A marketing plan intelligent push method, system and storage medium

Abstract

The present invention relates to the field of program push, and discloses a marketing program intelligent push method, system and storage medium, including the following steps: obtaining user information, and conducting in-depth analysis of the user information, dividing the users into new users and old users; classifying the content information of the marketing program, and combining the user's personal information, conducting a feasibility analysis of the marketing program for each user; combining the user's current status and the marketing program conditions in the push website, constructing a marketing program push sequence, and optimizing the marketing program push sequence according to the consumption situation and user feedback of the pushed user. The present invention can accurately recommend suitable marketing programs to different users based on the personal situations of different users by conducting in-depth learning of the marketing program push method, thereby improving the user's purchasing efficiency and enabling the user to obtain a comfortable purchasing experience.

Description

Technical Field

[0001] The present invention relates to the field of program push, and in particular to a marketing program intelligent push method, system and storage medium. Background Art

[0002] In today's society, there are many promotional activities, and usually multiple items are sold in bundles, so that merchants can obtain higher profits, and users can buy more and more affordable items at more favorable prices, so there are many marketing plans. A marketing plan contains multiple marketing items. For example, the marketing plan for buying a mobile phone usually comes with a charging cable, a charging head, headphones, etc. When users buy items, they usually choose marketing plans with greater discounts, so it is necessary to push marketing plans that are more in line with the user's personal needs to users for users to choose to buy. In the push process, by analyzing user personal information, preference values, and the evaluation of the push website, the push rules can be optimized, and the marketing plan can also be updated and optimized. Therefore, it is necessary to obtain a marketing plan intelligent push method, system and storage medium. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a marketing plan intelligent push method, system and storage medium.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] A first aspect of the present invention provides a marketing plan intelligent push method, comprising the following steps:

[0006] Acquire user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results;

[0007] Classify the content information of the marketing plan to obtain marketing plans of different categories, and combine the user information of new and old users to conduct feasibility analysis on the marketing plan;

[0008] In the push website, the marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence;

[0009] Analyze the consumption and feedback of the users being pushed, and optimize the push order of the marketing plans based on the analysis results.

[0010] Furthermore, in a preferred embodiment of the present invention, the user information is obtained, the user information is deeply analyzed, and the users are classified according to the in-depth analysis results, specifically:

[0011] Get the target website for marketing plan push, define it as the push website, and get the historical user information in the push website;

[0012] Real-time acquisition of user information of online users in the current push website, defined as online user information, converting historical user information and online user information into feature data, respectively calculating the inner product and respective modulus of the historical user information and online user information, and calculating the cosine similarity of the historical user information and online user information based on the inner product and modulus;

[0013] A cosine similarity interval is preset, and if the cosine similarity between the historical user information and the online user information is within the cosine similarity interval, the user represented by the corresponding online user information is marked as an old user;

[0014] If the cosine similarity between the historical user information and the online user information is not within the cosine similarity interval, the user represented by the corresponding online user information is marked as a new user.

[0015] Furthermore, in a preferred embodiment of the present invention, the content information of the marketing plan is classified to obtain marketing plans of different categories, and the feasibility analysis of the marketing plan is performed in combination with the user information of new users and old users, specifically:

[0016] In the push website, all marketing plans are obtained, and content information of each marketing plan is obtained respectively;

[0017] Extract features of marketing plan content information through Fourier transform to obtain marketing plan content feature data;

[0018] Constructing fuzzy clusters and initializing the fuzzy clusters, wherein the fuzzy clusters contain fuzzy centers, and one fuzzy cluster represents the content of a marketing plan of one category;

[0019] Using fuzzy clustering method and iterative calculation method, the Euclidean distance between the marketing plan content feature data and the fuzzy center is calculated, and based on the Euclidean distance between the marketing plan content feature data and the fuzzy center, the membership degree of the marketing plan content information and the fuzzy cluster is obtained;

[0020] According to historical data retrieval, the marketing plan content information is sorted by importance to obtain a marketing plan content information sorting table, based on the marketing plan content information sorting table, the membership degree of the marketing plan content information and the fuzzy cluster is updated, and the marketing plans are classified to obtain marketing plans of different categories;

[0021] Obtaining user information of new and old users, wherein the user information of new and old users includes user consumption upper limit, user consumption interest, and user gender and age;

[0022] The user information of the new and old users and different types of marketing plans are imported into the CNN neural network to perform user feasibility analysis to obtain user feasibility analysis results. Based on the user feasibility analysis results, suitable marketing plans and inappropriate marketing plans for different users are screened.

