A big data driven intelligent marketing prediction method

Abstract

The application relates to the technical field of marketing prediction, in particular to a big data driven intelligent marketing prediction method, which comprises the following steps: determining a prediction mode for a user based on the comprehensive evaluation value of the historical behavior data of a plurality of users; determining the appeal degree of the user based on the feature complexity characteristic parameter of a keyword feature; determining the classification of a demand product based on the dimension number of the acquired demand product information data; determining the eligibility of the extracted demand product parameter quantity based on the attention degree of the appeal degree user; determining the eligibility of the extracted core function feature quantity of the demand product based on the feature discrete coefficient of the degree user; acquiring real-time behavior data of a plurality of users; determining the eligibility of the prediction mode based on the prediction accuracy of the prediction mode; and determining the increase of the keyword feature extraction frequency or the increase of the demand product information data acquisition frequency according to the ratio of the prediction accuracy to a prediction accuracy threshold value, so that the accuracy of marketing prediction is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of marketing prediction, and particularly relates to a big data driven intelligent marketing prediction method. BACKGROUND

[0002] With the increasing complexity of consumer behavior and the massive growth of market data, traditional methods rely on a small amount of historical sales data, simple market research and experience judgment, which have significant limitations. Historical data cannot keep up with the current consumer demand and new market trends. The research sample is limited and subjective. Experience judgment lacks objective basis and is difficult to cope with complex market patterns.

[0003] Chinese patent application publication No. CN118780845A discloses a method for generating accurate marketing strategies based on big data, comprising the following steps: S1: collecting real-time data of target users, processing the real-time data of target users, and determining target user feature data; S2: constructing an accurate marketing strategy generation model, testing and optimizing the accurate marketing strategy generation model, and determining the best accurate marketing strategy generation model; S3: analyzing and predicting target user feature data based on the best accurate marketing strategy generation model, and generating an accurate marketing strategy. However, the prior art has the following problems: the prior art does not fully analyze user behavior data in depth, resulting in low accuracy of marketing prediction. SUMMARY

[0004] Therefore, the present application provides a big data driven intelligent marketing prediction method to overcome the problem of insufficient depth analysis of user behavior data in the prior art, resulting in low accuracy of marketing prediction.

[0005] To achieve the above purpose, the present application provides a big data driven intelligent marketing prediction method, comprising:

[0006] Obtain historical behavior data of a plurality of users, and determine a prediction mode for the users based on a comprehensive evaluation value of the historical behavior data;

[0007] Obtain chat record data of the user, extract keyword features related to user demand in the chat record data, and determine a user demand degree based on a feature complexity representation parameter of the keyword features, including a low demand degree or a high demand degree;

[0008] Under the condition of obtaining user demand product information data, determine whether the demand product is a single demand product or a complex demand product based on a dimension number of the demand product information data;

[0009] Under the condition of determining that the demand product is a single demand product, extract demand product parameters, and determine the eligibility of the number of extracted demand product parameters based on the attention degree of the user with a low demand degree or a high demand degree;

[0010] under the condition of determining that the demand product is a complex demand product, extracting core functional features of the demand product, and determining the eligibility of the number of core functional features of the demand product based on the feature discrete coefficient of the user with low or high appeal degree;

[0011] obtaining real-time behavior data of a plurality of users, determining the eligibility of the corresponding prediction mode based on the prediction accuracy of the prediction mode of the user, and determining the increase of the keyword feature extraction frequency or the increase of the demand product information data collection frequency according to the ratio of the prediction accuracy to the prediction accuracy threshold.

[0012] Further, the number of dimensions of the keyword features is also adjusted by the first preset keyword dimension adjustment coefficient or the second preset keyword dimension adjustment coefficient based on the comparison result of the correlation degree determined based on the purchase decision change rate of the purchase decision data and the marketing strategy change rate of the marketing strategy data and the preset correlation degree.

[0013] Further, the first prediction mode is determined for the user based on the comparison result that the comprehensive evaluation value of the historical behavior data is less than or equal to the preset comprehensive evaluation value.

[0014] Further, the second prediction mode is determined for the user based on the comparison result that the comprehensive evaluation value of the historical behavior data is greater than the preset comprehensive evaluation value.

