Intelligent marketing management system based on Internet data analysis

Abstract

The invention provides an intelligent marketing management system based on Internet data analysis, and relates to the technical field of intelligent marketing, the intelligent marketing management system comprises an Internet information retrieval platform, a marketing management module and an Internet platform, the Internet information retrieval platform comprises an information acquisition unit, an information classification unit, a user evaluation unit and a flow statistics unit, the marketing management module comprises an information processing unit, a function establishment unit, a strategy selection unit and a delivery establishment unit; various marketing information is obtained from an internet platform through an internet information retrieval platform, an information processing unit and a function establishment unit in a marketing management module analyze and process the various marketing information, and a putting confirmation unit confirms whether to put commodity or service information needing to be marketed or not. According to the technical scheme, diversified marketing information can be obtained, reference is provided for marketing information release based on various marketing strategies and marketing effect information, different customer groups can be met, and marketing efficiency is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent marketing technology, and in particular to an intelligent marketing management system based on Internet data analysis. Background Art

[0002] Intelligent marketing is a new concept of innovative marketing that uses human creativity, innovation and creative wisdom to integrate advanced computer, network, mobile Internet, Internet of Things and other scientific and technological applications into the field of contemporary brand marketing with new thinking, new concepts, new methods and new tools.

[0003] Currently, intelligent marketing primarily focuses on meeting consumers' ever-changing, personalized, and fragmented needs. This new marketing model, built on Industry 4.0 (mobile internet, the Internet of Things, big data, and cloud computing), flexible production, and data supply chains, integrates consumers into the production and marketing processes of enterprises, achieving comprehensive business integration. Examples include Uber, Xiaomi, and Kute Intelligence / Magic Factory. Intelligent marketing is people-centered, based on network technology, with creativity at its core, content as its support, and marketing as its fundamental purpose. It offers personalized consumer marketing, achieving a perfect combination of brand and effectiveness. It grounds consumers' subjective perceptions of experience, context, perception, and aesthetics in the corporate ecosystem of cultural heritage, technological evolution, and commercial interests, ultimately integrating virtual and real-world innovative marketing concepts and technologies.

[0004] Intelligent customer management and accurate telephone marketing are provided. The intelligent management system for enterprise customer marketing can establish a center for enterprises that combines customer management and interactive marketing, integrate customer management, pre-sales, sales and after-sales, effectively distinguish and manage customers and conduct one-to-one marketing with customers, improve service marketing efficiency and maintain, develop and manage good customers.

[0005] The patent document with announcement number CN107392450B is a customer relationship management system based on cloud computing, which is connected to the embedded telephone marketing system of intelligent communication terminals through the Internet and can automatically and accurately identify customers, automatically

[0006] It can automatically make calls to marketing personnel, synchronize marketing results, display customer information on incoming calls, display last contact records on incoming calls, automatically assign to account managers, upload and retrieve call recordings, automatically remind customers, automatically compile marketing reports, and automatically update functions, thereby improving the company's work efficiency, facilitating the company's management of marketing representatives and customers, and reducing the time and cost of manual marketing.

[0007] However, in the process of implementing the above technical solution, it was found that the above technical solution had the following technical problems:

[0008] This cloud computing-based customer relationship management system connects to the embedded telemarketing system of intelligent communication terminals through the Internet to improve the work efficiency of the enterprise. However, in the actual application process, with the continuous development of Internet technology and the long application time of traditional telemarketing methods, the use of a single optimized telemarketing method is limited by customer rejection, blacklists, and relatively unified speech techniques. It is difficult to meet the needs of different customer groups, and the marketing efficiency is likely to be lower. Summary of the Invention

[0009] In order to overcome the shortcomings of existing cloud computing-based customer relationship management systems in actual applications, as Internet technology continues to develop and traditional telephone marketing methods have been used for a long time, the use of a single optimized telephone marketing method is limited by customer rejection, blacklists, and relatively uniform sales pitches, making it difficult to meet the needs of different customer groups and leading to increasingly low marketing efficiency. The present application embodiment provides an intelligent marketing management system based on Internet data analysis. By using functions to establish units and constructing a hierarchical analysis method, complex decision-making problems are hierarchically quantified to determine indicator weights. Based on various marketing information on the Internet platform, the Bayesian inference theorem is used to calculate the posterior distribution of parameters using prior distributions and sample data. That is, using existing fixed marketing information, it is calculated whether the same marketing strategy can achieve the same marketing effect as the existing case. In combination with the SWOT analysis matrix, the optimal marketing strategy can be determined and more precise delivery targets and delivery content can be selected.

