponding consumption parameters; S143: Determine the consumption scenario of the user according to the consumption portrait, consumption parameters, and the corresponding scenario weights; S144: Determine the place of residence of the user according to the residential data of the user; S145: Determine the corresponding store distribution map based on the place of residence of the user, the surrounding drawings of the place of residence, and the store markings; S146: Determine the corresponding marketing scenario according to the consumption scenario of the user, the place of residence of the user, and the corresponding store distribution map.

[0061] In the embodiment of the present application, the consumption portrait and the corresponding consumption parameters are obtained; the corresponding scenario weights are matched to the consumption portrait and the corresponding consumption parameters; the consumption scenario of the user is determined according to the consumption portrait, consumption parameters, and the corresponding scenario weights. The consumption portrait, consumption parameters, and the corresponding scenario weights are introduced, and the overall control of the consumption portrait, consumption parameters, and the corresponding scenario weights is carried out, ensuring the accuracy of the consumption scenario of the user.

[0062] At this time, deeply understand the user's consumption portrait and consumption parameters, and match this information with different consumption scenarios, so as to assign a weight to each scenario; this weight reflects the likelihood or preference degree of the user's consumption in different scenarios; First, we need to review the user's consumption portrait, which usually includes information such as the user's age, gender, income level, occupation, hobbies, consumption habits, etc.; Then, we interpret the specific parameters related to the user's consumption, such as consumption frequency, average consumption amount, consumption hotspots (i.e., the fields or goods that the user consumes most frequently), consumption elasticity (i.e., the sensitivity of consumption to changes in income level), etc.; According to the user's consumption portrait and consumption parameters, we match the user with different consumption scenarios; These scenarios include online shopping, offline shopping (such as supermarkets, shopping malls, specialty stores, etc.), leisure and entertainment (such as cinemas, gyms, spas, etc.), travel, etc.; Finally, we assign a weight to each matched consumption scenario; The assignment of weights can be based on various factors such as the user's historical consumption data, consumption preferences, market trends, and competitive environment in this scenario; The weight value is usually a number between 0 and 1, indicating the likelihood or preference degree of the user's consumption in this scenario.

[0063] Combined with the user's consumption portrait, consumption parameters, and the previously assigned scenario weights, we will determine the user's most likely consumption scenario; This helps us understand the user's consumption behavior in different situations more accurately and provides a basis for formulating subsequent marketing strategies; First, we summarize the information such as the user's consumption portrait, consumption parameters, and scenario weights; Then, we analyze the weight values of each scenario and find the scenario with the highest weight; This scenario is most likely to become the user's main consumption scenario; When determining the main consumption scenario, we also need to consider other factors, such as the user's time arrangement, geographical location restrictions, etc.; These factors will affect the user's actual consumption behavior in different scenarios; Finally, we comprehensively consider all factors to determine the user's most likely consumption scenario.

[0064] Specifically, suppose there is a user, Mr. Li. His consumption portrait shows that he is a middle-aged, high-income male who likes outdoor activities and high-end shopping; His consumption parameters include an average monthly consumption amount of 5000 yuan in high-end shopping malls, an average monthly consumption amount of 2000 yuan on online shopping platforms, and he has a low sensitivity to prices.

[0065] Based on this information, we can match Mr. Li with the following consumption scenarios and assign corresponding weights: High-end mall shopping: weight is 0.6; because Mr. Li likes to shop in high-end malls and has a relatively high consumption amount in this scenario; Online shopping: weight is 0.3; although Mr. Li also shops online, his online consumption amount is lower compared to high-end malls; Outdoor gear store shopping: weight is 0.1; since Mr. Li likes outdoor activities, he will occasionally shop in outdoor gear stores, but this is not his main consumption scenario.

[0066] Based on his consumption profile, consumption parameters, and scenario weights, we can determine that his main consumption scenario is high-end mall shopping; this is because his average monthly consumption amount in high-end malls is relatively high, and the weight value of this scenario is 0.6, which is the highest among all scenarios; at the same time, we also need to note that although Mr. Li also consumes on online shopping platforms and the weight value is 0.3, his online consumption amount is still lower compared to high-end malls; therefore, when formulating marketing strategies, we can pay more attention to this scenario of high-end malls and provide Mr. Li with personalized shopping experiences and promotional activities.

[0067] Furthermore, determine the user's place of residence based on the user's residential data; based on the user's place of residence, the surrounding maps of the place of residence, and store markings, determine the corresponding store distribution map. By introducing the user's place of residence, the surrounding maps of the place of residence, and store markings, and considering them as a whole, the accuracy of the store distribution map is ensured.

[0068] At this time, determine the user's precise place of residence based on the residential data provided by the user; this data includes specific information such as the user's home address, postal code, city, region, or street, etc.; by analyzing and processing this data, we can clarify the approximate geographical location where the user lives, which is crucial for subsequent market segmentation, marketing strategy formulation, and store selection.

[0069] First, we need to collect the user's residential data from multiple channels such as user registration information, purchase records, or questionnaires; then, clean the collected data to remove duplicate, incorrect, or incomplete information to ensure the accuracy and reliability of the data; then, use Geographic Information System (GIS) technology to convert the user's residential data into specific geographical location information, such as longitude and latitude coordinates; finally, based on the geolocation results and combined with administrative division data, determine the user's precise place of residence, such as city, region, street, etc.