[0023] Furthermore, in a preferred embodiment of the present invention, in the push website, a marketing plan push sequence is constructed in combination with the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence, specifically:

[0024] Set up a questionnaire on the push website login interface and guide users to fill in the questionnaire to get the user's current status;

[0025] Acquire the number of marketing items currently in stock at the merchant in real time, and use the number of marketing items currently in stock at the merchant as a condition for the marketing plan;

[0026] Based on the marketing plan conditions, suitable marketing plans for different users are analyzed. If the number of marketing items required in the suitable marketing plan is greater than the number of marketing items currently available in the merchant, the corresponding suitable marketing plan is defined as a marketing plan that cannot be pushed;

[0027] In combination with the current state of the user, the remaining suitable marketing plans are predicted and processed in a convolutional neural network to obtain a marketing plan that can be pushed;

[0028] Obtaining the popularity of pushable marketing plans, where the popularity of pushable marketing plans includes the number of views and the number of consumptions of the pushable marketing plans, and constructing a pushable marketing plan popularity ranking table for different users based on the popularity of the pushable marketing plans;

[0029] If the user is a new user, the pushable marketing plan is pushed to the new user based on the descending order of popularity in the pushable marketing plan popularity ranking table corresponding to the new user;

[0030] If the user is an old user, a push sequence table of marketing plans that can be pushed to old users is constructed according to the consumption records of the old users, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to the old users.

[0031] Furthermore, in a preferred embodiment of the present invention, if the user is an old user, a push sequence table of marketing plans that can be pushed to the old user is constructed according to the consumption record of the old user, and based on the push sequence table of marketing plans that can be pushed to the old user, the pushable marketing plans are pushed to the old user, specifically:

[0032] If the user is an old user, the consumption record of the old user in the push website is obtained, and based on the consumption record of the old user in the push website, the consumption frequency of the old user for the pushable marketing plan is obtained, and the consumption frequency of the old user for the pushable marketing plan is sorted based on the descending principle to construct the first push sequence table of the pushable marketing plan for the old user;

[0033] Obtain pushable plans that old users have not consumed, sort the pushable plans that old users have not consumed based on the popularity ranking table of pushable marketing plans and based on the principle of descending popularity, and obtain a second push order table of pushable marketing plans for old users;

[0034] Combined with the first push sequence table of marketing plans that can be pushed to old users and the second push sequence table of marketing plans that can be pushed to old users, a push sequence table of marketing plans that can be pushed to old users is constructed, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to old users.

[0035] Furthermore, in a preferred embodiment of the present invention, the consumption situation and user feedback of the pushed users are analyzed, and the push order of the marketing plan is optimized according to the analysis results, specifically:

[0036] After pushing the pushable marketing plan to the user, collect the consumption status and user feedback of the pushed user;

[0037] Based on the pushed user consumption situation, obtain the consumption status of each marketing plan within a preset time, wherein the consumption status of each marketing plan includes the consumption quantity and consumption frequency of the marketing plan;

[0038] Arrange the consumption quantity and consumption frequency of each marketing plan in descending order to obtain a marketing plan consumption quantity ranking table and a marketing plan consumption frequency ranking table, and compare and analyze the marketing plan consumption quantity ranking table and the marketing plan consumption frequency ranking table with the marketing plan push sequence table. If the descending sorting results of the consumption quantity and consumption frequency of the marketing plan are the same as the push sequence of the marketing plan, there is no need to optimize the push sequence of each marketing plan;

[0039] If the descending sorting result of the consumption quantity and consumption frequency of the marketing plan is different from the push order of the marketing plan, then the deviation rate between the descending sorting result of the consumption quantity and consumption frequency of the marketing plan and the push order of the marketing plan is obtained, which is defined as the first deviation rate;

[0040] Analyze the first deviation rate, and if the first deviation rate is less than a preset value, optimize the order of pushing the marketing plans to new and old users respectively based on the results of the descending sorting of the consumption quantity and the consumption frequency of the marketing plans;

[0041] If the first deviation rate is greater than a preset value, the user feedback is analyzed, and the user feedback is the user's score on the marketing plan, and a qualified score is preset. If the user's score on the marketing plan is less than the qualified score, the corresponding marketing plan is eliminated and cannot be pushed;

[0042] The marketing plans with scores greater than the qualified score are defined as qualified marketing plans, and the qualified marketing plans are sorted in descending order of scores to obtain a push order table of qualified marketing plans based on scores;

[0043] According to the qualified marketing plan push sequence table based on scoring, the marketing plan push sequence for new and old users is optimized respectively.

[0044] The second aspect of the present invention further provides a marketing plan intelligent push system, the intelligent push system comprising a memory and a processor, the memory storing a marketing plan intelligent push method, and when the marketing plan intelligent push method is executed by the processor, the following steps are implemented:

[0045] Acquire user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results;

[0046] Classify the content information of the marketing plan to obtain marketing plans of different categories, and combine the user information of new and old users to conduct feasibility analysis on the marketing plan;

[0047] In the push website, the marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence;

[0048] Analyze the consumption and feedback of the users being pushed, and optimize the push order of the marketing plans based on the analysis results.

