Intelligent enterprise marketing data analysis system and method based on artificial intelligence

Abstract

The invention discloses an intelligent enterprise marketing data analysis system and method based on artificial intelligence, and belongs to the technical field of enterprise marketing, and the method comprises the steps: obtaining the historical enterprise marketing data of a target product, and setting the customer classification and the classification range of the corresponding customer classification according to the historical enterprise marketing data; enterprise marketing data of the target product in the analysis stage are obtained in real time, the enterprise marketing data are identified according to the customer classification and the classification range, customer classification information at the corresponding time in the analysis stage is obtained, a comprehensive monitoring graph is set according to the customer classification information, and the comprehensive monitoring graph is displayed to the user in real time; performing real-time identification on the comprehensive monitoring atlas, determining change reasons of customer classification, and performing marketing early warning according to the comprehensive monitoring atlas; and displaying the change reasons of the corresponding customer classifications to the user, and performing marketing suggestions according to the change reasons of the corresponding customer classifications.

Description

Technical Field

[0001] The present invention belongs to the field of enterprise marketing technology, and specifically is an enterprise marketing data intelligent analysis system and method based on artificial intelligence. Background Art

[0002] In today's highly competitive market environment, customers have become the core resource for the survival and development of enterprises. In-depth understanding of customer needs, behavior patterns and preferences, as well as accurate prediction of customers' future consumption trends, are essential for enterprises to formulate effective marketing strategies and enhance market competitiveness. However, with the advent of the big data era, the amount of data faced by enterprises is growing exponentially, and traditional marketing data analysis methods can no longer meet the needs of enterprises for deep insights and precision marketing. Customers are not only the final consumers of corporate products or services, but also the creators and disseminators of corporate brand value. Customer satisfaction and loyalty directly affect the company's market share, brand reputation and long-term development. Therefore, how to efficiently and accurately analyze customer data and tap into customer value has become the key to corporate marketing decisions.

[0003] Although there are a variety of marketing data analysis tools and methods on the market, they still have obvious shortcomings in processing massive data, mining deep customer insights, and realizing personalized marketing. For example, some tools can only provide basic data statistics and reports, lack in-depth data analysis capabilities, and it is difficult to adjust marketing strategies in a timely manner according to customer changes.

[0004] Based on this, the present invention provides an enterprise marketing data intelligent analysis system and method based on artificial intelligence. Summary of the invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an enterprise marketing data intelligent analysis system and method based on artificial intelligence.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An enterprise marketing data intelligent analysis system based on artificial intelligence, including a customer analysis module and a marketing analysis module;

[0008] The customer analysis module is used to perform customer analysis, obtain historical enterprise marketing data of the target product, and set customer categories and classification ranges of corresponding customer categories according to the historical enterprise marketing data;

[0009] Obtain the enterprise marketing data of the target product in the analysis stage in real time, identify the enterprise marketing data according to customer classification and classification scope, and obtain the customer classification information at the corresponding time within the analysis stage. The customer classification information includes the number of customers, the amount of consumption completed, and the potential consumption amount; set a comprehensive monitoring graph according to the customer classification information and display the comprehensive monitoring graph to the user in real time.

[0010] Further, set customer classifications and the classification scopes of the corresponding customer classifications according to historical enterprise marketing data, including:

[0011] Determine a classification index set according to a classification index library preset by the platform side. The classification index set is composed of corresponding classification indexes;

[0012] Perform feature recognition on the historical enterprise marketing data according to the classification index set to obtain a customer index feature set of the corresponding historical customers; remove duplicates from the historical customers according to the customer index feature set;

[0013] Calculate the classification vectors between the corresponding historical customers according to the customer index feature set. The classification vectors are set according to the similarity of the corresponding classification indexes between the historical customers;

[0014] Judge whether the corresponding historical customers belong to the same customer classification according to the classification vectors, classify the historical customers according to the judgment result to obtain the corresponding customer classification, and set the classification scope according to the customer classification.