[0015] Further, the appeal degree of the user is determined to be low based on the comparison result that the feature complexity representation parameter of the extracted keyword feature is less than or equal to the preset feature complexity representation parameter.

[0016] Further, the appeal degree of the user is determined to be high based on the comparison result that the feature complexity representation parameter of the extracted keyword feature is greater than the preset feature complexity representation parameter.

[0017] Further, the demand product is determined to be a single demand product based on the comparison result that the number of dimensions of the demand product information data is less than or equal to the preset dimension number, and the number of extracted demand product parameters is determined to be eligible based on the comparison result that the attention degree of the corresponding appeal degree user is greater than the preset attention degree.

[0018] Further, the demand product is determined to be a complex demand product based on the comparison result that the number of dimensions of the demand product information data is greater than the preset dimension number, and the number of core functional features of the demand product is determined to be eligible based on the comparison result that the feature discrete coefficient of the corresponding appeal degree user is greater than the preset feature discrete coefficient.

[0019] Further, the corresponding prediction mode is determined to be unqualified based on a comparison result of a prediction accuracy of the corresponding prediction mode being less than or equal to a prediction accuracy threshold, and the keyword feature extraction frequency is increased by a preset extraction frequency adjustment coefficient according to a comparison result of the prediction accuracy and the prediction accuracy threshold being less than or equal to a preset ratio.

[0020] Further, the demand product information data acquisition frequency is increased by a preset acquisition frequency adjustment coefficient according to a comparison result of the prediction accuracy and the prediction accuracy threshold being greater than the preset ratio.

[0021] Compared with the prior art, the present application has the beneficial effects that the present application determines the prediction mode by acquiring user historical behavior data, judges the user demand degree by extracting the chat record keyword feature, distinguishes the product type according to the dimension number of the demand product information data, determines the eligibility of the extraction parameter or the number of functional features according to the user demand for different product types, verifies the prediction mode according to the real-time behavior data and determines the data processing frequency adjustment strategy, accurately grasps the user demand, and the product information processing is more targeted, thereby improving the efficiency of marketing prediction and the accuracy of marketing prediction.

[0022] Further, the present application determines to adopt the first or second prediction mode by the comprehensive evaluation value of the user historical behavior data, acquires the chat record data and extracts the keyword feature in the determined prediction mode, and determines the user demand degree according to the feature complexity representation parameter, accurately predicts according to the historical behavior of the user, and improves the accuracy of prediction.

[0023] Further, the present application determines the classification of the product by the dimension number of the demand product information data, and judges whether the number of extracted product parameters is qualified according to the comparison result of the user attention degree and the preset attention degree for a single demand product, efficiently classifies the demand product, ensures that the number of extracted product parameters meets the user demand for a single demand product, and improves the accuracy of product recommendation.

[0024] Further, the present application determines the demand product to be a complex demand product, extracts the core functional features, and judges whether the number of extracted core functional features is qualified by comparing the user feature dispersion coefficient with the preset feature dispersion coefficient, extracts the core functional features for the complex demand product, which meets the basic cognition of the user and does not overly lengthy or miss the key information, thereby improving the accuracy of product recommendation.

[0025] Further, the present application adjusts the dimension number of the keyword feature according to the correlation degree of the change rates of the user purchase decision data for the demand product and the marketing strategy data of the enterprise by capturing and analyzing the user purchase decision data for the demand product and the marketing strategy data of the enterprise, more effectively matches the user demand and the marketing strategy, and improves the accuracy of marketing prediction.

[0026] Further, the application determines the eligibility of the prediction mode by comparing the prediction accuracy of the prediction mode with a prediction accuracy threshold, if not eligible, selects the way of adjusting the keyword feature extraction frequency or the demand product information data collection frequency to optimize the prediction mode, evaluates the performance of the prediction mode, and adjusts the optimization strategy according to the evaluation result, thereby improving the accuracy and adaptability of the prediction mode. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 Flowchart of the big data driven intelligent marketing prediction method of the embodiment of the application;

[0028] Figure 2 Flowchart of the extraction of the demand product parameter quantity eligibility of the embodiment of the application;

[0029] Figure 3 Flowchart of the extraction of the core function feature quantity eligibility of the embodiment of the application;

[0030] Figure 4 Flowchart of the prediction mode eligibility of the embodiment of the application. DETAILED DESCRIPTION

[0031] In order to make the objects and advantages of the application clearer, the application will be further described below with reference to the embodiments. It should be recognized that the specific embodiments described herein are only used to explain the application and do not limit the application.