[0049] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: obtaining user information, and conducting in-depth analysis of the user information, dividing users into new users and old users; classifying the content information of the marketing plan, and combining the user's personal information, conducting a feasibility analysis of the marketing plan for each user; combining the user's current status and the marketing plan conditions in the push website, constructing the marketing plan push sequence, and optimizing the marketing plan push sequence according to the consumption of the pushed user and the user feedback. The present invention can accurately recommend suitable marketing plans to different users based on the personal situations of different users through in-depth learning of the marketing plan push method, thereby improving the user's purchasing efficiency and enabling the user to obtain a comfortable purchasing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A flow chart showing a method for intelligently pushing a marketing plan is shown;

[0052] Figure 2 A flow chart showing a method for pushing marketing plans to new and old users according to the order in which marketing plans are pushed;

[0053] Figure 3 A view of a marketing plan intelligent push system is shown. DETAILED DESCRIPTION

[0054] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0056] Figure 1 A flowchart of a marketing plan intelligent push method is shown, comprising the following steps:

[0057] S102: Obtain user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results;

[0058] S104: Classify the content information of the marketing plan to obtain marketing plans of different categories, and perform feasibility analysis on the marketing plan in combination with user information of new users and old users;

[0059] S106: In the push website, a marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence;

[0060] S108: Analyze the consumption and feedback of the users to whom the marketing plans are pushed, and optimize the order of pushing the marketing plans based on the analysis results.

[0061] Furthermore, in a preferred embodiment of the present invention, the user information is obtained, the user information is deeply analyzed, and the users are classified according to the in-depth analysis results, specifically:

[0062] Get the target website for marketing plan push, define it as the push website, and get the historical user information in the push website;

[0063] Real-time acquisition of user information of online users in the current push website, defined as online user information, converting historical user information and online user information into feature data, respectively calculating the inner product and respective modulus of the historical user information and online user information, and calculating the cosine similarity of the historical user information and online user information based on the inner product and modulus;

[0064] A cosine similarity interval is preset, and if the cosine similarity between the historical user information and the online user information is within the cosine similarity interval, the user represented by the corresponding online user information is marked as an old user;

[0065] If the cosine similarity between the historical user information and the online user information is not within the cosine similarity interval, the user represented by the corresponding online user information is marked as a new user.

[0066] It should be noted that a large amount of user data is stored in the marketing plan push website, including user browsing data and ordering data. Users who have browsed and placed orders on the push website have a certain understanding of the marketing plan and are defined as old users; new users are users who come to the website for the first time, and the website does not store the user's information. During the user login process, the user's login information is online user information. The website's database can be called to determine whether the user is a new user or an old user. The cosine similarity calculation method is used to calculate the similarity between the user information at the time of user login and the information in the database. If the similarity is greater than the preset value, it proves that the user is an old user, otherwise it is a new user. The present invention can classify users by calculating the cosine similarity between online user information and historical user information.

[0067] Furthermore, in a preferred embodiment of the present invention, the content information of the marketing plan is classified to obtain marketing plans of different categories, and the feasibility analysis of the marketing plan is performed in combination with the user information of new users and old users, specifically:

[0068] In the push website, all marketing plans are obtained, and content information of each marketing plan is obtained respectively;

[0069] Extract features of marketing plan content information through Fourier transform to obtain marketing plan content feature data;

[0070] Constructing fuzzy clusters and initializing the fuzzy clusters, wherein the fuzzy clusters contain fuzzy centers, and one fuzzy cluster represents the content of a marketing plan of one category;

[0071] Using fuzzy clustering method and iterative calculation method, the Euclidean distance between the marketing plan content feature data and the fuzzy center is calculated, and based on the Euclidean distance between the marketing plan content feature data and the fuzzy center, the membership degree of the marketing plan content information and the fuzzy cluster is obtained;

[0072] According to historical data retrieval, the marketing plan content information is sorted by importance to obtain a marketing plan content information sorting table, based on the marketing plan content information sorting table, the membership degree of the marketing plan content information and the fuzzy cluster is updated, and the marketing plans are classified to obtain marketing plans of different categories;

[0073] Obtaining user information of new and old users, wherein the user information of new and old users includes user consumption upper limit, user consumption interest, and user gender and age;

[0074] The user information of the new and old users and different types of marketing plans are imported into the CNN neural network to perform user feasibility analysis to obtain user feasibility analysis results. Based on the user feasibility analysis results, suitable marketing plans and inappropriate marketing plans for different users are screened.

[0075] It should be noted that the marketing items included in different marketing plans may be the same. For example, if the marketing plan is a food set, there are staple foods and snacks in a set, and the staple foods or snacks in different sets may be the same, so the set can be classified and processed, and can be classified based on staple foods or snacks. The marketing plan can be classified using the fuzzy clustering method. The Euclidean distance between the content features of the marketing plan and the fuzzy center can be calculated to obtain the membership of the marketing plan content information and the fuzzy cluster, thereby achieving the classification effect. The user may not buy a certain commodity in the marketing plan, or there may be a situation where a certain commodity in the marketing plan must be purchased by the user, so a feasibility analysis is performed on the user information of new and old users, and the feasibility analysis includes analyzing whether the user's consumption ability can purchase a certain commodity, or whether the user has a rigid demand for the commodity, etc. According to the feasibility analysis results, suitable marketing plans and inappropriate marketing plans for different users can be obtained. The present invention can classify the marketing plan by the fuzzy clustering method, and perform user feasibility analysis on different marketing plans to obtain suitable marketing plans and inappropriate marketing plans.