[0015] Further, calculate the classification vectors between the corresponding historical customers according to the customer index feature set, including:

[0016] Quantify the customer index feature set to obtain an index quantization feature set, which is composed of index quantization values of the corresponding classification indexes;

[0017] Mark the classification indexes corresponding to the customer index feature set as i, where i = 1, 2,..., n, and n is the number of classification indexes; mark the index quantization values of the corresponding classification indexes as LK according to the index quantization feature set i ;

[0018] Calculate the positioning values of the corresponding historical customers according to the positioning formula. The positioning formula is:

[0019]

[0020] In the formula: DA is the positioning value; λ i represents the weight coefficient of the corresponding classification index;

[0021] Sort the historical customers in ascending order of the positioning values to obtain a first sequence;

[0022] Determine potential classified customers of corresponding historical customers according to the first sequence, and calculate the classification vector between the historical customers and the potential classified customers according to the metric quantization feature set.

[0023] Furthermore, set up a comprehensive monitoring graph according to the customer classification information, including:

[0024] Generate a classification monitoring graph according to the customer classification information. The classification monitoring graph includes a customer quantity change graph and a consumption monitoring graph. The customer quantity change graph includes a customer quantity change curve. The horizontal axis of the customer quantity change graph is time, and the vertical axis is the customer quantity; the consumption monitoring graph includes a completed consumption curve and a potential consumption curve. The horizontal axis of the consumption monitoring graph is time, and the vertical axis is the amount;

[0025] Calculate the customer value of the customer classification according to the classification monitoring graph. The customer value calculation formula is:

[0026]

[0027] In the formula: KH is the customer value; j represents the corresponding customer classification, KS j represents the customer quantity of the corresponding customer classification; β j represents the weight coefficient of the customer classification;

[0028] Generate a customer monitoring curve according to the customer value. The horizontal axis of the customer monitoring curve is time, and the vertical axis is the customer value;

[0029] Integrate the customer monitoring curve and the classification monitoring graph of each customer classification, and set up a comprehensive monitoring graph.

[0030] The marketing analysis module is used to conduct marketing analysis, identify the comprehensive monitoring graph in real time, determine the reasons for the changes in customer classification, and conduct marketing early warning according to the comprehensive monitoring graph; display the reasons for the changes in the corresponding customer classification to the user, and give marketing suggestions according to the reasons for the changes in the corresponding customer classification.

[0031] Furthermore, determining the reasons for the changes in customer classification includes:

[0032] Identify the classification monitoring graph of the customer classification according to the comprehensive monitoring graph; analyze the classification monitoring graph of the corresponding customer classification through a preset change analysis model to obtain the reasons for the changes in the customer classification.

[0033] Furthermore, conducting marketing early warning according to the comprehensive monitoring graph includes:

[0034] Identify the customer monitoring curve based on the comprehensive monitoring map, and obtain the marketing expectation curve of the user in the evaluation stage. The horizontal axis of the marketing expectation curve is time, and the vertical axis is the customer value;

[0035] Fit the customer monitoring curve and the marketing expectation curve to obtain the customer monitoring function and the marketing expectation function. Mark the customer monitoring function and the marketing expectation function as HA(t) and QA(t) respectively, where t is time;

[0036] Calculate the curve slope at the corresponding time according to the customer monitoring function, and mark the curve slope as k t ;

[0037] Establish an early warning analysis model. The expression of the early warning analysis model is:

[0038]

[0039] In the formula: [k t , HA(t), QA(t)] are input data; k x is the early warning slope; HQ is the early warning change value; the output data is the early warning analysis value YG[k t , HA(t), QA(t)], and the early warning analysis value is 1, 2, or 0;

[0040] Analyze through the early warning analysis model to obtain the corresponding early warning analysis value, and perform early warning processing according to the obtained early warning analysis value.