[0032] The preferred embodiments of the application will be described below with reference to the accompanying drawings. Those skilled in the art should recognize that these embodiments are only used to explain the technical principles of the application and do not limit the protection scope of the application.

[0033] It should be noted that the data in this embodiment are obtained by comprehensive analysis and evaluation of the historical detection data and the corresponding historical detection results in the previous three months before the detection.

[0034] Please refer to Figures 1-4 as shown, Figure 1 Flowchart of the big data driven intelligent marketing prediction method of the embodiment of the application; Figure 2 Flowchart of the extraction of the demand product parameter quantity eligibility of the embodiment of the application; Figure 3A flowchart for extracting the number of core function features qualified for the embodiment of the application; Figure 4 A flowchart for predicting the eligibility of the mode for the embodiment of the application.

[0035] The big data driven intelligent marketing prediction method of the embodiment of the application comprises:

[0036] Step S1, historical behavior data of a plurality of users is acquired, and a first prediction mode or a second prediction mode is determined for the users based on a comprehensive evaluation value of the historical behavior data;

[0037] Step S2, chat record data of the user is acquired, keyword features about the user demand in the chat record data are extracted, and the user appeal degree is determined to be a low appeal degree or a high appeal degree based on a feature complexity representation parameter of the keyword features;

[0038] Step S3, under the condition of acquiring user demand product information data, the demand product is determined to be a single demand product or a complex demand product based on the dimension number of the demand product information data;

[0039] Step S4, under the condition of determining that the demand product is a single demand product, demand product parameters are extracted, and the eligibility of the number of extracted demand product parameters is determined based on the attention degree of the user with the low appeal degree or the high appeal degree;

[0040] Step S5, under the condition of determining that the demand product is a complex demand product, core function features of the demand product are extracted, and the eligibility of the number of extracted core function features of the demand product is determined based on the feature discrete coefficient of the user with the low appeal degree or the high appeal degree;

[0041] Step S6, real-time behavior data of a plurality of users is acquired, the eligibility of the second prediction mode or the first prediction mode is determined based on the prediction accuracy of the second prediction mode or the first prediction mode, and the key word feature extraction frequency or the demand product information data acquisition frequency is determined according to the ratio of the prediction accuracy to a prediction accuracy threshold.

[0042] In the embodiment of the application, the behavior data is various behavior record information data generated by the user in the relevant business activities of the enterprise, including browsing time data, browsing frequency data and purchase amount data.

[0043] In the embodiment of the application, the chat record data is text information record data generated when the user communicates and interacts with the relevant platform within an extraction period of one month.

[0044] In the embodiment of the application, the demand product information data is all-around descriptive data about the demand product within an extraction period of one month, including basic attribute data, function parameter data and performance parameter data.

[0045] In the embodiment of the present application, the user's behavior data and chat record data need to be authorized by the user.

[0046] Specifically, the present application determines the prediction mode by obtaining the user's historical behavior data, judges the user's appeal degree by extracting the chat record keyword features, distinguishes the product types according to the number of product information data dimensions, determines the eligibility of the extraction parameters or the number of functional features according to the user's appeal for different product types, verifies the prediction mode according to the real-time behavior data and determines the data processing frequency adjustment strategy, accurately grasps the user's demand, and is more targeted for product information processing, improves the efficiency of marketing prediction, and improves the accuracy of marketing prediction.

[0047] Specifically, the embodiment of the present application determines the prediction mode for the user according to the comparison result of the comprehensive evaluation value of the historical behavior data and the preset comprehensive evaluation value 0.85 under the condition of obtaining the historical behavior data of a plurality of users.

[0048] When the comprehensive evaluation value is less than or equal to the preset comprehensive evaluation value, the first prediction mode is determined for the user.

[0049] When the comprehensive evaluation value is greater than the preset comprehensive evaluation value, the second prediction mode is determined for the user.