[0070] Based on the user's place of residence, the geographical maps of the place of residence, and the marked store information, draw a distribution map of the stores around the user; this map will help us intuitively understand the business environment in the user's area, including key information such as the types, quantities, distribution locations of the stores, and their relative distances from the user's place of residence.

[0071] First, we need to collect the geographical maps of the user's place of residence, which usually includes geographical information such as streets, buildings, and public facilities. At the same time, we also need to collect the marked store information, such as the name, type, and address of the store. Then, integrate the collected geographical maps and store information to ensure the accuracy of their corresponding relationships. Next, use GIS technology or professional map-drawing software to draw the integrated data into a store distribution map. This map should clearly show the types, quantities, and distribution locations of the stores around the user's place of residence. Finally, optimize and adjust the distribution map as needed, such as adding labels, marking distances, or providing other relevant information to better meet the subsequent analysis and marketing needs.

[0072] Specifically, assume that the residential data of Ms. Wang includes her home address: "A certain community in Sanlitun Street, Chaoyang District, Beijing". Through data cleaning and geolocation, we can determine that Ms. Wang's place of residence is Sanlitun Street, Chaoyang District, Beijing. This information is crucial for subsequent analysis of Ms. Wang's consumption habits, formulating targeted marketing strategies, and selecting nearby stores for promotion.

[0073] After determining that her place of residence is Sanlitun Street, Chaoyang District, Beijing, we collected the geographical maps and marked store information of this area. By integrating these data, we drew a distribution map of the stores around Ms. Wang. This map shows the distribution of various types of stores in Sanlitun Street, such as high-end shopping malls, fashion brand stores, gourmet restaurants, and coffee shops. Through this map, we can intuitively understand the business environment in the area where Ms. Wang is located, providing strong support for subsequent marketing strategy formulation. For example, we can select nearby fashion brand stores or high-end shopping malls for targeted promotion and marketing based on Ms. Wang's consumption profile and preferences.

[0074] Therefore, determining the corresponding marketing scenario based on the user's consumption scenario, the user's place of residence, and the corresponding store distribution map, and multi-dimensionally controlling the user's consumption scenario, the user's place of residence, and the corresponding store distribution map ensure the accuracy of the marketing scenario.

[0075] The consumption scenario refers to the specific environment and situation when consumers purchase, use products or services. Different consumption scenarios will stimulate different needs of consumers, thus affecting their purchase decisions. Analyze the daily living habits, interests, and potential needs of the target user group. Combine the characteristics of the product or service to determine the applicable consumption scenario. Through market research, user interviews, etc., verify and optimize the positioning of the consumption scenario.

[0076] The user's place of residence is one of the important factors to consider when formulating a marketing strategy; by understanding the geographical distribution of users, enterprises can promote the market and allocate resources more targeted; collect and analyze the geographical location data of users, including cities, regions, streets, etc.; use tools such as Geographic Information System (GIS) to visualize user data and form a user distribution map; identify key regions and potential markets according to the user distribution map.

[0077] The store distribution map shows the layout of the enterprise's physical stores in different regions; by analyzing the store distribution map, the enterprise can understand its market coverage and potential market gaps; collect and organize the store information of the enterprise, including store location, area, business status, etc.; use map software or GIS tools to visualize the store information and form a store distribution map; combine with the user distribution map to analyze the rationality of the store layout and potential market gaps.

[0078] After determining the user's consumption scenarios, place of residence, and store distribution map, the enterprise needs to combine this information to determine the corresponding marketing scenarios; marketing scenarios refer to the marketing activities carried out by the enterprise for specific user groups within a specific time and space; comprehensively analyze the user's consumption scenarios, place of residence, and store distribution map to identify potential marketing opportunities; according to the marketing opportunities, formulate specific marketing activity plans, including activity themes, times, locations, participation methods, etc.; through a combination of online and offline methods, conduct activity publicity and promotion to attract users to participate.

[0079] Taking a smart home company as an example, we opened a new physical store in a large shopping mall in the central city of a first-tier city; there are multiple large residential communities around the shopping mall, and users have a relatively high demand for our smart home products; combining with the user's consumption scenarios such as home entertainment and working from home, we plan to hold a "Smart Home Experience Day" activity during the opening period of the new physical store; the activity content includes smart product display, on-site experience, expert explanation, preferential promotions, etc.

[0080] We carried out activity publicity and promotion through various channels such as social media, offline advertising, and partners, attracting a large number of users to participate; during the activity, our smart home products received extensive attention and recognition, achieving good marketing results; by deeply analyzing the user's consumption scenarios, place of residence, and store distribution map, and combining with specific product features and market demands, the enterprise can determine the corresponding marketing scenarios and formulate corresponding marketing strategies and activity plans; this not only helps to enhance the enterprise's brand awareness and market share, but also better meets the needs and expectations of users.

[0081] Specifically, design a marketing scenario matching table to associate the user's consumption scenarios, place of residence, store distribution, and corresponding marketing scenarios. The following is a simplified example of the marketing scenario matching table:

[0082] User ID: The unique identifier of the user; Consumption scenario: The consumption situations that occur in the user's daily life; Place of residence: The specific geographical location where the user is located; Store distribution: The quantity and distribution of stores near the user's place of residence; Marketing scenario: The targeted marketing activities determined based on the user's consumption scenario and place of residence, combined with the store distribution.