[0010] At the same time, by using the SWOT analysis matrix and RFM model in the open channel, we can use the user churn prediction

[0011] The warning model is used to predict the attenuation of personnel flow, and with the help of data-driven marketing management, the customer acquisition ability of the marketing model is reversed, and the model and delivery plan that are easy to obtain positive returns are selected, which is conducive to selecting the best marketing strategy and marketing information for delivery among numerous marketing methods.

[0012] The technical solution adopted by the embodiment of the present application to solve the technical problem is:

[0013] An intelligent marketing management system based on Internet data analysis, including an Internet information retrieval platform, a marketing management module and an Internet platform, wherein the marketing management module is connected to the Internet information retrieval platform;

[0014] The Internet information retrieval platform includes an information acquisition unit, an information classification unit, a user evaluation unit, and a traffic statistics unit, and the marketing management module includes an information processing unit, a function establishment unit, a strategy selection unit, and a delivery establishment unit;

[0015] The information acquisition unit is used to acquire various types of marketing models on major Internet platforms, as well as user evaluation information allowed in the corresponding models and personnel flow information in open channels;

[0016] The information classification unit is used to classify the information acquired by the information acquisition unit according to different types of marketing models;

[0017] The user evaluation unit is used to identify the actual content of user evaluations in various types of marketing models, and classify the evaluation content according to positive and negative methods within the framework of the corresponding type of marketing model;

[0018] The traffic statistics unit is used to count the personnel traffic on a certain network platform within a unit time and infer the personnel traffic on the network platform within a period of time;

[0019] The information processing unit is used to organize the various types of marketing models obtained by the information acquisition unit on major Internet platforms, as well as the user evaluation information allowed in the corresponding model and the personnel flow information in the corresponding model;

[0020] The function establishment unit is used to establish a function to obtain various types of marketing models on major Internet platforms, as well as user evaluation information and personnel flow information allowed in the corresponding models, to analyze the target groups and advantages of various types of marketing models;

[0021] The strategy selection unit is used to select the corresponding marketing model required for marketing according to the target groups and advantages of the various marketing models displayed by the function establishment unit;

[0022] The delivery confirmation unit is used to deliver the commodity information and service information to be sold in an optimal marketing mode and expose them to the public.

[0023] In one possible implementation, information acquisition logic is set for the information acquisition unit, and these logics include (picture ∧ address ∧ price ∧ phone number), (video ∧ link ∧ price) and platform. The platform logic in the information acquisition logic also includes (product information ∧ evaluation information).

[0024] In a possible implementation, when the user evaluation unit categorizes the content into positive and negative evaluations, it records positive and negative user evaluations as (0, 1), respectively, while more neutral user evaluations are ignored.

[0025] In a possible implementation, the function establishment unit quantifies the complex decision problem in layers and determines the indicator weights by constructing a hierarchical analysis method. The specific formula of the hierarchical analysis method is as follows:

[0026] Construct judgment matrix:

[0027] A = [a ij ] n×n

[0028] (a ij indicates the relative importance of index i to j, 1-9 scale method)

[0029] Calculate weight vector:

[0030]

[0031] Consistency check:

[0032]

[0033] (through consistency check).