[0076] Furthermore, in a preferred embodiment of the present invention, the consumption situation and user feedback of the pushed users are analyzed, and the push order of the marketing plan is optimized according to the analysis results, specifically:

[0077] After pushing the pushable marketing plan to the user, collect the consumption status and user feedback of the pushed user;

[0078] Based on the pushed user consumption situation, obtain the consumption status of each marketing plan within a preset time, wherein the consumption status of each marketing plan includes the consumption quantity and consumption frequency of the marketing plan;

[0079] Arrange the consumption quantity and consumption frequency of each marketing plan in descending order to obtain a marketing plan consumption quantity ranking table and a marketing plan consumption frequency ranking table, and compare and analyze the marketing plan consumption quantity ranking table and the marketing plan consumption frequency ranking table with the marketing plan push sequence table. If the descending sorting results of the consumption quantity and consumption frequency of the marketing plan are the same as the push sequence of the marketing plan, there is no need to optimize the push sequence of each marketing plan;

[0080] If the descending sorting result of the consumption quantity and consumption frequency of the marketing plan is different from the push order of the marketing plan, then the deviation rate between the descending sorting result of the consumption quantity and consumption frequency of the marketing plan and the push order of the marketing plan is obtained, which is defined as the first deviation rate;

[0081] Analyze the first deviation rate, and if the first deviation rate is less than a preset value, optimize the order of pushing the marketing plans to new and old users respectively based on the results of the descending sorting of the consumption quantity and the consumption frequency of the marketing plans;

[0082] If the first deviation rate is greater than a preset value, the user feedback is analyzed, where the user feedback is the user's score on the marketing plan, and a qualified score is preset. If the user's score on the marketing plan is less than the qualified score, the corresponding marketing plan is eliminated and cannot be pushed;

[0083] The marketing plans with scores greater than the qualified score are defined as qualified marketing plans, and the qualified marketing plans are sorted in descending order of scores to obtain a push order table of qualified marketing plans based on scores;

[0084] According to the qualified marketing plan push sequence table based on scoring, the marketing plan push sequence for new and old users is optimized respectively.

[0085] It should be noted that after the marketing plan is pushed to the user, it is necessary to judge whether the user places an order according to the push order. The user's consumption situation can be obtained. If the descending sorting result of the consumption quantity and consumption frequency in the user's consumption situation is the same as the marketing plan push order, it proves that the user consumes according to the marketing plan push order, and there is no need to optimize the marketing plan push order. If the descending sorting result of the consumption quantity and consumption frequency in the user's consumption situation is different from the marketing plan push order, it is necessary to optimize the marketing plan push order. Analyze the first deviation rate. When the first deviation rate is less than the preset value, it proves that the user basically consumes according to the marketing plan push order, and the descending sorting result of the consumption quantity and consumption frequency of the marketing plan can be used as the push order. When the first deviation rate is greater than the preset value, it proves that the user basically does not consume according to the marketing plan push order. Obtain user feedback and score the marketing plan. When the score is less than the qualified score, it proves that the user is unwilling to buy the marketing plan and the quality is poor, so it is eliminated; when the score is greater than the qualified score, it proves that the marketing plan is qualified, and the qualified marketing plan is ranked in descending order of scores to obtain a qualified marketing plan push order table based on the score and push it. The present invention can optimize the push sequence of marketing plans by analyzing the consumption situation of the pushed users and the user feedback situation.

[0086] Figure 2 A flow chart of a method for pushing marketing plans to new and old users according to the order in which marketing plans are pushed is shown, including the following steps:

[0087] S202: Based on the current status of the user and the marketing plan conditions, obtain the marketing plans that can be pushed and the marketing plans that cannot be pushed, and build a popularity ranking table of the marketing plans that can be pushed for different users;

[0088] S204: based on the popularity ranking table of pushable marketing plans, and in descending order of popularity, push the pushable marketing plans to the new user;

[0089] S206: Pushing the pushable marketing plans to the old users according to the push order table of the pushable marketing plans to the old users.

[0090] Furthermore, in a preferred embodiment of the present invention, the marketing plans that can be pushed and the marketing plans that cannot be pushed are obtained by combining the current status of the user and the marketing plan conditions, and a popularity ranking table of marketing plans that can be pushed is constructed for different users, specifically:

[0091] Set up a questionnaire on the push website login interface and guide users to fill in the questionnaire to get the user's current status;

[0092] Acquire the number of marketing items currently in stock at the merchant in real time, and use the number of marketing items currently in stock at the merchant as a condition for the marketing plan;

[0093] Based on the marketing plan conditions, suitable marketing plans for different users are analyzed. If the number of marketing items required in the suitable marketing plan is greater than the number of marketing items currently available in the merchant, the corresponding suitable marketing plan is defined as a marketing plan that cannot be pushed;

[0094] In combination with the current state of the user, the remaining suitable marketing plans are predicted and processed in a convolutional neural network to obtain a marketing plan that can be pushed;

[0095] The popularity of the pushable marketing plan is obtained, where the popularity of the pushable marketing plan includes the number of views and the number of consumptions of the pushable marketing plan. Based on the popularity of the pushable marketing plan, a pushable marketing plan popularity ranking table is constructed for different users.