[0083] In step S15, in this marketing scenario, determine the user's consumption route based on the user's current location, corresponding demand information, and current time, and match the corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and present each piece of marketing information to the user in turn; In the specific implementation process of the present invention, the specific steps may be: S151: In this marketing scenario, collect the user's demand information; S152: Determine the user's current location based on the user's location data; S153: Determine the user's consumption route according to the user's current location, corresponding demand information, and current time; S154: Determine multiple consumption nodes in the user's consumption route based on the user's consumption route, store distribution map, and the user's consumption budget; S155: Associate multiple consumption nodes in the user's consumption route and the marketing scenario; S156: Match the corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and present each piece of marketing information to the user in turn.

[0084] In the embodiment of the present application, in this marketing scenario, collect the user's demand information; determine the user's current location based on the user's location data; determine the user's consumption route according to the user's current location, corresponding demand information, and current time, which takes into account the overall consideration of the user's current location, corresponding demand information, and current time, and ensures the accuracy of the user's consumption route.

[0085] At this time, in this marketing scenario, collect the user's demand information. Optionally, obtain the user's demand information in a specific marketing scenario through methods such as questionnaire surveys, user interviews, online behavior tracking (such as browsing records, search keywords, etc.), or collection by intelligent devices (such as usage habits of smart home systems). The demand information includes the user's preferences for products, price sensitivity, purchase frequency, specific function requirements, etc.

[0086] It is necessary to collect location data from the user's devices (such as smartphones, in-vehicle navigation systems, etc.); these data include GPS data, Wi-Fi hotspot information, mobile device signals, etc.; the collected location data needs to be processed and analyzed to determine the user's precise location; when processing the user's location data, privacy protection regulations must be strictly adhered to to ensure the security and privacy of the user's data.

[0087] Based on the user's current location, corresponding demand information, and current time, determine the user's consumption route, introducing the user's current location, corresponding demand information, and current time; the demand information includes the user's clearly expressed demands (such as finding nearby restaurants, shopping centers, etc.) and potential demands obtained through data analysis (such as the types of goods the user likes, consumption habits, etc.).

[0088] The current time has a great impact on the consumption route; for example, at lunchtime, users are more inclined to look for restaurants, while in the evening, they are more inclined to look for entertainment venues; combine the user's current location, demand information, and current time to facilitate the user to plan the optimal consumption route.

[0089] Specifically, assume a user is using a navigation application; the application collects GPS data through the user's smartphone and uploads it to the server in real time; after receiving the data, the server uses map matching technology to determine the user's current location and displays it on the map; in this way, the user can clearly know their location and information about the surrounding roads and buildings, etc.; assume a user opens the application at lunchtime and expresses the need to find a nearby restaurant; the application first filters out a list of nearby restaurants based on the two factors of the user's current location and lunchtime; then, uses path search technology to plan an optimal route for the user to reach the nearest restaurant; at the same time, the application can also consider traffic conditions (such as congestion, road construction, etc.) to provide the user with a more accurate estimated arrival time; we can determine the user's current location based on the user's location data and plan the optimal consumption route for them according to their demand information and current time; this not only improves the convenience of the user but also provides more marketing opportunities for merchants.

[0090] Furthermore, based on the user's consumption route, store distribution map, and the user's consumption budget, determine multiple consumption nodes in the user's consumption route, introducing the interaction of the user's consumption route, store distribution map, and the user's consumption budget, realizing the division of multiple consumption nodes in the consumption route, accommodating the overall consideration of the user's consumption route, store distribution map, and the user's consumption budget, and ensuring the accuracy of multiple consumption nodes in the user's consumption route.

[0091] For the consumption route, first, based on the user's starting location, destination location, and intermediate stop points, combined with map data and traffic information, one or more feasible consumption routes are planned; considering the time period of the user's travel, such as morning rush hour, lunch break, evening shopping, etc., and the business hours of each consumption node, ensure that the nodes on the route are within the user's available time.

[0092] For the store distribution map, overlay the planned consumption route with the store distribution map to identify the store locations around the route; according to the user's consumption preferences (such as dining, shopping, entertainment, etc.), filter out the store types that meet the user's needs.

[0093] For the consumption budget, understand the user's consumption budget, including the overall budget and item budgets (such as dining budget, shopping budget, etc.); according to the price information of the stores (such as per capita consumption, commodity price range, etc.), filter out the stores that meet the user's budget.

[0094] For the consumption nodes, combine factors such as the user's preferences, store ratings, and promotional activities to rank the filtered stores in order of priority; according to the route plan, time nodes, budget constraints, and priority ranking, determine multiple consumption nodes for the user on the consumption route.

[0095] Specifically, assume that a user plans to have a one-day shopping and entertainment activity on the weekend, starting from home and ending at the cinema, with a consumption budget of 500 yuan; the following is a specific example: The user departs from home and plans to first go to the shopping mall, then to the food street, and finally arrive at the cinema; considering the traffic conditions and store business hours, an optimal consumption route is planned; around the shopping mall, stores such as fashion clothing, electronics, and beauty products are filtered out; around the food street, various restaurants and food stalls are filtered out; the user sets an overall budget of 500 yuan, with a shopping budget of 300 yuan and a dining budget of 200 yuan; according to the price information of the stores, stores that meet the budget are filtered out, such as restaurants with a per capita consumption of less than 100 yuan.