[0041] An intelligent analysis method for enterprise marketing data based on artificial intelligence, the method includes:

[0042] Obtain the historical enterprise marketing data of the target product, and set customer classifications and the classification ranges of the corresponding customer classifications according to the historical enterprise marketing data;

[0043] Obtain the enterprise marketing data of the target product in real time during the analysis stage, identify the enterprise marketing data according to the customer classifications and the classification ranges, obtain the customer classification information at the corresponding time during the analysis stage, set a comprehensive monitoring map according to the customer classification information, and display the comprehensive monitoring map to the user in real time;

[0044] Perform real-time identification on the comprehensive monitoring map, determine the reasons for changes in customer classifications, and perform marketing early warnings according to the comprehensive monitoring map; display the reasons for changes in the corresponding customer classifications to the user, and provide marketing suggestions according to the reasons for changes in the corresponding customer classifications.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] In the increasingly fierce market competition, whether an enterprise can quickly respond to market changes and accurately grasp customer needs often determines whether it can stand out. This system enables enterprises to gain the upper hand in the competition by providing real-time and comprehensive customer data analysis and marketing strategy suggestions. In addition, the intelligent and automated features of the system also enable enterprises to process massive amounts of data more efficiently, improving the overall operational efficiency and decision-making ability. With the powerful analysis ability of artificial intelligence, this system can quickly identify market trends and customer behavior patterns, providing data support for enterprises to formulate more accurate and efficient marketing strategies. By monitoring the feedback of marketing activities in real time, the system can also adjust strategies in a timely manner to ensure the maximum utilization of marketing resources. This not only reduces the marketing costs of enterprises, but also significantly improves the conversion rate and ROI of marketing activities; BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0048] Figure 1 It is a block diagram of the principle of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] The following will clearly and completely describe the technical solutions of the present invention in conjunction with the embodiments. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0050] As Figure 1 shown, an intelligent enterprise marketing data analysis system based on artificial intelligence includes a customer analysis module and a marketing analysis module;

[0051] The customer analysis module is used to conduct customer analysis, where the customer refers to the customer targeted by the user for selling corresponding products; obtain the historical enterprise marketing data of the target product, where the target product refers to the user product for which marketing analysis is carried out; set customer classifications and the corresponding classification ranges according to the historical enterprise marketing data;

[0052] Obtain the enterprise marketing data of the target product in the analysis stage in real time. The analysis stage is set by the user, such as monthly, quarterly, custom time period, promotion stage, etc.; identify the enterprise marketing data according to customer classification and classification scope, and obtain the customer classification information corresponding to the corresponding time within the analysis stage. The customer classification information includes the number of customers, customer information, completed consumption amount, potential consumption amount, etc. corresponding to the customer classification within the corresponding time analysis stage; the potential consumption amount refers to the amount of intended consumption but not consumed, and the potential consumption amount can be determined according to browsing, consulting, adding to the shopping cart, etc.

[0053] Generate a classification monitoring graph according to the customer classification information. The classification monitoring graph includes a customer number change graph and a consumption monitoring graph. The customer number change graph includes a customer number change curve. The horizontal axis of the customer number change graph is time, and the vertical axis is the number of customers; the consumption monitoring graph includes a completed consumption curve and a potential consumption curve. The horizontal axis of the consumption monitoring graph is time, and the vertical axis is the amount. Substitute the completed consumption amount and potential consumption amount corresponding to the corresponding time to generate the corresponding consumption curve and potential consumption curve.

[0054] Calculate the customer value according to the classification monitoring graphs of each customer classification. The customer value calculation formula is:

[0055]

[0056] In the formula: KH is the customer value; j represents the corresponding customer classification, KS j represents the number of customers of the corresponding customer classification; β j represents the weight coefficient of the customer classification, which is set by the user. Generally, it is set according to the consumption ability, share, etc. of the customer classification. For example, the weight coefficient is set according to the profit ratio of the historical customer classification.