[0096] It should be noted that the purpose of setting up the questionnaire is to obtain the current status of the user, which includes the user's consumption upper limit, the user's location, the user's consumption willingness, etc. The number of existing marketing items of the merchant is the marketing plan condition. Different marketing plans have different popularity. For example, the marketing plan where popular products are located will have higher sales. Analyze the marketing plan conditions. If the number of marketing items required for the marketing plan is greater than the number of existing marketing items of the merchant, it proves that the marketing plan cannot be satisfied, and the corresponding suitable marketing plan is marked as a non-push plan. Among the remaining suitable marketing plans, there may be some marketing plans that are temporarily not needed by the user. For example, in the catering industry, a user terminal is set up with five people dining, and some marketing plans are double sets, then the corresponding marketing plans cannot be pushed. After obtaining the pushable marketing plan based on the convolutional neural network, according to the number of views and the number of consumption, according to the different new and old users, a pushable marketing plan heat ranking table can be constructed for different users. The present invention can obtain pushable marketing plans and non-pushable marketing plans by analyzing the current state of the user and the marketing plan conditions, and construct a pushable marketing plan heat ranking table for different users.

[0097] Furthermore, in a preferred embodiment of the present invention, the marketing scheme that can be pushed to the old user according to the push sequence table of the marketing scheme that can be pushed to the old user is pushed as follows:

[0098] If the user is an old user, the consumption record of the old user in the push website is obtained, and based on the consumption record of the old user in the push website, the consumption frequency of the old user for the pushable marketing plan is obtained, and the consumption frequency of the old user for the pushable marketing plan is sorted based on the descending principle to construct the first push sequence table of the pushable marketing plan for the old user;

[0099] Obtain pushable plans that old users have not consumed, sort the pushable plans that old users have not consumed based on the popularity ranking table of pushable marketing plans and based on the principle of descending popularity, and obtain a second push order table of pushable marketing plans for old users;

[0100] Combined with the first push sequence table of marketing plans that can be pushed to old users and the second push sequence table of marketing plans that can be pushed to old users, a push sequence table of marketing plans that can be pushed to old users is constructed, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to old users.

[0101] It should be noted that when the user is an old user, it proves that the user has browsing and consumption records at present, and the marketing plans that the user has repeatedly consumed are pushed first. The reason is that the old user's repeated purchases prove that the old user has a higher preference value for the marketing plan, and the consumption frequency of the old user for the pushable marketing plans is sorted according to the descending principle, and the first push sequence table of the old user's pushable marketing plans is constructed. For the pushable plans that the old user has not consumed, the second push sequence table of the old user's pushable marketing plans is constructed according to the descending principle of popularity. Finally, the two push sequence tables are combined to obtain the push sequence table of the old user's pushable marketing plans, and the pushable marketing plans are pushed to the old users. The present invention can analyze the old user's consumption records and the popularity of the marketing plans, and push the pushable marketing plans to the old users according to the analysis results.

[0102] In addition, the marketing plan intelligent push method further includes the following steps:

[0103] Obtaining the quantity of raw materials required for each marketing item, and combining it with the quantity of marketing items currently in stock by the merchant, obtaining the total number of raw materials for each marketing item;

[0104] According to the push order of the marketing plans, the types of marketing items in each marketing plan are obtained in descending order, and the sales records of each marketing plan are obtained;

[0105] Importing the sales records of each marketing plan into a convolutional neural network for convolution processing, pooling processing, and reverse training processing to obtain a sales forecast value of the marketing plan;

[0106] Based on the sales forecast value of the marketing plan, a popularity ranking table of marketing items is obtained, and based on the popularity ranking table of marketing items, the total number of raw materials for each marketing item is combined to perform raw material allocation processing for the marketing items;

[0107] Obtain the marketing rules of each marketing plan, optimize the marketing plan based on the raw material allocation results of the marketing items, obtain an optimized marketing plan, perform sales forecast processing on the optimized marketing plan, obtain the sales forecast processing results of the optimized marketing plan, and push the optimized marketing plan on the push website in descending order of sales.