[0096] At the shopping mall, the user preferentially selects fashion clothing stores and electronics stores with high ratings and promotional activities; at the food street, the user selects a restaurant with a good reputation and moderate prices; finally, the user goes to the cinema to watch a movie, completing a one-day shopping and entertainment activity.

[0097] In another embodiment of the present application, a consumption node matching table is constructed, which is used to record the potential consumption nodes in the user's consumption route and perform a preliminary screening according to factors such as the user's consumption budget, store type, location, etc.; the following is an example of a consumption node matching table:

[0098] In this consumption node matching table: The consumption node number is used to uniquely identify each consumption node; the store type represents the type of store, such as a coffee shop, a bookstore, etc.; the location information describes the location of the store relative to the user's consumption route; the estimated consumption amount is the amount the user expects to spend at the store; whether it meets the budget is a boolean value indicating whether the consumption node is within the user's consumption budget; the user preference level represents the degree of the user's preference for the store type, which can be high, medium, low levels.

[0099] Through preliminary screening, we can exclude consumption nodes that do not meet the budget or have a low user preference level, such as the "high-end restaurant" in the above consumption node matching table.

[0100] Therefore, multiple consumption nodes in the user's consumption route are associated with marketing scenarios; corresponding marketing information is matched based on multiple consumption nodes in the user's consumption route and marketing scenarios, and each piece of marketing information is presented to the user in turn, achieving the matching of multiple consumption nodes in the user's consumption route and marketing scenarios, ensuring the presentation of marketing information, and improving the accuracy of marketing information.

[0101] At this time, first, based on the user's consumption route and multiple previously determined consumption nodes (such as stores, restaurants, cinemas, etc.), clarify the location and type of each node; for each consumption node, define the associated marketing scenario; the marketing scenario is based on various factors such as festivals (such as Christmas promotions), events (such as new product launches), user behavior (such as first-time shopping discounts), etc.; associate each consumption node with the corresponding marketing scenario, which usually involves data analysis to understand the user's consumption behavior and historical preferences in specific scenarios; dynamically adjust the association between the marketing scenario and the consumption node according to the user's real-time feedback and location changes to ensure the timeliness and relevance of marketing activities.

[0102] Establish a database containing various marketing information, including coupons, discounts, gifts, event details, etc.; according to the user's consumption route and associated marketing scenarios, match the marketing information related to each consumption node from the marketing information database; formulate a presentation strategy for marketing information, considering the presentation time point (such as before the user arrives at the node, when consuming at the node, etc.), method (such as text message, APP push, on-site display, etc.) and frequency (such as presenting once or reminding multiple times); make personalized adjustments to the presented marketing information according to the user's consumption history, preferences and real-time feedback to improve the user's participation and satisfaction; monitor the presentation effect of marketing information, collect user feedback data for optimizing future marketing activities.

[0103] Specifically, assume that a user plans to leave home, go shopping at a shopping mall first, then have a meal at a restaurant, and finally watch a movie at a cinema. In this scenario: the shopping mall node is associated with the "Winter Big Promotion" marketing scenario because multiple merchants are promoting winter clothing and household items.

[0104] The restaurant node is associated with the "Food Festival" marketing scenario because the restaurant participates in the local food festival activities and offers special dishes and preferential packages; the cinema node is associated with the "New Movie Release" marketing scenario because there are new movies that the user is interested in showing.

[0105] Assume that the user has planned a consumption route from the shopping mall to the restaurant and then to the cinema. When the user approaches the shopping mall, a coupon message for the "Winter Big Promotion" is pushed through the APP to encourage the user to consume in the shopping mall; before the user has a meal, a detailed message about the "Food Festival" activity, including special dishes and preferential packages, is sent via text message to attract the user to dine at the designated restaurant; when the user arrives at the cinema, a preferential message for the "New Movie Release" is displayed on the self-service ticket machine at the cinema, such as getting popcorn or a small gift when buying a ticket, to enhance the user's movie-watching experience.

[0106] In another embodiment of the present application, in order to clearly illustrate how to match marketing information based on the user's consumption nodes and marketing scenarios, we can construct a marketing information matching table; the following is an example of a marketing information matching table:

[0107] In this marketing information matching table: the consumption node number corresponds to the store selected by the user in the consumption route; the store type describes the specific type of each consumption node; the marketing scenario is a marketing theme set according to factors such as store type, season, and holidays; the marketing information is the preferential activities or promotional information specifically provided to the user.

[0108] In step S16, determine the grade coefficients of each marketing information according to each marketing information, the user's viewing duration, and the user's click times. Determine the training set according to the content of each marketing information, the grade coefficients of each marketing information, and the user's consumption portrait in the city. Perform intelligent training on the marketing scenario according to the training set and the marketing scenario trigger. In the specific implementation process of the present invention, the specific steps can be: S161: Obtain each marketing information, and each marketing information is presented to the user based on a mobile device; S162: Collect the user's viewing duration and the user's click times based on the presentation of each marketing information; S163: Perform multiple interactions on each marketing message, the viewing duration of the user, and the number of clicks of the user, and determine the ranking coefficient of each marketing message according to the multiple interactions of each marketing message, the viewing duration of the user, and the number of clicks of the user; S164: Determine the content of each marketing message by traversing each marketing message; S165: Interact the content of each marketing message, the ranking coefficient of each marketing message, and the consumption portrait of the user in this city; determine the training set according to the interaction of the content of each marketing message, the ranking coefficient of each marketing message, and the consumption portrait of the user in this city; S166: Determine multiple training dimensions based on the matching of the training set and the marketing scenario, and mark the corresponding training data in the multiple training dimensions; trigger the intelligent training of the marketing scenario according to the multiple training dimensions, the corresponding training data, and the corresponding marketing scenario to output an optimized marketing scenario to achieve the optimized management of the marketing scenario.