[0057] Generate a customer monitoring curve according to the customer value. The horizontal axis of the customer monitoring curve is time, and the vertical axis is the customer value.

[0058] Integrate and set the customer monitoring curve and the classification monitoring graphs of each customer classification to form a comprehensive monitoring graph, that is, conduct a summary display.

[0059] In one embodiment, set the customer classification and the corresponding classification scope according to the historical enterprise marketing data. The existing classification methods can be used for classification to obtain several customer classifications, and mark the classification scope of the customer classification. For example, classification is carried out based on the RFM model, cluster analysis, etc.

[0060] In one embodiment, setting the customer classification and classification scope according to the historical enterprise marketing data includes:

[0061] The platform party presets a classification index library, that is, the platform party presets a number of classification indexes according to the customer classification requirements that users may have, such as relevant indexes such as gender, age, consumption frequency, consumption amount, etc.; a classification index library is established according to each classification index, and explanatory data for the corresponding classification indexes are set to facilitate users' understanding;

[0062] The user selects a number of classification indexes from the classification index library for classification to form a classification index set; it can be selected by the user himself, or the platform party can make intelligent recommendations according to the target product, and the user selects according to the recommended classification index set, or the classification index set can also be determined by other means;

[0063] Identify each classification index corresponding to the classification index set, perform feature recognition on the historical enterprise marketing data according to the classification index set, and obtain the customer index feature set of each historical customer. The customer index feature set is composed of customer information corresponding to the corresponding classification indexes; de-duplicate the historical customers according to the customer index feature set, that is, de-duplicate the historical customers with the same customer index feature set;

[0064] Calculate the similarity between the corresponding historical customers for each classification index according to the customer index feature set, and form a classification vector between the corresponding historical customers according to the similarity of each classification index. For example, if the similarities are 0.7, 0.2, 0.5, and 0.9 respectively, the classification vector is (0.7, 0.2, 0.5, 0.9);

[0065] Judge whether the corresponding historical customers belong to the same customer classification according to the classification vector, classify the corresponding historical customers according to the judgment result, obtain the corresponding customer classification, and set the classification range according to the customer classification.

[0066] In one embodiment, to calculate the similarity between the corresponding historical customers for each classification index according to the customer index feature set, the corresponding historical customers can be analyzed based on existing methods, such as pairwise calculation.

[0067] In one embodiment, because of the large number of historical customers, if calculated one by one, the efficiency is not high. Because when the customer index feature sets of the two are quite different, it can be determined that they do not belong to the same customer classification, and in order to reduce the amount of analysis, the calculation can be omitted. Therefore, in this embodiment, the following method is adopted for processing, including:

[0068] When the platform party sets the classification index library, it synchronously sets corresponding quantization methods for each classification index in the classification index library for quantifying non-numerical data into corresponding numerical values; quantify the customer index feature set according to the preset quantization method to obtain an index quantization feature set;

[0069] Mark the classification indicators corresponding to the customer metric feature set as i, where i = 1, 2, ……, n, and n is the number of classification indicators; mark the metric quantization values corresponding to the corresponding classification indicators according to the metric quantization feature set as LK i ;

[0070] Calculate the positioning value of the corresponding historical customer according to the positioning formula, and the positioning formula is:

[0071]

[0072] In the formula: DA is the positioning value; λ i represents the weight coefficient of the corresponding classification indicator, which can be set by the platform side or can be set and adjusted by the user;

[0073] Sort the historical customers in ascending or descending order of the positioning value to obtain the first sequence; it can be sorted in descending order or in ascending order;

[0074] Determine the potential classification customers of the corresponding historical customers according to the first sequence, that is, the historical customers within the range where the absolute value of the difference between the positioning values is not greater than the preset value. Generally, it is set according to the absolute value of the difference that is most likely to be classified into the same customer classification. For example, the preset value is set to be slightly larger than its value; calculate the classification vector between the corresponding historical customer and the potential classification customer according to the corresponding metric quantization feature set.