[0050] In the embodiment of the present application, the preset comprehensive evaluation value is 0.85, but the above value is not limited thereto, and the skilled person in the art can also adjust the value according to the actual needs.

[0051] In the embodiment of the present application, the first prediction mode is to analyze the user's behavior data, and the second prediction mode is to analyze the user's behavior data, the user's purchase decision data, and the marketing strategy data respectively.

[0052] Specifically, the embodiment of the present application calculates the comprehensive evaluation value according to the following formula, and sets:

[0053]

[0054] Wherein, P represents the comprehensive evaluation value, T max is the maximum browsing time value, T min is the minimum browsing time value, μ T is the average browsing time value, F max is the maximum browsing frequency value, F min is the minimum browsing frequency value, μ F is the average browsing frequency value, M max is the maximum purchase amount value, M min is the minimum purchase amount value, μ M is the average purchase amount value.

[0055] Specifically, in the condition of determining to take different prediction modes for the user, the chat record data of the user is acquired, and the keyword features about the demand of the user in the chat record data are extracted at a preset keyword feature extraction frequency of 20 times / min, which is an average of historical keyword feature extraction frequencies.

[0056] In the embodiment of the application, the dimension of the keyword feature is the appearance frequency and the number of the keyword feature, and the dimension number of the keyword feature is 2.

[0057] Specifically, the user appeal degree is determined according to the comparison result of the feature complexity representation parameter of the keyword feature and a preset feature complexity representation parameter 0.83.

[0058] When the feature complexity representation parameter is less than or equal to the preset feature complexity representation parameter, the user appeal degree is determined to be a low appeal degree.

[0059] When the feature complexity representation parameter is greater than the preset feature complexity representation parameter, the user appeal degree is determined to be a high appeal degree.

[0060] In the embodiment of the application, the preset feature complexity representation parameter is 0.83, but the above value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0061] In the embodiment of the application, the keyword feature is a phrase that can reflect the purchase intention of the user, such as the words “like”, “plan” or “intend” and the like.

[0062] In the embodiment of the application, the low appeal degree is an appeal with a small number of keyword features, and the high appeal degree is an appeal with a large number of keyword features.

[0063] Specifically, the feature complexity representation parameter is calculated according to the following formula, and is set as:

[0064]

[0065] Wherein, Q represents the feature complexity representation parameter, f i is the frequency of the i-th keyword feature appearing in the chat record data, n represents the total number of extracted keyword features, max(f i ) is the frequency of the keyword feature with the highest frequency among all keyword features appearing in the chat record data, k is the number of types of keyword features in the chat record data, and k0 is the maximum number of types of keyword features in the chat record data of a plurality of users.

[0066] Specifically, the application determines to take the first or second prediction mode through the comprehensive evaluation value of user historical behavior data, acquires chat record data and extracts keyword features in the determined prediction mode, and determines the user appeal degree according to the feature complexity representation parameter, accurately predicts according to the historical behavior of the user, and improves the prediction accuracy.

[0067] Specifically, the demand product information data of the embodiment of the application is acquired at a frequency of 5 times / s, and the preset demand product information data acquisition frequency is the average value of historical demand product information data acquisition frequencies.

[0068] Specifically, the embodiment of the application determines the classification of the demand product according to the comparison result of the dimension number of the demand product information data and the preset dimension number 3.

[0069] When the dimension number is less than or equal to the preset dimension number, the demand product is determined to be a single demand product.

[0070] When the dimension number is greater than the preset dimension number, the demand product is determined to be a complex demand product.

[0071] In the embodiment of the application, the preset dimension number is 3, but the above-mentioned value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0072] In the embodiment of the application, the single demand product is a product including only basic attribute data, and the complex demand product is a product including basic attribute data, function parameter data and performance parameter data.

[0073] In the embodiment of the application, the dimension number is the inherent attribute number of the demand product information data, such as the size and weight of the basic attribute data, the control mode and compatibility of the function parameter data, and the power and efficiency of the performance parameter data.

[0074] Specifically, the embodiment of the application determines the dimension number according to the listed inherent attribute number of the demand product information data.