[0109] In the specific implementation process of the present invention, each marketing message is obtained, and each marketing message is presented to the user based on a mobile device; the viewing duration of the user and the number of clicks of the user are collected based on the presentation of each marketing message; perform multiple interactions on each marketing message, the viewing duration of the user, and the number of clicks of the user, and determine the ranking coefficient of each marketing message according to the multiple interactions of each marketing message, the viewing duration of the user, and the number of clicks of the user, realizing the multiple interactions of each marketing message, the viewing duration of the user, and the number of clicks of the user, and ensuring the accuracy of the ranking coefficient of each marketing message.

[0110] At this time, each marketing message is obtained, and each marketing message is presented to the user based on a mobile device, further controlling the interaction between the mobile device and the user. At the same time, the marketing content shown to the user by the enterprise on various marketing channels (such as social media, search engines, emails, APP push, etc.), including advertisements, coupons, event information, etc.; the time spent by the user browsing or viewing a certain marketing message; this reflects the degree of interest and attention of the user to this information; the number of times the user clicks on a certain marketing message, which usually indicates the user's interest in the information or the willingness to take action; in actual operation, the enterprise can collect these data through data analysis tools or platforms; for example, by embedding tracking codes in web pages or APPs, the click behavior and browsing duration of the user can be recorded.

[0111] The multiple interaction of each marketing information, the viewing duration of users, and the number of clicks of users refers to the comprehensive analysis of these three variables, namely marketing information, user viewing duration, and the number of clicks, considering their mutual influence and correlation. For example, if a marketing information has a high number of clicks but a short viewing duration, it means that the information attracts users to click, but the content is not attractive enough. On the contrary, if the viewing duration is long but the number of clicks is small, it means that the information content is attractive, but lacks the elements to prompt users to take action. Based on the results of the multiple interaction analysis, a grade coefficient is assigned to each marketing information to evaluate the attractiveness and effect of the information. The grade coefficient can be a numerical value or a grade classification (such as A, B, C, etc.), depending on the enterprise's evaluation criteria and requirements. Optionally, in actual operation, enterprises can use data analysis models or algorithms to conduct multiple interaction analysis and determine the grade coefficient. For example, machine learning algorithms can be used to train a large amount of historical data to identify the attractiveness and effect patterns of different marketing information, and accordingly assign grade coefficients to new information.

[0112] Specifically, assume that an e-commerce website displays an advertisement for a new mobile phone on its home page. Through data analysis tools, the website records that user A viewed the advertisement for 30 seconds and clicked 2 times, while user B only viewed it for 5 seconds and did not click. These data provide the basis for subsequent analysis. Based on the data of user A and user B, and the data of more similar users, the e-commerce website uses a data analysis model for analysis. Assume that the data analysis model finds that marketing information with a viewing duration between 15 - 30 seconds and a number of clicks greater than 1 usually has a higher conversion rate. Therefore, for the advertisement viewed and clicked by user A, the data analysis model will assign a higher grade coefficient (such as grade A), indicating that the information has high attractiveness and effect. For the information only briefly viewed and not clicked by user B, the data analysis model will assign a lower grade coefficient (such as grade C), suggesting that the information needs to be optimized or adjusted.

[0113] Furthermore, the content of each marketing information is determined by traversing each marketing information. The content of each marketing information, the grade coefficient of each marketing information, and the consumption portrait of users in this city are interacted. The training set is determined based on the interaction of the content of each marketing information, the grade coefficient of each marketing information, and the consumption portrait of users in this city, realizing the interaction of the content of each marketing information, the grade coefficient of each marketing information, and the consumption portrait of users in this city, and ensuring the richness of the training set. At this time, all available marketing information is traversed, that is, each piece of information is checked one by one. The purpose of traversing is to understand the content of each marketing information in detail, which includes text descriptions, pictures, videos, links to product or service details, etc. The determination of the content is crucial for subsequent steps, because it will directly affect the matching degree between the marketing information and the user portrait, as well as the formulation of the final marketing strategy.

[0114] Perform an interaction analysis on the marketing information content determined in the previous step, the rank coefficient of the marketing information (reflecting the attractiveness and effectiveness of the information), and the consumption portrait of the user in this city (reflecting information such as the user's consumption habits, preferences, purchasing power, etc.); the purpose of this interaction analysis is to find the correlation between the marketing information and the user portrait, and which information is more likely to attract specific types of users.

[0115] Based on this interaction analysis, the system or analyst will construct a training set; the training set is a data set containing multiple data points, and each data point contains the content of the marketing information, the rank coefficient, and the user portrait features that may match this information; this training set will be used for subsequent machine learning model training or strategy formulation.

[0116] Specifically, assume that an e-commerce platform is preparing a marketing campaign for a summer promotion; in step S164, the analyst will go through all the marketing information related to the summer promotion, such as discount information on sunscreen, recommendations for summer clothing, advertisements for beach vacation packages, etc.; for each piece of information, the analyst will record its specific content, such as the discount rate of the sunscreen, the style description of the summer clothing, the price of the beach vacation package and the included services, etc.