[0075] In one embodiment, to determine whether the corresponding historical customers belong to the same customer classification according to the classification vector, it can be judged based on existing judgment methods and classification criteria or requirements. For example, an intelligent judgment model is established based on neural networks such as CNN network or DNN network, and a corresponding training set is established and trained manually. The training set includes input data and output data. The input data is the classification standard and the classification vector, and the output data is the judgment result. Analyze and judge through the intelligent judgment model after successful training; judge based on the similarity of the classification vector, etc.

[0076] The marketing analysis module is used to conduct marketing analysis, identify the comprehensive monitoring map in real time, determine the reasons for the changes in customer classification, and issue marketing warnings according to the comprehensive monitoring map; display the reasons for the changes in the corresponding customer classification to the user and provide marketing suggestions according to the reasons for the changes in the corresponding customer classification.

[0077] In one embodiment, determining the reasons for the changes in customer classification includes:

[0078] Identify the classification monitoring map of the corresponding customer classification according to the comprehensive monitoring map; analyze the classification monitoring map of the corresponding customer classification through a preset change analysis model to determine the corresponding reasons for the changes;

[0079] The change analysis model mainly takes the changes in the classification monitoring map as the result, infers the reasons for the changes, and obtains the reasons for the changes, such as enterprise marketing promotion, market environment, product price adjustment, etc.; specifically, it is established based on current intelligent technologies, such as establishing a change analysis model based on machine learning, deep learning models, etc.

[0080] Exemplarily, the platform party sets the training data, and the training data includes the classification monitoring map, the collected data of relevant potential influencing factors, and the reasons for the changes.

[0081] Select machine learning or deep learning models suitable for processing time series data and multivariate variable relationships, such as recurrent neural network models such as LSTM (Long Short-Term Memory Network), GRU (Gated Recurrent Unit), or ensemble learning methods, etc.; design the model structure, including the input layer, hidden layer, and output layer, as well as the connection methods between the layers.

[0082] Train the model with the training data, continuously adjust the model parameters to minimize the prediction error; optimize the model using methods such as cross-validation and grid search to improve the prediction accuracy and stability of the model.

[0083] In one embodiment, marketing early warning is performed according to the comprehensive monitoring map, including:

[0084] Identify the customer monitoring curve according to the comprehensive monitoring map, and obtain the marketing expectation curve of the user in the evaluation stage. The horizontal axis of the marketing expectation curve is time, and the vertical axis is the customer value, that is, the expected customer value to be achieved.

[0085] Fit the customer monitoring curve and the marketing expectation curve to obtain the customer monitoring function and the marketing expectation function, and mark the customer monitoring function and the marketing expectation function as HA(t) and QA(t) respectively, where t is time.

[0086] Calculate the curve slope at the corresponding time according to the customer monitoring function, and mark this curve slope as k t ;

[0087] Establish an early warning analysis model, and the expression of the early warning analysis model is:

[0088]

[0089] In the formula: [k t , HA(t), QA(t)] are input data; k xis the warning slope, which is set according to the user's changing warning requirements. For example, if the customer value reduction rate is listed, the user selects the reduction rate interval that needs to be warned, and the warning slope is set according to the reduction rate interval; HQ is the warning change value, which is set according to the cumulative expected deviation allowed by the user; the output data is the warning analysis value YG[k t , HA(t), QA(t)], and the warning analysis value is 1, 2, or 0;

[0090] Analyze through the warning analysis model to obtain the corresponding warning analysis value, and perform warning processing according to the obtained warning analysis value. That is, when the warning analysis value is 0, no warning processing is performed; when the warning analysis value is 1 or 2, warning processing is performed according to the preset warning processing method.