[0034] In one possible implementation, on the basis of the analytic hierarchy process which quantifies complex decision-making problems in layers, the Bayesian inference formula is used to update the probability of event occurrence, as follows:

[0035]

[0036] Where P(A|B) is the posterior probability (updated probability) of event A occurring after observing evidence B, P(A) is the prior probability of event A occurring (initial probability without considering evidence B), P(B|A) is the likelihood of event B occurring under the condition that event A occurs, and P(B) is the marginal probability of event B occurring (total probability), which can be calculated by the total probability formula:

[0037] P(B) = ∑ i P(B|A i )·P(A i )({A i});

[0038] is a partition of the sample space (i.e. mutually exclusive and exhaustive of all possible events).

[0039] In one possible implementation, the function establishment unit comprehensively evaluates the advantages (S) and disadvantages (W) of various types of marketing modes, external opportunities (O) and threats (T) by establishing a SWOT analysis matrix, and its expression is:

[0040] Strategy priority = f (S-W, O-T);

[0041] Where the advantage-disadvantage difference (internal competitiveness) and the opportunity-threat difference (external environment) cross-position the strategy direction.

[0042] In a possible implementation, the function building unit further establishes an RFM model to quantify customer value based on recent purchases (R), purchase frequency (F), and purchase amount (M), as follows:

[0043] Customer value score = ∝ × R + β × F + γ × M

[0044] (∝, β, and γ are the weights of each dimension, which are set according to business objectives);

[0045] Among them, reverse inference through customer value score status and quantity allows users to evaluate the feasibility of relevant patterns in information.

[0046] In one possible implementation, the strategy selection unit selects more accurate delivery targets and delivery content based on the hierarchical analysis method, Bayesian inference formula and SWOT analysis matrix logic (picture ∧ address ∧ price ∧ phone) and (video ∧ link ∧ price), and selects models and delivery plans that are easy to obtain positive returns from relevant models that allow users to evaluate the existence of information based on the SWOT analysis matrix and RFM model.

[0047] In a possible implementation, the personnel flow information in the open channel is used to predict the personnel flow attenuation state through a user churn warning model, as follows:

[0048]

[0049] The customer retention income after the early warning system is deployed is calculated using the ROI function.

[0050] In one possible implementation, the Internet information retrieval platform obtains various marketing information from the Internet platform through the information acquisition unit, and organizes the various marketing information based on the information classification unit, the user evaluation unit and the traffic statistics unit. The information processing unit and the function establishment unit in the marketing management module analyze and process the various marketing information, provide selection criteria for the strategy selection unit, and finally the delivery confirmation unit confirms whether to deliver the required marketing goods or service information.

[0051] The beneficial effects of this application are:

[0052] First, in this solution, various marketing information is obtained from the Internet platform based on the Internet information retrieval platform. The information processing unit and function establishment unit in the marketing management module analyze and process the various marketing information, provide selection criteria for the strategy selection unit, and finally the delivery confirmation unit confirms whether to deliver the required product or service information. This can achieve the acquisition of diversified marketing information and provide reference for the delivery of marketing information based on various marketing strategies and marketing effect information, which is conducive to satisfying different customer groups.

[0053] Second, this solution uses a hierarchical analysis method (AHP) to quantify complex decision-making problems in layers and determine indicator weights by building units with the help of functions. Using the Bayesian inference theorem, we use the prior distribution and sample data to calculate the posterior distribution of parameters based on various marketing information on the internet platform. This means that we can use existing fixed marketing information to calculate whether the same marketing strategy can achieve the same marketing results as the existing case. Combined with the SWOT analysis matrix, we can determine the optimal marketing strategy and select more precise delivery targets and content.

[0054] Third, in this plan, by using the SWOT analysis matrix and RFM model in the open channel, the user churn warning model is used to predict the attenuation status of personnel traffic. With the help of data-driven marketing management, the customer acquisition ability of the marketing model is reversed, and the model and delivery plan that are easy to obtain positive returns are selected, which is conducive to selecting the best marketing strategy and marketing information for delivery among numerous marketing methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a schematic diagram of the overall framework of the intelligent marketing management system based on Internet data analysis of the present invention;

[0056] Figure 2 This is a schematic diagram of the workflow of the intelligent marketing management system based on Internet data analysis of the present invention. DETAILED DESCRIPTION