[0117] In the example of the e-commerce platform's summer promotion, the analyst will perform an interaction analysis on each piece of marketing information (such as the discount information on sunscreen) with its rank coefficient (such as the evaluation result based on user clicks and purchase data) and the consumption portrait of the user in this city (such as whether the user often buys sunscreen products, whether they prefer high-end brands, whether they often participate in promotional activities, etc.); through this analysis, the analyst may find that the discount information on sunscreen is particularly effective for user groups who often buy sunscreen products and are price-sensitive.

[0118] Based on this discovery, the analyst will construct a training set, which contains multiple data points similar to "sunscreen discount information + high rank coefficient + user portrait of often buying sunscreen products and being price-sensitive"; this training set can then be used to train a machine learning model, which can predict which marketing information is most likely to attract which types of users, thus helping the e-commerce platform formulate more precise marketing strategies; In another embodiment of the present application, a relationship matching table is established to record the corresponding relationship between the content of the marketing information, the rank coefficient, and the user consumption portrait; this relationship matching table can be used to construct a training set. Example of the relationship matching table:

[0119] In this relationship matching table, each row represents a training data point; for example, the first piece of data represents a marketing message about summer clothing discounts, with a grade coefficient of A, applicable to user groups who like summer clothing and have a medium consumption level, and this data point is classified into training set 1.

[0120] Therefore, multiple training dimensions are determined based on the matching of this training set and the marketing scenario, and the corresponding training data is marked in multiple training dimensions; intelligent training of the marketing scenario is triggered according to multiple training dimensions, the corresponding training data, and the corresponding marketing scenario to output an optimized marketing scenario, so as to realize the optimized management of the marketing scenario, realize the interaction of multiple training dimensions, the corresponding training data, and the corresponding marketing scenario, ensure the accurate triggering of the intelligent training of the marketing scenario, and further control the optimized management of the marketing scenario.

[0121] At this time, the training set (including data such as user consumption portraits and marketing information) is matched with a specific marketing scenario to identify multiple key training dimensions; these dimensions may include user interests, consumption behaviors, product characteristics, marketing channels, etc.; by analyzing the data in the training set and combining the specific characteristics of the marketing scenario, the factors that have an important impact on the marketing effect are extracted as training dimensions.

[0122] After determining the training dimensions, it is necessary to mark the data under each dimension so that the subsequent intelligent training can accurately identify and utilize this data; assign labels or codes to the data under each training dimension to ensure the accuracy and traceability of the data; for example, for the dimension of user interests, the preference degrees of users for different types of products can be encoded and marked; further, use machine learning or deep learning algorithms, combine multiple training dimensions and the corresponding training data, and perform simulation and optimization training on the marketing scenario; select a suitable machine learning model, input the marked training data into the model for training; by adjusting the model parameters and optimizing the algorithm, improve the prediction and adaptation ability of the model to the marketing scenario.

[0123] After intelligent training, the model will output an optimized marketing scenario plan; these plans are designed to improve the marketing effect, meet user needs, and at the same time reduce the marketing cost; apply the results of the intelligent training to the actual marketing scenario, and continuously adjust and optimize the plan through methods such as A / B testing and user feedback, and finally realize the optimized management of the marketing scenario.

[0124] Specifically, assume that we have an online retail platform that hopes to optimize the scenario of its email marketing campaign; the following is a specific example of operating according to the above steps: Determine multiple training dimensions based on the matching of the training set and the marketing scenario: The training set includes data such as the user's purchase history, browsing behavior, and hobbies; Marketing scenario: An email marketing campaign aimed at promoting new products on the platform; Training dimensions: According to the characteristics of the training set and the marketing scenario, determine the following training dimensions: User purchase preferences (such as product type, price sensitivity), browsing behavior characteristics (such as page dwell time, click-through rate), hobbies (such as sports, fashion, etc.).

[0125] Encode and label the user purchase preferences, such as labeling users who have purchased sports products as "sports preference"; Quantify the browsing behavior characteristics, such as calculating the average page dwell time and click-through rate of each user; Classify and label the hobbies, such as labeling users who like fashion content as "fashion lovers".

[0126] Select a suitable machine learning model (such as logistic regression, random forest, etc.) for training; Input the labeled training data into the model for training, and adjust the model parameters to optimize the prediction ability for the email marketing campaign effect; After intelligent training, the model outputs an optimized email marketing campaign plan; For example, send emails containing the promotion information of the latest sports products to the user group with "sports preference" and "long page dwell time"; Collect data through A / B testing and user feedback, and adjust and optimize the optimized marketing scenario; For example, when it is found that a certain type of user has a poor response to the promotion of a specific type of product, adjust the promotion strategy or replace the promoted product in a timely manner.

[0127] The above marketing scenario intelligent training management method, device, and storage medium collect the consumption data set of users in different cities; Determine the user's consumption portrait in the city according to the consumption data set, the user's living information, and the corresponding city; Determine the consumption parameters of the user in the city based on the user's income level, the corresponding consumption time, and the user's preferences in the city; Determine the user's consumption scenario according to the consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario according to the user's consumption scenario, the user's place of residence, and the corresponding store distribution map, which takes into account the overall consideration of the user's consumption scenario, the user's place of residence, and the corresponding store distribution map, and ensures the accuracy of the marketing scenario.