[0091] In one embodiment, marketing warning can be performed according to the comprehensive monitoring map, and warning can be based on the existing warning methods, such as the customer value cannot be lower than the preset value.

[0092] In one embodiment, marketing suggestions are made according to the reasons for the changes in the corresponding customer classification. That is, improvements are made according to the suggestions for customer changes caused by adverse effects, or the user selects the reasons for the changes of the customer service, and targeted marketing suggestions are made according to the customer classification corresponding to the change reasons. For example, marketing is carried out according to the shopping characteristics of the customer classification; specifically, suggestion analysis is carried out based on the existing marketing suggestion methods.

[0093] An intelligent analysis method for enterprise marketing data based on artificial intelligence includes:

[0094] Obtain the historical enterprise marketing data of the target product, and set the customer classification and the classification range of the corresponding customer classification according to the historical enterprise marketing data;

[0095] Obtain the enterprise marketing data of the target product in real time during the analysis stage, identify the enterprise marketing data according to the customer classification and the classification range, obtain the customer classification information at the corresponding time during the analysis stage, set the comprehensive monitoring map according to the customer classification information, and display the comprehensive monitoring map to the user in real time;

[0096] Perform real-time identification on the comprehensive monitoring map, determine the reasons for the changes in the customer classification, and perform marketing warning according to the comprehensive monitoring map; display the reasons for the changes in the corresponding customer classification to the user, and make marketing suggestions according to the reasons for the changes in the corresponding customer classification.

[0097] The above formulas are all calculated by removing the dimension and taking their numerical values. The formula is obtained by collecting a large amount of data and performing software simulation to get a formula that is closest to the real situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained by simulating a large amount of data.

[0098] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent analysis system for enterprise marketing data based on artificial intelligence, characterized in that, It includes a customer analysis module and a marketing analysis module; The customer analysis module is used to conduct customer analysis, obtain historical enterprise marketing data of the target product, and set customer classifications and the classification ranges of the corresponding customer classifications according to the historical enterprise marketing data; It obtains the enterprise marketing data of the target product in the analysis stage in real time, identifies the enterprise marketing data according to the customer classifications and the classification ranges, and obtains the customer classification information at the corresponding time within the analysis stage. The customer classification information includes the number of customers, the amount of consumption completed, and the potential consumption amount; it sets a comprehensive monitoring graph according to the customer classification information and displays the comprehensive monitoring graph to the user in real time; The marketing analysis module is used to conduct marketing analysis, identify the comprehensive monitoring graph in real time, determine the reasons for changes in customer classifications, and issue marketing warnings according to the comprehensive monitoring graph; it displays the reasons for changes in the corresponding customer classifications to the user and provides marketing suggestions according to the reasons for changes in the corresponding customer classifications.

2. The intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 1, characterized in that, Setting customer classifications and the classification ranges of the corresponding customer classifications according to historical enterprise marketing data includes: Determining a classification index set according to a classification index library preset by the platform party. The classification index set consists of corresponding classification indexes; Conducting feature recognition on the historical enterprise marketing data according to the classification index set to obtain a customer index feature set of the corresponding historical customers; removing duplicates from the historical customers according to the customer index feature set; Calculating a classification vector between the corresponding historical customers according to the customer index feature set. The classification vector is set according to the similarity of the corresponding historical customers for the corresponding classification indexes; Judging whether the corresponding historical customers belong to the same customer classification according to the classification vector, classifying the historical customers according to the judgment result to obtain the corresponding customer classifications, and setting the classification ranges according to the customer classifications.

