Precision marketing recommendation system based on AI large model

Abstract

The invention discloses a precision marketing recommendation system based on an AI large model, relates to the technical field of information recommendation, is used for improving the problem of product information recommendation efficiency reduction caused by tag recommendation delay, and comprises the steps of collecting user data, analyzing user information frequency modulation potentials by using a K nearest neighbor algorithm, performing classification, screening out users with high frequency modulation potentials, and performing recommendation on the users with high frequency modulation potentials. The method comprises the following steps: setting detection time to collect recommendation information data of a user, screening key tags from the recommendation information data, calculating a tag demand coefficient, detecting a user emptying behavior, carrying out secondary screening on the key tags to obtain evaluation tags, and judging whether demand down-regulation processing is carried out on the evaluation tags by utilizing a corrected geometric average method. And marking the evaluation labels requiring down-regulation processing, obtaining label attributes of the marked labels, calculating down-regulation coefficients of different marked labels according to the label attributes of the marked labels and the label demand coefficients, and setting skipping time for the marked labels according to the down-regulation coefficients.

Description

Technical Field

[0001] The present invention relates to the field of information recommendation technology, and more specifically, to a precision marketing recommendation system based on an AI big model. Background Art

[0002] Information recommendation technology uses data analysis and algorithms to recommend products or services to users on a marketing platform, thereby improving user satisfaction. When information recommendation technology is applied to a marketing recommendation system, product information can be accurately recommended to high-demand users.

[0003] The prior art has the following deficiencies: In the past, when marketing platforms recommended product information to users, they determined user needs based on user behavior and made accurate recommendations, without taking into account the different timeliness of products under different tags. When user needs shifted, the platform still provided the user with a large number of products from past tags, resulting in a decrease in the efficiency of product information recommendations. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a precision marketing recommendation system based on an AI big model, which evaluates and classifies the user's frequency tuning potential by analyzing the degree of user demand transfer, multiple-screens product tags based on user categories and user behaviors, and sets different skip times for different product tags to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: Precision marketing recommendation system based on AI big model, including user data collection module, frequency modulation potential analysis module, label demand evaluation module and information frequency modulation processing module; The user data collection module is used to collect user data and transfer the user data to the frequency modulation potential analysis module. After receiving the user classification result returned by the frequency modulation potential analysis module, the user data collection module sets the detection time to collect the recommended information data and transfer it to the label demand evaluation module; The frequency modulation potential analysis module analyzes the user's information frequency modulation potential using the K nearest neighbor algorithm based on the user data transmitted by the user data collection module and classifies the user, and transmits the user classification result back to the user data collection module; The tag demand evaluation module is used to receive recommended information data, screen out key tags according to the recommended information data and calculate the corresponding tag demand coefficient, detect the user's clearing behavior and perform secondary screening on the key tags to obtain evaluation tags, use the modified geometric mean method to determine whether to adjust the demand of the evaluation tags downward, mark the evaluation tags that are to be adjusted downward, and send the tag demand coefficient of the marked tag to the information frequency modulation processing module; After receiving the marking tag, the information frequency modulation processing module obtains the tag attribute of the marking tag, calculates the down-adjustment coefficient of the corresponding marking tag based on the tag attribute of the marking tag and the tag requirement coefficient, and sets the skip time for the marking tag according to the down-adjustment coefficient.

[0006] In a preferred embodiment, the user data in the user data collection module includes the number of tag categories occupied by the user's search keywords and the browsing time of tags of different categories; When a user enters a search keyword through the search bar, it will be recorded in the search history. In the search history, the search keyword entered by the user is identified and the tag category to which it belongs is determined. The total number of different tag categories is counted to obtain the number of tag categories occupied by the user's search keyword; When a user browses in the marketing platform, the tag where the user's browsing information is located is identified and the browsing time is recorded to obtain the browsing time of different categories of tags.

[0007] In a preferred embodiment, the frequency modulation potential analysis module comprehensively analyzes the user's information frequency modulation potential using the K nearest neighbor algorithm based on the number of tag categories occupied by the user's search keywords and the browsing time of different categories and classifies the user. The specific steps are as follows: Collect the number of tag categories and browsing time of different categories of search keywords of multiple users, and perform standardization processing respectively, merge the number of tag categories of search keywords into a search data set, merge the browsing time of different categories into a browsing data set, take the ratio of the data in the search data set or the browsing data set to the maximum value in the search data set or the browsing data set as the standardization result of the corresponding data, randomly select a data in the search data set and the browsing data set as the reference data, set the nearest neighbor coefficient K, select K data with the smallest absolute value of the difference with the reference data in each data set as the marked data, calculate the average value of the marked data in the search data set and the browsing data set, and sum them to obtain the user classification threshold; The user's information frequency tuning potential is obtained by summing up the number of search keyword categories and the standardized browsing time of different categories, and the information frequency tuning potential is compared with the user classification threshold; when the user's information frequency tuning potential exceeds the user classification threshold, the user is judged to be a user with high information frequency tuning potential; when the user's information frequency tuning potential is lower than the user classification threshold, the user is judged to be a user with low information frequency tuning potential.