[0075] Specifically, the embodiment of the application extracts demand product parameters under the condition that the demand product is determined to be a single demand product, and determines the eligibility of the number of extracted demand product parameters according to the comparison result of the attention degree of the corresponding appeal degree user and the preset attention degree 0.79.

[0076] When the attention degree is less than or equal to the preset attention degree, the number of extracted demand product parameters is determined to be eligible.

[0077] When the attention degree is greater than the preset attention degree, the number of extracted demand product parameters is determined to be unqualified.

[0078] In the embodiment of the present application, the preset attention value is 0.79, which is obtained by averaging the attention values of a plurality of historical qualified demand product parameter quantities, but the above value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0079] In the embodiment of the present application, the demand product parameter is a parameter contained in the basic attribute data.

[0080] Specifically, the attention value is the difference between the maximum attention frequency of the parameter in the basic attribute data and the minimum attention frequency of the parameter, divided by the average frequency of the parameter.

[0081] Specifically, the present application determines the classification of the product by the number of dimensions of the demand product information data, and determines whether the number of extracted product parameters is qualified according to the comparison result of the user attention value and the preset attention value for a single demand product, efficiently classifies the demand product, and ensures that the number of extracted product parameters meets the user demand for a single demand product, thereby improving the accuracy of product recommendation.

[0082] Specifically, in the embodiment of the present application, the core function features are extracted under the condition that the demand product is a complex demand product, and the qualification of the number of extracted core function features is determined according to the comparison result of the feature dispersion coefficient of the corresponding appeal degree user and the preset feature dispersion coefficient 0.81.

[0083] When the feature dispersion coefficient is less than or equal to the preset feature dispersion coefficient, it is determined that the number of extracted core function features is qualified.

[0084] When the feature dispersion coefficient is greater than the preset feature dispersion coefficient, it is determined that the number of extracted core function features is unqualified.

[0085] In the embodiment of the present application, the preset feature dispersion coefficient is 0.81, which is obtained by averaging the feature dispersion coefficients of a plurality of historical core function feature quantities, but the above value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0086] In the embodiment of the present application, the core function feature is a feature contained in the function parameter data and the performance parameter data.

[0087] Specifically, the feature dispersion coefficient is calculated according to the following formula in the embodiment of the present application, and is set as:

[0088]

[0089] Wherein, D represents the feature dispersion coefficient, A max is the maximum attention frequency of the parameter in the function parameter data, and Amin is the minimum attention frequency of the parameter in the functional parameter data, is the average attention frequency of the parameter in the functional parameter data, B max is the maximum attention frequency of the parameter in the performance parameter data, B min is the minimum attention frequency of the parameter in the performance parameter data, is the average attention frequency of the parameter in the performance parameter data.

[0090] Specifically, the present application determines that the demand product is a complex demand product, extracts core functional features, and judges whether the number of extracted core functional features is qualified through comparison of the user feature dispersion coefficient and the preset feature dispersion coefficient. For complex demand products, the extracted core functional features meet the basic cognition of users and are not too long or lack key information, thereby improving the accuracy of product recommendation.

[0091] Specifically, the present application extracts purchase decision data and marketing strategy data of users on demand products in big data. The purchase decision data is data reflecting the decision-making process of users purchasing demand products. The marketing strategy data is data of enterprises formulating and implementing marketing strategies, including advertising, promotion activities, product recommendation and other strategies.

[0092] Specifically, the present application determines the marketing strategy change rate by analyzing the purchase decision change rate of users on demand products under the condition of determining the classification of demand products. The purchase decision change rate is the ratio of the number of purchase decision changes to the total number of purchase decisions. The purchase decision change is the change of purchase decision of users from determining a number of demand products to paying due to the influence of other factors. The total number of purchase decisions is the number of times of purchasing demand products by users. The marketing strategy change rate is the ratio of the number of marketing strategy changes to the total number of marketing strategies. The number of marketing strategy changes is the change of advertising, promotion activities, product recommendation and other strategies in the marketing process. The total number of marketing strategies is the total number of executed marketing strategies.

[0093] Specifically, the present application determines the number of dimensions of the keyword feature according to the comparison result of the correlation degree of the marketing strategy change rate and the purchase decision change rate and the preset correlation degree.