[0108] It should be noted that, taking the catering industry as an example, the raw materials of marketing items may have something in common, for example, both dishes need to use the same ingredient, the same seasoning, etc., so the total number of raw materials for each marketing item is obtained. The marketing items may be the same in different marketing plans. The sales records of each marketing plan are predicted and processed to obtain the predicted value. According to the predicted value, the popularity ranking table of the marketing items is obtained. The popularity ranking table of the marketing items can judge the user's preference for the marketing items in the marketing plan, and based on the preference, the marketing items are redistributed to the raw materials, so that the number of popular marketing items is more, and the number of corresponding less popular marketing items is reduced, so the marketing plan will change accordingly. The marketing plan rule is not to change the principles of marketing benefits, activities, profits, etc., and then based on the distribution results of the raw materials of the marketing items, an optimized marketing plan is obtained. And the optimized marketing plan is diverse, and it is necessary to obtain the sales forecast value, and give priority to the push of the optimized marketing plan with a large sales forecast value. The present invention can redistribute the raw materials based on the total number of raw materials of each marketing item, obtain the optimized marketing plan and push it.

[0109] In addition, the marketing plan intelligent push method further includes the following steps:

[0110] Obtain the marketing plan order of the user on the push website, and extract information about the marketing items in the marketing plan order to obtain the marketing items ordered by the user;

[0111] Performing an appearance frequency analysis on the marketing items ordered by the user, defining the marketing items ordered by the user with an appearance frequency greater than a preset appearance frequency as the first category of marketing items, calculating the association value between all marketing items in the push website and the first category of marketing items by using the grey association method, and screening the marketing items with an association value greater than a preset value and marking them as the second category of marketing items;

[0112] On the purchase page of the second-category marketing items, obtain the user's rating of the second-category marketing items, filter out the second-category marketing items with a rating less than a preset value, and obtain qualified second-category marketing items;

[0113] Based on big data retrieval, a vocabulary sentiment intensity score table is constructed to obtain user evaluation sentences on qualified second-category marketing items, and the user evaluation sentences on qualified second-category marketing items are imported into the vocabulary sentiment intensity score table to perform sentiment intensity scoring of the evaluation sentences, and the qualified second-category marketing items are ranked by user sentiment intensity according to the scoring results to obtain a user sentiment intensity ranking table;

[0114] Based on historical data retrieval, the order quantities of qualified second-class marketing items at different times are obtained, and based on the order quantities of qualified second-class marketing items at different times, a ranking table of order quantities of qualified second-class marketing items at different times is generated;

[0115] Based on the attention mechanism, the user emotion intensity ranking table and the order quantity ranking table of qualified Class II marketing items at different times are fused to generate a qualified Class II commodity recommendation sequence table, and qualified Class II commodities are recommended to users based on the qualified Class II commodity recommendation sequence table.

[0116] It should be noted that there are multiple marketing items in the marketing plan, and users may have preferences for certain marketing items. For example, in a mobile phone marketing plan, the main purpose of the user is to buy a mobile phone, then the mobile phone is the user's first-class marketing product, and the mobile phone case, charging cable, plug and other marketing items are second-class marketing items. The second-class marketing items are determined by calculating the correlation with the first-class marketing items. If the score of the second-class marketing item is low, it is not suitable for recommendation to the user, so a qualified second-class marketing item is obtained. Different users have different preferences for qualified second-class marketing items. The vocabulary emotion intensity score table reflects the emotion intensity of different words. According to the vocabulary emotion intensity score table, the evaluation statements of qualified second-class marketing items can be scored to obtain a user emotion intensity ranking table. The user's desire to purchase qualified second-class products may be different at different times, so it is necessary to obtain the order volume placed at different times, and merge the two through the attention mechanism to obtain a qualified second-class product recommendation sequence table, and recommend qualified second-class products to users. The present invention can recommend suitable marketing items to users by performing correlation analysis and emotion intensity analysis on marketing items.

[0117] like Figure 3 As shown, the second aspect of the present invention further provides a marketing plan intelligent push system, the intelligent push system includes a memory 31 and a processor 32, the memory 31 stores a marketing plan intelligent push method, and when the marketing plan intelligent push method is executed by the processor 32, the following steps are implemented:

[0118] Acquire user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results;

[0119] Classify the content information of the marketing plan to obtain marketing plans of different categories, and combine the user information of new and old users to conduct feasibility analysis on the marketing plan;

[0120] In the push website, the marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence;

[0121] Analyze the consumption and feedback of the users being pushed, and optimize the push order of the marketing plans based on the analysis results.

[0122] The third aspect of the present invention further provides a computer-readable storage medium containing a marketing plan intelligent push method program, and when the marketing plan intelligent push method program is executed by a processor, the steps of any one of the marketing plan intelligent push methods are implemented.