[0057] The technical solution in the embodiments of the present application is to solve the problems of the above-mentioned background technology, and the overall idea is as follows:

[0058] Example 1:

[0059] This embodiment introduces the specific structure of the intelligent marketing management system based on Internet data analysis. Figure 1-Figure 2 As shown, it includes an Internet information retrieval platform, a marketing management module connected to the Internet information retrieval platform, and an Internet platform. The Internet information retrieval platform includes an information acquisition unit, an information classification unit, a user evaluation unit, and a traffic statistics unit. The marketing management module includes an information processing unit, a function establishment unit, a strategy selection unit, and a delivery establishment unit.

[0060] Among them, the information acquisition unit is used to obtain various types of marketing models on major Internet platforms, as well as user evaluation information allowed in the corresponding models and personnel flow information in open channels; the information classification unit is used to classify the information obtained by the information acquisition unit according to different types of marketing models; the user evaluation unit is used to identify the actual content of user evaluations in various types of marketing models, and classify the evaluation content in a positive or negative manner within the framework of the corresponding type of marketing model; the traffic statistics unit is used to count the personnel flow on a certain network platform within a unit time and infer the personnel flow on the network platform within a period of time;

[0061] At the same time, information acquisition logic is set for the information acquisition unit. This logic includes (picture ∧ address ∧ price ∧ phone number), (video ∧ link ∧ price) and platform. The platform logic in the information acquisition logic also includes (product information ∧ evaluation information). Specifically, it can be expressed as searching for corresponding web pages and platforms through logical information such as pictures, addresses, prices and phone numbers; searching for corresponding web pages and platforms through logical information such as videos, links, prices; and searching for various platforms that release product information, such as e-commerce platforms, shopping platforms, live broadcast platforms, etc., through platform logical information.

[0062] Secondly, the information processing unit is used to organize the various types of marketing models obtained by the information acquisition unit on major Internet platforms, as well as the user evaluation information and personnel flow information allowed in the corresponding models; the function establishment unit is used to establish a function to analyze the various types of marketing models obtained on major Internet platforms, as well as the user evaluation information and personnel flow information allowed in the corresponding models, and display the target groups and advantages of various types of marketing models; the strategy selection unit is used to select the corresponding marketing model required for marketing according to the target groups and advantages of various marketing models displayed by the function establishment unit; the delivery confirmation unit is used to deliver the product information and service information to be sold in the optimal marketing model and expose them to the public;

[0063] At the same time, when the user evaluation unit categorizes the content into positive and negative ways, it records positive and negative user evaluations as (0, 1) respectively, while the more neutral user evaluations are ignored here, which is convenient for the marketing management module to process information and reduce the pressure;

[0064] The above design is based on the Internet information retrieval platform, various marketing information is acquired from the Internet platform by the information acquisition unit, and based on the information classification unit, the user evaluation unit and the flow statistics unit, the various marketing information is sorted out, and the information processing unit and the function establishment unit in the marketing management module analyze and process the various marketing information, provide selection criteria for the strategy selection unit, and finally confirm whether to put the required marketing goods or service information by the putting confirmation unit. The acquisition work of diversified marketing information can be realized, and based on various marketing strategies and marketing effect information, the marketing information can be provided for reference for the marketing information, which is beneficial to meet the needs of different customer groups and ensure the marketing efficiency.

[0065] Embodiment 2:

[0066] Based on embodiment 1, this embodiment introduces the detailed structure of the function establishment unit. As shown in Figure 2 The function establishment unit quantifies the complex decision-making problem by constructing the analytic hierarchy process, determines the index weight, and the specific formula of the analytic hierarchy process is as follows:

[0067] Construct the judgment matrix:

[0068] A=[a ij ] n×n (a ij represents the relative importance of index i to j, and the 1-9 scale method);

[0069] Calculate the weight vector:

[0070]

[0071] Consistency check:

[0072] (through the consistency check);