[0094] When the correlation degree is less than or equal to the preset correlation degree, the number of dimensions of the keyword feature is increased to the corresponding value by the first preset keyword dimension adjustment coefficient 2.

[0095] When the correlation degree is greater than the preset correlation degree, the number of dimensions of the keyword feature is increased to the corresponding value by the second preset keyword dimension adjustment coefficient 4.

[0096] In the embodiment of the present application, the preset correlation degree value is 0.52, but the above value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0097] In the embodiment of the present application, the increased dimension number of the keyword feature is the product of the dimension number and the first preset keyword dimension adjustment coefficient, the value of l is 1 or 2, U1 is the first preset keyword dimension adjustment coefficient 2, and U2 is the second preset keyword dimension adjustment coefficient 4.

[0098] Specifically, the correlation degree is calculated according to the following formula, and the following is set:

[0099]

[0100] wherein, represents the correlation degree, is the purchase decision change rate at the tth time, is the marketing strategy change rate at the tth time, is the average of the purchase decision change rate, is the average of the marketing strategy change rate.

[0101] Specifically, the present application adjusts the dimension number of the keyword feature according to the correlation degree of the change rates of the purchase decision data of the user for the demand product and the marketing strategy data of the enterprise by capturing and analyzing the data, more effectively matches the user demand and the marketing strategy, and improves the accuracy of the marketing prediction.

[0102] Specifically, the embodiment of the present application determines the eligibility of the prediction mode according to the comparison result of the prediction accuracy of the prediction mode and the prediction accuracy threshold 95% under the condition of obtaining real-time behavior data of a plurality of users;

[0103] When the prediction accuracy is less than or equal to the prediction accuracy threshold, it is determined that the prediction mode is unqualified;

[0104] When the prediction accuracy is greater than the prediction accuracy threshold, it is determined that the prediction mode is qualified.

[0105] In the embodiment of the present application, the prediction accuracy threshold value is 95%, but the above value is not limited thereto, and the value can be adjusted according to actual needs by those skilled in the art.

[0106] Specifically, the prediction accuracy is the ratio of the result of the real-time behavior data of the prediction mode to the actual result of the real-time behavior data.

[0107] Specifically, the embodiment of the present application determines the adjustment mode of the prediction mode according to a comparison result of a ratio of the prediction accuracy of the prediction mode to a prediction accuracy threshold value and a preset ratio 0.55 under the condition that the corresponding prediction mode is determined to be unqualified.

[0108] When the ratio is less than or equal to the preset ratio, the keyword feature extraction frequency is increased to a corresponding value by a preset extraction frequency adjustment coefficient 1.04.

[0109] When the ratio is greater than the preset ratio, the demand product information data collection frequency is increased to a corresponding value by a preset collection frequency adjustment coefficient 1.07.

[0110] The ratio is a ratio of the prediction accuracy to the prediction accuracy threshold value.

[0111] In the embodiment of the present application, the preset ratio is 0.55, but the above value is not limited thereto, and a person skilled in the art can adjust the value according to actual needs.

[0112] In the embodiment of the present application, the increased keyword feature extraction frequency is a product of the preset keyword feature extraction frequency and the preset extraction frequency adjustment coefficient 1.04, and the increased demand product information data collection frequency is a product of the preset demand product information data collection frequency and the preset collection frequency adjustment coefficient 1.07.

[0113] Specifically, the present application determines the qualification of the prediction mode by comparing the prediction accuracy of the prediction mode with the prediction accuracy threshold value, and if unqualified, selects the way of adjusting the keyword feature extraction frequency or the demand product information data collection frequency to optimize the prediction mode, evaluates the performance of the prediction mode, and adjusts the optimization strategy according to the evaluation result, thereby improving the accuracy and adaptability of the prediction mode.