[0123] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0124] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0125] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0126] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0127] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0128] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A marketing plan intelligent push method, characterized in that: The following steps are involved: Acquire user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results; Classify the content information of the marketing plan to obtain marketing plans of different categories, and combine the user information of new and old users to conduct feasibility analysis on the marketing plan; In the push website, the marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence; Analyze the consumption and feedback of users who receive push notifications, and optimize the order of marketing plan push based on the analysis results; The content information of the marketing plan is classified to obtain marketing plans of different categories, and the feasibility analysis of the marketing plan is performed in combination with the user information of new users and old users, specifically: In the push website, all marketing plans are obtained, and content information of each marketing plan is obtained respectively; Extract the features of the marketing plan content information through Fourier transform to obtain the marketing plan content feature data; Constructing fuzzy clusters and initializing the fuzzy clusters, wherein the fuzzy clusters contain fuzzy centers, and one fuzzy cluster represents the content of a marketing plan of one category; Using fuzzy clustering method and iterative calculation method, the Euclidean distance between the marketing plan content feature data and the fuzzy center is calculated, and based on the Euclidean distance between the marketing plan content feature data and the fuzzy center, the membership degree of the marketing plan content information and the fuzzy cluster is obtained; According to historical data retrieval, the marketing plan content information is sorted by importance to obtain a marketing plan content information sorting table, based on the marketing plan content information sorting table, the membership degree of the marketing plan content information and the fuzzy cluster is updated, and the marketing plans are classified to obtain marketing plans of different categories; Obtaining user information of new and old users, wherein the user information of new and old users includes user consumption upper limit, user consumption interest, and user gender and age; Importing the user information of the new and old users and different types of marketing plans into the CNN neural network to perform user feasibility analysis, obtain user feasibility analysis results, and screen suitable marketing plans and inappropriate marketing plans for different users based on the user feasibility analysis results; Among them, in the push website, the marketing plan push sequence is constructed in combination with the user's current status and the marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence, specifically: Set up a questionnaire on the push website login interface and guide users to fill in the questionnaire to get the user's current status; Acquire the number of marketing items currently in stock at the merchant in real time, and use the number of marketing items currently in stock at the merchant as a condition for the marketing plan; Based on the marketing plan conditions, suitable marketing plans for different users are analyzed. If the number of marketing items required in the suitable marketing plan is greater than the number of marketing items currently available in the merchant, the corresponding suitable marketing plan is defined as a marketing plan that cannot be pushed; In combination with the current state of the user, the remaining suitable marketing plans are predicted and processed in a convolutional neural network to obtain marketing plans that can be pushed; Obtaining the popularity of pushable marketing plans, where the popularity of pushable marketing plans includes the number of views and the number of consumptions of the pushable marketing plans, and constructing a pushable marketing plan popularity ranking table for different users based on the popularity of the pushable marketing plans; If the user is a new user, the pushable marketing plan is pushed to the new user based on the descending order of popularity in the pushable marketing plan popularity ranking table corresponding to the new user; If the user is an old user, a push sequence table of marketing plans that can be pushed to old users is constructed according to the consumption records of the old users, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to the old users.

2. According to the method for intelligently pushing a marketing plan as described in claim 1, it is characterized in that: The obtaining of user information, performing in-depth analysis on the user information, and classifying the users according to the in-depth analysis results are specifically as follows: Get the target website for marketing plan push, define it as the push website, and get the historical user information in the push website; Real-time acquisition of user information of online users in the current push website, defined as online user information, converting historical user information and online user information into feature data, respectively calculating the inner product and respective modulus of the historical user information and online user information, and calculating the cosine similarity of the historical user information and online user information based on the inner product and modulus; A cosine similarity interval is preset, and if the cosine similarity between the historical user information and the online user information is within the cosine similarity interval, the user represented by the corresponding online user information is marked as an old user; If the cosine similarity between the historical user information and the online user information is not within the cosine similarity interval, the user represented by the corresponding online user information is marked as a new user.

3. According to the marketing plan intelligent push method described in claim 1, it is characterized in that: If the user is an old user, a push sequence table of marketing plans that can be pushed to old users is constructed according to the consumption records of the old users, and based on the push sequence table of marketing plans that can be pushed to old users, the marketing plans that can be pushed are pushed to the old users, specifically: If the user is an old user, the consumption record of the old user in the push website is obtained, and based on the consumption record of the old user in the push website, the consumption frequency of the old user for the pushable marketing plan is obtained, and the consumption frequency of the old user for the pushable marketing plan is sorted based on the descending principle to construct the first push sequence table of the pushable marketing plan for the old user; Obtain pushable plans that old users have not consumed, sort the pushable plans that old users have not consumed based on the popularity ranking table of pushable marketing plans and based on the principle of descending popularity, and obtain a second push order table of pushable marketing plans for old users; Combined with the first push sequence table of marketing plans that can be pushed to old users and the second push sequence table of marketing plans that can be pushed to old users, a push sequence table of marketing plans that can be pushed to old users is constructed, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to old users.