[0128] Furthermore, in this marketing scenario, determine the user's consumption route based on the user's current location, corresponding demand information, and current time, and match the corresponding marketing information according to the multiple consumption nodes in the user's consumption route and the marketing scenario. Each marketing information is presented to the user in turn, so as to present various aspects of marketing information according to the consumption route, which is convenient for subsequent control of each marketing information.

[0129] Therefore, the grade coefficients of each marketing message are determined based on each marketing message, the viewing duration of the user, and the number of clicks of the user. A training set is determined based on the content of each marketing message, the grade coefficient of each marketing message, and the consumption portrait of the user in this city. The intelligent training of the marketing scenario is triggered based on this training set and this marketing scenario, so as to accommodate the considerations of the user in each city and ensure the applicability of this marketing scenario in each city. Embodiment III

[0130] In this embodiment, as Figure 3 shown, a marketing scenario intelligent training management device is provided, including: A collection module, configured to collect the consumption data set of the user in different cities; A consumption portrait module, configured to determine the consumption portrait of the user in this city according to the consumption data set, the living information of the user, and the corresponding city; A consumption parameter module, configured to determine the consumption parameters of the user in this city based on the income level of the user, the corresponding consumption time, and the preferences of the user in this city; A marketing scenario module, configured to determine the consumption scenario of the user according to this consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario according to the consumption scenario of the user, the place of residence of the user, and the corresponding store distribution map; A marketing message module, configured to determine the consumption route of the user based on the current location of the user, the corresponding demand information, and the current time in this marketing scenario, and match the corresponding marketing messages according to multiple consumption nodes in the consumption route of the user and the marketing scenario, and each marketing message is presented to the user in turn; An intelligent training module, configured to determine the grade coefficients of each marketing message according to each marketing message, the viewing duration of the user, and the number of clicks of the user, determine a training set according to the content of each marketing message, the grade coefficient of each marketing message, and the consumption portrait of the user in this city, and trigger the intelligent training of the marketing scenario according to this training set and this marketing scenario. Embodiment IV

[0131] In this embodiment, a marketing scenario intelligent training management device is provided; its internal structure diagram can be as Figure 4As shown in the figure; the intelligent training management device for marketing scenarios includes a processor, a memory, a network interface, a display screen, and an input device connected through a system bus; among them, the processor of the intelligent training management device for marketing scenarios is used to provide computing and control capabilities; the memory of the intelligent training management device for marketing scenarios includes a non-volatile storage medium and an internal memory; the non-volatile storage medium stores an operating system and a computer program, and a database is deployed on the non-volatile storage medium, and the database is used to store user behavior data and user portraits; the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium; the network interface of the intelligent training management device for marketing scenarios is used to communicate with other devices on which application software is deployed; when the computer program is executed by the processor, it implements an intelligent training management method for marketing scenarios; the display screen of the intelligent training management device for marketing scenarios can be a liquid crystal display screen or an electronic ink display screen, and the input device of the intelligent training management device for marketing scenarios can be a touch layer covered on the display screen, or a button, a trackball or a touchpad provided on the device shell, or an external keyboard, touchpad or mouse, etc.

[0132] The above embodiments only represent several implementation manners of the present application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent; it should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several deformations and improvements can still be made, and these all belong to the protection scope of the present application; therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A marketing scenario intelligent training management method, characterized in that: include: Collect consumption data of users in different cities; Determine the consumption profile of the user in the city based on the consumption data set, the user's life information and the corresponding city; Determine the user's consumption parameters in the city based on the user's income level, corresponding consumption time, and the user's preferences in the city; Determine the user's consumption scenario based on the consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario based on the user's consumption scenario, the user's place of residence, and the corresponding store distribution map; In this marketing scenario, the user's consumption route is determined based on the user's current location, corresponding demand information, and current time, and corresponding marketing information is matched according to multiple consumption nodes in the user's consumption route and marketing scenarios, and each piece of marketing information is presented to the user in sequence; The level coefficient of each marketing message is determined based on the user's viewing time and the user's click times. A training set is determined based on the content of each marketing message, the level coefficient of each marketing message and the user's consumption profile in the city. Intelligent training of the marketing scenario is triggered based on the training set and the marketing scenario.

2. The marketing scenario intelligent training management method according to claim 1, characterized in that: The collection of consumption data of users in different cities includes: Mark the user's payment platform; Export the user's previous payment data based on the user's payment platform; Determine the data set for each city based on the user's previous payment data and the matching of the corresponding payment location; The consumption data of users in different cities are determined based on the data sets of each city and the consumption types of users.

3. The marketing scenario intelligent training management method according to claim 2 is characterized in that: Determining the consumption profile of the user in the city according to the consumption data set, the user's life information and the corresponding city includes: Get consumption data collection; Collecting user's life information based on traversal of user's life database; Associate consumption data sets, user life information and corresponding cities; Determine a first consumption feature according to the consumption data set and the user's life information; Determine a second consumption characteristic according to the consumption data set and the corresponding city; The consumption profile of the user in the city is determined based on the first consumption characteristic, the second consumption characteristic, and the length of stay of the user in the city.