[0114] The technical solutions of the present application have been described in connection with the preferred embodiments shown in the drawings, but a person skilled in the art can easily recognize that the protection scope of the present application is obviously not limited to these specific embodiments. A person skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

[0115] The above description is only the preferred embodiments of the present application and is not used to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A big data driven intelligent marketing prediction method, characterized by: include: Obtain historical behavior data of several users and determine a prediction model for the users based on the comprehensive evaluation value of the historical behavior data; Based on the comparison result that the comprehensive evaluation value of the historical behavior data is less than or equal to the preset comprehensive evaluation value, a first prediction mode is determined for the user, where the first prediction mode is to analyze the user behavior data; Based on the comparison result that the comprehensive evaluation value of the historical behavior data is greater than the preset comprehensive evaluation value, a second prediction model is adopted for the user, and the second prediction model is to analyze the user's behavior data, the user's purchase decision data, and the marketing strategy data respectively; Obtaining user chat record data, extracting keyword features related to user needs from the chat record data, and determining the user's demand level based on feature complexity representation parameters of the keyword features, including low demand level or high demand level; Under the condition of obtaining the user's demand product information data, determine whether the demand product is a single demand product or a complex demand product based on the number of dimensions of the demand product information data; Under the condition that the demand product is determined to be a single demand product, the demand product parameters are extracted, and the eligibility of the number of extracted demand product parameters is determined based on the attention of users with low or high demand levels; Under the condition that the demand product is determined to be a complex demand product, the core functional features of the demand product are extracted, and the eligibility of the number of extracted core functional features of the demand product is determined based on the characteristic dispersion coefficient of users with low or high demand levels; Acquire real-time behavioral data of a number of users, determine that the corresponding prediction model is unqualified based on a comparison result that the prediction accuracy of the corresponding prediction model is less than or equal to a prediction accuracy threshold, and determine to increase the keyword feature extraction frequency by a preset extraction frequency adjustment coefficient based on a comparison result that the ratio of the prediction accuracy to the prediction accuracy threshold is less than or equal to a preset ratio; According to a comparison result that the ratio of the prediction accuracy rate to the prediction accuracy rate threshold is greater than a preset ratio, determining to increase the frequency of collecting demand product information data by a preset collection frequency adjustment coefficient; Furthermore, by capturing the purchase decision data and marketing strategy data of users on the demand products in the big data, and comparing the correlation determined based on the purchase decision change rate of the purchase decision data and the marketing strategy change rate of the marketing strategy data with the preset correlation, it is determined that the number of dimensions of the keyword feature is increased by the first preset keyword dimension adjustment coefficient or the second preset keyword dimension adjustment coefficient; The purchase decision change rate is the ratio of the number of purchase decision changes to the total number of purchase decisions; The marketing strategy change rate is the ratio of the number of times the marketing strategy changes to the total number of times the marketing strategy is formulated.

2. The big data driven intelligent marketing prediction method according to claim 1, characterized in that: Based on the comparison result that the comprehensive evaluation value of the historical behavior data is less than or equal to the preset comprehensive evaluation value, it is determined that the first prediction mode is adopted for the user.

3. The big data driven intelligent marketing prediction method according to claim 1, characterized in that: Based on the comparison result that the comprehensive evaluation value of the historical behavior data is greater than the preset comprehensive evaluation value, it is determined that the second prediction mode is adopted for the user.

4. The big data driven intelligent marketing prediction method according to claim 1, characterized in that: Based on a comparison result that a feature complexity representation parameter of the extracted keyword feature is less than or equal to a preset feature complexity representation parameter, it is determined that the user appeal level is low.

5. The big data driven intelligent marketing prediction method according to claim 1, characterized in that: The user demand degree is determined to be high based on a comparison result that the feature complexity representation parameter of the extracted keyword feature is greater than the preset feature complexity representation parameter.

6. The big data driven intelligent marketing prediction method according to claim 4 or 5, characterized in that: Based on the comparison result that the number of dimensions of the demand product information data is less than or equal to the preset number of dimensions, the demand product is determined to be a single demand product, and based on the comparison result that the user's attention to the corresponding demand level is greater than the preset attention level, the number of extracted demand product parameters is determined to be qualified.

7. The big data driven intelligent marketing prediction method according to claim 4 or 5, characterized in that: Based on the fact that the number of dimensions of the demand product information data is greater than the preset number of dimensions, the demand product is determined to be a complex demand product, and based on the comparison result that the feature dispersion coefficient of users with corresponding demand levels is greater than the preset feature dispersion coefficient, it is determined that the number of core functional features extracted from the demand product is qualified.

Images