4. According to the method for intelligently pushing a marketing plan as described in claim 1, it is characterized in that: The consumption and feedback of the users being pushed are analyzed, and the order of pushing the marketing plan is optimized according to the analysis results, specifically: After pushing the pushable marketing plan to the user, collect the consumption status and user feedback of the pushed user; Based on the pushed user consumption situation, obtain the consumption status of each marketing plan within a preset time, wherein the consumption status of each marketing plan includes the consumption quantity and consumption frequency of the marketing plan; Arrange the consumption quantity and consumption frequency of each marketing plan in descending order to obtain a marketing plan consumption quantity ranking table and a marketing plan consumption frequency ranking table, and compare and analyze the marketing plan consumption quantity ranking table and the marketing plan consumption frequency ranking table with the marketing plan push sequence table. If the descending sorting results of the consumption quantity and consumption frequency of the marketing plan are the same as the push sequence of the marketing plan, there is no need to optimize the push sequence of each marketing plan; If the descending sorting result of the consumption quantity and consumption frequency of the marketing plan is different from the push order of the marketing plan, then the deviation rate between the descending sorting result of the consumption quantity and consumption frequency of the marketing plan and the push order of the marketing plan is obtained, which is defined as the first deviation rate; Analyze the first deviation rate, and if the first deviation rate is less than a preset value, optimize the order of pushing the marketing plans to new and old users respectively based on the results of the descending sorting of the consumption quantity and the consumption frequency of the marketing plans; If the first deviation rate is greater than a preset value, the user feedback is analyzed, and the user feedback is the user's score on the marketing plan, and a qualified score is preset. If the user's score on the marketing plan is less than the qualified score, the corresponding marketing plan is eliminated and cannot be pushed; The marketing plans with scores greater than the qualified score are defined as qualified marketing plans, and the qualified marketing plans are sorted in descending order of scores to obtain a push order table of qualified marketing plans based on scores; According to the qualified marketing plan push sequence table based on scoring, the marketing plan push sequence for new and old users is optimized respectively.

5. A marketing plan intelligent push system, characterized in that: The intelligent push system includes a memory and a processor. The memory stores a marketing plan intelligent push method. When the marketing plan intelligent push method is executed by the processor, the following steps are implemented: Acquire user information, conduct in-depth analysis on the user information, and classify the users according to the in-depth analysis results; Classify the content information of the marketing plan to obtain marketing plans of different categories, and combine the user information of new and old users to conduct feasibility analysis on the marketing plan; In the push website, the marketing plan push sequence is constructed based on the user's current status and marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence; Analyze the consumption and feedback of users who receive push notifications, and optimize the order of marketing plan push based on the analysis results; The content information of the marketing plan is classified to obtain marketing plans of different categories, and the feasibility analysis of the marketing plan is performed in combination with the user information of new users and old users, specifically: In the push website, all marketing plans are obtained, and content information of each marketing plan is obtained respectively; Extract the features of the marketing plan content information through Fourier transform to obtain the marketing plan content feature data; Constructing fuzzy clusters and initializing the fuzzy clusters, wherein the fuzzy clusters contain fuzzy centers, and one fuzzy cluster represents the content of a marketing plan of one category; Using fuzzy clustering method and iterative calculation method, the Euclidean distance between the marketing plan content feature data and the fuzzy center is calculated, and based on the Euclidean distance between the marketing plan content feature data and the fuzzy center, the membership degree of the marketing plan content information and the fuzzy cluster is obtained; According to historical data retrieval, the marketing plan content information is sorted by importance to obtain a marketing plan content information sorting table, based on the marketing plan content information sorting table, the membership degree of the marketing plan content information and the fuzzy cluster is updated, and the marketing plans are classified to obtain marketing plans of different categories; Obtaining user information of new and old users, wherein the user information of new and old users includes user consumption upper limit, user consumption interest, and user gender and age; Importing the user information of the new and old users and different types of marketing plans into the CNN neural network to perform user feasibility analysis, obtain user feasibility analysis results, and screen suitable marketing plans and inappropriate marketing plans for different users based on the user feasibility analysis results; Among them, in the push website, the marketing plan push sequence is constructed in combination with the user's current status and the marketing plan conditions, and appropriate marketing plans are pushed to different users based on the marketing plan push sequence, specifically: Set up a questionnaire on the push website login interface and guide users to fill in the questionnaire to get the user's current status; Acquire the number of marketing items currently in stock at the merchant in real time, and use the number of marketing items currently in stock at the merchant as a condition for the marketing plan; Based on the marketing plan conditions, suitable marketing plans for different users are analyzed. If the number of marketing items required in the suitable marketing plan is greater than the number of marketing items currently available in the merchant, the corresponding suitable marketing plan is defined as a marketing plan that cannot be pushed; In combination with the current state of the user, the remaining suitable marketing plans are predicted and processed in a convolutional neural network to obtain marketing plans that can be pushed; Obtaining the popularity of pushable marketing plans, where the popularity of pushable marketing plans includes the number of views and the number of consumptions of the pushable marketing plans, and constructing a pushable marketing plan popularity ranking table for different users based on the popularity of the pushable marketing plans; If the user is a new user, the pushable marketing plan is pushed to the new user based on the descending order of popularity in the pushable marketing plan popularity ranking table corresponding to the new user; If the user is an old user, a push sequence table of marketing plans that can be pushed to old users is constructed according to the consumption records of the old users, and based on the push sequence table of marketing plans that can be pushed to old users, the pushable marketing plans are pushed to the old users.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a marketing plan intelligent push method program, and when the marketing plan intelligent push method program is executed by a processor, the steps of the marketing plan intelligent push method described in any one of claims 1-4 are implemented.

Images