4. The marketing scenario intelligent training management method according to claim 3 is characterized in that: The determining of the consumption parameters of the user in the city based on the user's income level, corresponding consumption time and the user's preference in the city includes: Collecting user income data based on the user's traversal of payment platforms; Determine the user's income level based on the user's income data, the user's job title in the company, and the user's place of residence; Determine multiple consumption categories and corresponding consumption frequencies based on the traversal of the user's consumption data in the city; Determine the user's preference in a city based on multiple consumption categories, corresponding consumption frequencies, and corresponding cities; Interact with the user's income level, corresponding consumption time, and the user's preferences in the city; The consumption parameters of the user in the city are determined according to the user's income level, the corresponding consumption time and the interaction of the user's preferences in the city.

5. The marketing scenario intelligent training management method according to claim 1, characterized in that: Determining the user's consumption scenario based on the consumption portrait and the corresponding consumption parameters, and determining the corresponding marketing scenario based on the user's consumption scenario, the user's residence, and the corresponding store distribution map, includes: Get the consumption profile and corresponding consumption parameters; Match the corresponding scenario weight to the consumption profile and corresponding consumption parameters; Determine the user's consumption scenario based on the consumption profile, consumption parameters, and corresponding scenario weights; Determine the user's place of residence based on the user's residential data; Determine a corresponding store distribution map based on the user's residence, a drawing of the surrounding area of ​​the residence, and store marks; Determine the corresponding marketing scenario based on the user's consumption scenario, the user's place of residence and the corresponding store distribution map.

6. The marketing scenario intelligent training management method according to claim 1, characterized in that: In the marketing scenario, the user's consumption route is determined based on the user's current location, corresponding demand information, and current time, and corresponding marketing information is matched according to multiple consumption nodes in the user's consumption route and marketing scenarios. Each marketing information is presented to the user in sequence, including: In this marketing scenario, user demand information is collected; determining a current location of the user based on the user's location data; Determine the user's consumption route based on the user's current location, corresponding demand information, and current time; Determine multiple consumption nodes in the user's consumption route based on the user's consumption route, the store distribution map, and the user's consumption budget; Associate multiple consumption nodes and marketing scenarios in the user's consumption route; Based on multiple consumption nodes in the user's consumption route and the corresponding marketing information matched with the marketing scenario, each piece of marketing information is presented to the user in turn.

7. The marketing scenario intelligent training management method according to claim 1, characterized in that: The step of determining the level coefficient of each marketing information according to each marketing information, the viewing time of the user, and the number of clicks of the user, determining the training set according to the content of each marketing information, the level coefficient of each marketing information, and the consumption profile of the user in the city, and triggering the intelligent training of the marketing scenario according to the training set and the marketing scenario includes: Obtain various marketing information, and present the various marketing information to users based on mobile devices; Collect the user's viewing time and number of clicks based on the presentation of each marketing message; Perform multiple interactions on each marketing message, the viewing time of the user, and the number of clicks of the user, and determine the level coefficient of each marketing message based on the multiple interactions of each marketing message, the viewing time of the user, and the number of clicks of the user;.

8. The marketing scenario intelligent training management method according to claim 7, characterized in that: The step of determining the level coefficient of each marketing information according to each marketing information, the viewing time of the user, and the number of clicks of the user, determining the training set according to the content of each marketing information, the level coefficient of each marketing information, and the consumption profile of the user in the city, and triggering the intelligent training of the marketing scenario according to the training set and the marketing scenario, further includes: Determine the content of each marketing information according to the traversal of each marketing information; Interacting the content of each marketing information, the level coefficient of each marketing information, and the consumption profile of the user in the city; determining the training set according to the interaction of the content of each marketing information, the level coefficient of each marketing information, and the consumption profile of the user in the city; Based on the matching of the training set and the marketing scenario, multiple training dimensions are determined, and corresponding training data are marked in the multiple training dimensions; intelligent training of the marketing scenario is triggered according to the multiple training dimensions, the corresponding training data and the corresponding marketing scenario to output the optimized marketing scenario, so as to achieve optimized management of the marketing scenario.

9. A marketing scenario intelligent training management device, characterized in that: The marketing scenario intelligent training management device is applied to the marketing scenario intelligent training management method according to any one of claims 1 to 8, and the marketing scenario intelligent training management device includes: The collection module is used to collect consumption data sets of users in different cities; The consumption profile module is used to determine the consumption profile of the user in the city based on the consumption data set, the user's life information and the corresponding city; A consumption parameter module, used to determine the consumption parameters of a user in a city based on the user's income level, corresponding consumption time and the user's preferences in the city; The marketing scenario module is used to determine the user's consumption scenario based on the consumption portrait and the corresponding consumption parameters, and determine the corresponding marketing scenario based on the user's consumption scenario, the user's place of residence and the corresponding store distribution map; The marketing information module is used to determine the user's consumption route based on the user's current location, corresponding demand information and current time in the marketing scenario, and match corresponding marketing information according to multiple consumption nodes in the user's consumption route and the marketing scenario, and present each marketing information to the user in sequence; The intelligent training module is used to determine the grade coefficient of each marketing information based on the marketing information, the user's viewing time and the user's click times, determine the training set based on the content of each marketing information, the grade coefficient of each marketing information and the user's consumption profile in the city, and trigger intelligent training of the marketing scenario based on the training set and the marketing scenario.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the marketing scenario intelligent training management method described in any one of claims 1 to 8 are implemented.

Images