[0073] In the process of using the analytic hierarchy process, first, a hierarchical structure model is established, and the problem is divided into three levels of target layer (the final goal of decision-making, such as selecting the optimal marketing scheme), criterion layer (key factors or evaluation indexes affecting target realization, such as cost, benefit and risk, etc.), and scheme layer (specific schemes to be evaluated);

[0074] Then, the relative importance of the elements in each level is compared by two-by-two comparison, and the judgment matrix is constructed, such as the 1-9 scale method, which quantifies subjective judgment;

[0075] Then, the criterion layer judgment matrix (selection of the final marketing strategy) is constructed, the importance of the criterion layer indexes (picture, price, geographical position, click volume) is compared, and the matrix is obtained;

[0076] Then, the weight vector of the judgment matrix is calculated to reflect the relative importance of each element;

[0077] Further, the judgment matrix needs to satisfy logical consistency to avoid the contradiction that "A is more important than B, B is more important than C, and C is more important than A";

[0078] The test steps are as follows:

[0079] 1. Calculate the maximum eigenvalue λ max : obtained by matrix operation;

[0080] 2. Calculate the consistency index CI: (n is the order of the matrix);

[0081] 3. Find the average random consistency index RI;

[0082] 4. Calculate the consistency ratio CR: (if CR<0.1, the judgment matrix consistency is acceptable; otherwise, the judgment matrix needs to be adjusted);

[0083] Finally, the weights of each layer are passed layer by layer, the comprehensive weight of the scheme layer relative to the target layer is calculated, and the optimal scheme is determined according to the weight order.

[0084] Secondly, on the basis of the analytic hierarchy process dividing the complex decision-making problem into layers and quantifying, the Bayesian inference formula is used to update the event occurrence probability, which is as follows:

[0085]

[0086] P(A│B) is the posterior probability (updated probability) of event A occurring after observing evidence B, P(A) is the prior probability of event A occurring (initial probability without considering evidence B), P(B│A) is the likelihood of event B occurring under the condition that event A occurs, and P(B) is the marginal probability of event B occurring (total probability), which can be calculated by the total probability formula:

[0087] P(B)=∑ i P(B│A i )·P(A i );

[0088] ({A i} is a partition of the sample space (i.e. mutually exclusive and exhaustive of all possible events));

[0089] By using the Bayesian inference formula to update the event occurrence probability, when filtering the marketing information that cannot achieve the target marketing effect, the posterior distribution of the parameter is calculated through the prior distribution and sample data, that is, whether the same marketing strategy can achieve the same marketing effect as the existing case is calculated by using the existing various marketing information;

[0090] Further, the function establishing unit comprehensively evaluates the advantages (S) and disadvantages (W) of various types of marketing modes, external opportunities (O) and threats (T) by establishing a SWOT analysis matrix, and its expression is:

[0091] Strategy priority = f (S-W, O-T) ;

[0092] Wherein, the advantage-disadvantage difference (internal competitiveness) and the opportunity-threat difference (external environment) cross the positioning strategy direction;

[0093] At the same time, the function establishing unit also quantifies customer value by establishing an RFM model, with recent purchase (R), purchase frequency (F), and purchase amount (M), as follows:

[0094] Customer value score = a x R + b x F + g x M

[0095] (a, b, g are dimension weights, which are set according to business objectives);

[0096] Wherein, the customer value score state and quantity are used to infer the feasibility of the relevant mode of user evaluation information existing, and the marketing party can use data-driven marketing management to convert from "traffic operation" to "user value operation" to infer the customer acquisition ability of the marketing mode;

[0097] Further, the personnel flow information in the open channel is predicted by a user loss early warning model to predict the personnel flow attenuation state, as follows:

[0098]

[0099] And the customer retention revenue after deployment of the early warning system is calculated by the ROI function;

[0100] Specifically, first prepare the data and clean the data:

[0101] Data preparation includes user basic data in the fixed platform (registration information such as age, region, consumption ability, etc.; account status such as paid / free, membership level), behavior data (login frequency, usage time, function module access record, transaction record), interaction data (customer service communication record, complaint feedback, marketing activity participation), and external data (industry trends, competitor dynamics, holiday impact, etc.):