3. An intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 2, characterized in that Calculating a classification vector between the corresponding historical customers according to the customer index feature set includes: Quantifying the customer index feature set to obtain an index quantization feature set. The index quantization feature set consists of index quantization values of the corresponding classification indexes; Mark the classification index corresponding to the customer index feature set as i, where i = 1, 2, ……, n, and n is the number of classification indexes; mark the index quantization value of the corresponding classification index as LK according to the index quantization feature set i ; Calculating the positioning values of the corresponding historical customers according to a positioning formula. The positioning formula is: where: DA is the positioning value; λ i represents the weight coefficient of the corresponding classification index; Sorting the historical customers in ascending order of the positioning values to obtain a first sequence; Determining the potential classification customers of the corresponding historical customers according to the first sequence, and calculating the classification vector between the historical customers and the potential classification customers according to the index quantization feature set.

4. An intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 3, characterized in that, Setting a comprehensive monitoring graph according to the customer classification information includes: Generating a classification monitoring graph according to the customer classification information. The classification monitoring graph includes a customer number change graph and a consumption monitoring graph. The customer number change graph includes a customer number change curve. The horizontal axis of the customer number change graph is time, and the vertical axis is the number of customers; the consumption monitoring graph includes a completed consumption curve and a potential consumption curve. The horizontal axis of the consumption monitoring graph is time, and the vertical axis is the amount; Calculating the customer value of the customer classification according to the classification monitoring graph. The customer value calculation formula is: Where: KH is the customer value; j represents the corresponding customer classification, and KS j represents the number of customers in the corresponding customer classification; β j represents the weight coefficient of the customer classification; Generating a customer monitoring curve according to the customer value. The horizontal axis of the customer monitoring curve is time, and the vertical axis is the customer value; Integrate the customer monitoring curve and the classification monitoring atlas of each customer classification to set up a comprehensive monitoring atlas.

5. An intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 1, characterized in that, Determine the reasons for changes in customer classification, including: Identify the classification monitoring atlas of the customer classification according to the comprehensive monitoring atlas; analyze the classification monitoring atlas of the corresponding customer classification through a preset change analysis model to obtain the reasons for changes in the customer classification.

6. An intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 1, characterized in that Conduct marketing early warning according to the comprehensive monitoring atlas, including: Identify the customer monitoring curve according to the comprehensive monitoring atlas, obtain the marketing expectation curve of the user in the evaluation stage, where the horizontal axis of the marketing expectation curve is time and the vertical axis is the customer value; Fit the customer monitoring curve and the marketing expectation curve to obtain the customer monitoring function and the marketing expectation function, and mark the customer monitoring function and the marketing expectation function as HA(t) and QA(t) respectively, where t is time; Calculate the curve slope at the corresponding time according to the customer monitoring function, and the curve slope is marked as k t ; Establish an early warning analysis model, analyze through the early warning analysis model to obtain the corresponding early warning analysis value, and conduct early warning processing according to the obtained early warning analysis value.

7. An intelligent analysis system for enterprise marketing data based on artificial intelligence according to claim 6, characterized in that, The expression of the early warning analysis model is: Where: [k t , HA(t), QA(t)] are input data; k x is the warning slope; HQ is the warning change value; The output data is the warning analysis value YG[k t , HA(t), QA(t)], and the warning analysis value is 1, 2, or 0.

8. An intelligent analysis method for enterprise marketing data based on artificial intelligence, characterized in that, Applied to an intelligent analysis system for enterprise marketing data based on artificial intelligence as described in any one of claims 1 to 7, the method includes: Obtain the historical enterprise marketing data of the target product, and set customer classifications and the corresponding classification ranges of the customer classifications according to the historical enterprise marketing data; Obtain the enterprise marketing data of the target product in real time during the analysis stage, identify the enterprise marketing data according to the customer classification and the classification range to obtain the customer classification information at the corresponding time during the analysis stage, set up a comprehensive monitoring atlas according to the customer classification information, and display the comprehensive monitoring atlas to the user in real time; Identify the comprehensive monitoring atlas in real time, determine the reasons for changes in customer classification, and conduct marketing early warning according to the comprehensive monitoring atlas; display the reasons for changes in the corresponding customer classification to the user, and provide marketing suggestions according to the reasons for changes in the corresponding customer classification.

Images