[0102] Data cleaning includes processing missing values, abnormal values, unifying data formats, and removing noise data;

[0103] Then, perform feature work such as extracting key features, feature conversion, and feature selection:

[0104] Key features extracted include activity features (number of logins in the past 30 days, average daily usage time, depth of functional module usage), consumption features (last consumption time (R), consumption frequency (F), consumption amount (M), coupon usage rate), interaction features (customer service response time, number of complaints, message open rate) and life cycle features (user registration time, number of days since last active)

[0105] Feature transformation, i.e. one-hot encoding of categorical variables (such as region, user type) and standardization or binning of continuous variables;

[0106] Feature screening is to eliminate redundant features through correlation analysis (such as Pearson coefficient), mutual information method or model importance ranking;

[0107] As mentioned above, the strategy selection unit can select more accurate delivery targets and delivery content from the logic of (image ∧ address ∧ price ∧ phone number) and (video ∧ link ∧ price) based on the hierarchical analysis method, Bayesian reasoning formula and SWOT analysis matrix;

[0108] At the same time, based on the SWOT analysis matrix and RFM model, you can choose models and delivery plans that are easy to obtain positive returns from relevant models that allow users to evaluate the existence of information, which is conducive to selecting the best marketing strategies and marketing information for delivery among numerous marketing methods.

[0109] The above design uses functions to establish units and construct a hierarchical analysis method to quantify complex decision-making problems in layers and determine indicator weights. Based on the Bayesian inference theorem, based on various marketing information on the Internet platform, after filtering out marketing information that cannot achieve the target marketing effect, the prior distribution and sample data are used to calculate the posterior distribution of parameters. In other words, using existing fixed marketing information, it is calculated whether the same marketing strategy can achieve the same marketing effect as the existing case. Combined with the SWOT analysis matrix, the optimal marketing strategy can be determined and more precise marketing targets and content can be selected.

[0110] At the same time, with the help of the SWOT analysis matrix and RFM model in the open channel, the user churn warning model is used to predict the attenuation status of personnel traffic. With the help of data-driven marketing management and the transformation from "traffic operation" to "user value operation", the customer acquisition ability of the marketing model is reversed, and the model and delivery plan that are easy to obtain positive returns are selected, which is conducive to selecting the best marketing strategy and marketing information for delivery among numerous marketing methods.

[0111] Finally, it should be noted that the above embodiments are merely examples for the purpose of illustrating the present invention and are not intended to limit the embodiments. Those skilled in the art will readily appreciate that other variations or modifications based on the above description are possible. It is not necessary and impossible to provide an exhaustive list of all embodiments. However, obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. Intelligent marketing management system based on Internet data analysis, characterized by: include: Internet information retrieval platform; A marketing management module that is connected to an Internet information retrieval platform; Internet platforms; The Internet information retrieval platform includes an information acquisition unit, an information classification unit, a user evaluation unit, and a traffic statistics unit, and the marketing management module includes an information processing unit, a function establishment unit, a strategy selection unit, and a delivery establishment unit; The information acquisition unit is used to acquire various types of marketing models on major Internet platforms, as well as user evaluation information allowed in the corresponding models and personnel flow information in open channels; The information classification unit is used to classify the information acquired by the information acquisition unit according to different types of marketing models; The user evaluation unit is used to identify the actual content of user evaluations in various types of marketing models, and classify the evaluation content according to positive and negative methods within the framework of the corresponding type of marketing model; The traffic statistics unit is used to count the personnel traffic on a certain network platform within a unit time and infer the personnel traffic on the network platform within a period of time; The information processing unit is used to organize the various types of marketing models obtained by the information acquisition unit on major Internet platforms, as well as the user evaluation information allowed in the corresponding model and the personnel flow information in the corresponding model; The function establishment unit is used to establish a function to obtain various types of marketing models on major Internet platforms, as well as user evaluation information and personnel flow information allowed in the corresponding models, to analyze the target groups and advantages of various types of marketing models; The strategy selection unit is used to select the corresponding marketing model required for marketing according to the target groups and advantages of the various marketing models displayed by the function establishment unit; The delivery confirmation unit is used to deliver the commodity information and service information to be sold in an optimal marketing mode and expose them to the public.

2. The intelligent marketing management system based on Internet data analysis according to claim 1, characterized in that: Information acquisition logic is set for the information acquisition unit. These logics include (picture ∧ address ∧ price ∧ phone number), (video ∧ link ∧ price) and platform. The platform logic in the information acquisition logic also includes (product information ∧ evaluation information).

3. The intelligent marketing management system based on Internet data analysis according to claim 1, characterized in that: When the user evaluation unit categorizes the content into positive and negative evaluations, it records positive and negative user evaluations as (0, 1) respectively, while more neutral user evaluations are ignored.

4. The intelligent marketing management system based on Internet data analysis according to claim 1, characterized in that: The function establishment unit quantifies the complex decision-making problem by layers and determines the indicator weights by constructing the hierarchical analysis method. The specific formula of the hierarchical analysis method is as follows: Construct a judgment matrix: A=[a ij ] n×n ; (a ij Indicates the relative importance of indicator i to j, 1-9 scale) Calculate the weight vector: Consistency check: (Pass consistency check).

5. The intelligent marketing management system based on Internet data analysis according to claim 4, characterized in that: Based on the hierarchical quantification of complex decision-making problems using the AHP method, the Bayesian inference formula is used to update the probability of an event, as follows: Where P(A│B) is the posterior probability (updated probability) of event A occurring after observing evidence B, P(A) is the prior probability of event A occurring (the initial probability without considering evidence B), P(B│A) is the likelihood of event B occurring given event A, and P(B) is the marginal probability (total probability) of event B occurring, which can be calculated using the total probability formula: P(B)=∑ i P(B│A i )·P(A i )({A i }; It is a partition of the sample space (i.e., mutually exclusive and exhaustive of all possible events).

6. The intelligent marketing management system based on Internet data analysis according to claim 5, characterized in that: The function establishment unit comprehensively evaluates the strengths (S) and weaknesses (W), external opportunities (O) and threats (T) of various types of marketing models by establishing a SWOT analysis matrix, the expression of which is: Strategy priority = f(SW, OT); Among them, the advantage-disadvantage gap (internal competitiveness) and the opportunity-threat gap (external environment) cross-position the strategic direction.

7. The intelligent marketing management system based on Internet data analysis according to claim 6, characterized in that: The function building unit also quantifies customer value by building an RFM model based on recent purchases (R), purchase frequency (F), and purchase amount (M), as follows: Customer value score = ∝ × R + β × F + γ × M; (∝, β, and γ are the weights of each dimension, which are set according to business objectives); Among them, reverse inference through customer value score status and quantity allows users to evaluate the feasibility of relevant patterns in information.

8. The intelligent marketing management system based on Internet data analysis according to claim 7, characterized in that: The strategy selection unit selects more accurate delivery targets and delivery content from the logics of (picture ∧ address ∧ price ∧ phone) and (video ∧ link ∧ price) according to the hierarchical analysis method, the Bayesian reasoning formula and the SWOT analysis matrix, and selects models and delivery plans that are easy to obtain positive returns from relevant models that allow users to evaluate the existence of information according to the SWOT analysis matrix and the RFM model.

9. The intelligent marketing management system based on Internet data analysis according to claim 1, characterized in that: The personnel flow information in the open channel is used to predict the personnel flow attenuation state through the user churn warning model, as follows: The customer retention income after the early warning system is deployed is calculated using the ROI function.

10. The intelligent marketing management system based on Internet data analysis according to claim 1, characterized in that: The Internet information retrieval platform obtains various marketing information from the Internet platform through the information acquisition unit, and organizes the various marketing information based on the information classification unit, the user evaluation unit and the traffic statistics unit. The information processing unit and the function establishment unit in the marketing management module analyze and process the various marketing information, provide selection criteria for the strategy selection unit, and finally the delivery confirmation unit confirms whether to deliver the required marketing product or service information.

Images