[0008] In a preferred embodiment, the frequency tuning potential analysis module classifies the user and transmits the user classification result to the user data collection module. When the user is a high information frequency tuning potential user, the user data collection module sets a detection time to collect recommended information data, and the recommended information data is the browsing frequency of the user under each tag; The user data collection module selects a period of time as the detection time, records and counts the number of tags to which the recommended content clicked by the user during the detection time, obtains the user's browsing frequency under each tag during the detection time, and passes the user's browsing frequency under each tag to the tag demand evaluation module.

[0009] In a preferred embodiment, the tag demand evaluation module receives the browsing frequency of each tag by the user, selects the median of the browsing frequency under each tag as the screening threshold, screens out tags with browsing frequencies exceeding the screening threshold as key tags, and uses the ratio of the browsing frequency of the key tag to the screening threshold as the tag demand coefficient of the corresponding tag; User clearing behavior refers to the user's active deletion of keywords in the search history within the detection time. The tags of each keyword in the clear list are identified and matched with the key tags. The successfully matched tags are retained as evaluation tags, and the clearing ratio of evaluation tags is calculated. The number of keywords in the clear list with the same evaluation tag is accumulated and divided by the total number of keywords in the clear list to obtain the tag clearing ratio of the corresponding evaluation tag.

[0010] In a preferred embodiment, the label demand evaluation module comprehensively evaluates the label demand coefficient and the label clearing ratio of the label and uses the modified geometric mean method to determine whether to adjust the demand of the evaluation label downward. The specific steps are as follows: Data conversion: convert the label requirement coefficient and label clearing ratio of the evaluation label into the same dimension; Repeated conversion: Set n conversion parameters to obtain conversion results of multiple label requirement coefficients or label clearing ratios; Data connection: multiple connection values ​​are obtained by subtracting the label requirement coefficient and label clearing ratio after data conversion using the same conversion parameters; Connection processing: perform logarithmic operations on multiple connection values ​​and take the arithmetic mean as the connection processing amount; Antilogarithmic processing: Take the antilogarithm of the connection processing amount to obtain the demand evaluation value; When the demand evaluation value of an evaluation tag exceeds the preset evaluation threshold, the demand for the evaluation tag will not be adjusted downward; when the demand evaluation value of an evaluation tag is lower than the preset evaluation threshold, the demand for the evaluation tag will be adjusted downward.

[0011] In a preferred embodiment, the label attributes of the marking label are the shelf life and rating of the product under the corresponding label; the rating rating is a scoring system of the marketing platform, and different star ratings are set to achieve the purpose of evaluating different products.

[0012] In a preferred embodiment, when calculating the downward adjustment coefficient of the marked label, the information frequency modulation processing module randomly selects multiple products under the marked label as target products, collects the shelf time and evaluation star rating of each target product, calculates the average shelf time of all target products as the label shelf time, and counts the total number of evaluation stars of all target products as the label star value.

[0013] In a preferred embodiment, the label shelf time and label star value of the marked label are normalized and the geometric mean is taken as the demand adjustment amount, and the downward adjustment coefficient of the corresponding marked label is calculated by the harmonic formula: , where B is the tag demand coefficient of the tag, s is the preset reconciliation parameter, h is the demand adjustment of the tag, and T is the down-regulation coefficient of the tag; The product of the default skip time of the marker label and the down-adjustment coefficient is used as the skip time of the corresponding marker label.

[0014] The technical effects and advantages of the precision marketing recommendation system based on AI big model of the present invention are as follows: The present invention collects user data and analyzes the frequency tuning potential of user information to classify it, screens out users with high frequency tuning potential, screens out users to determine user directions, improves processing efficiency, sets detection time to collect user recommendation information data, screens out key tags in the recommendation information data and calculates tag demand coefficients, detects user clearing behavior and performs secondary screening on key tags to obtain evaluation tags, performs secondary screening on tags to improve the accuracy of tag analysis, determines whether to perform downward adjustment processing on evaluation tags, marks evaluation tags that require downward adjustment, obtains tag attributes of the marked tags, calculates downward adjustment coefficients of different marked tags based on the tag attributes of the marked tags and the tag demand coefficients, sets skip time for the marked tags according to the downward adjustment coefficients, thereby reducing the recommendation of invalid tags or keywords and improving the accuracy of marketing recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic diagram of the precision marketing recommendation system based on the AI ​​big model of the present invention.

[0016] Figure 2 This is a flow chart of the precision marketing recommendation system based on the AI ​​big model of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The present invention collects user data and analyzes the user information tuning potential to classify it, screens out users with high tuning potential, sets detection time to collect users' recommended information data, screens out key tags in the recommended information data and calculates tag demand coefficients, detects users' clearing behavior and performs secondary screening on key tags to obtain evaluation tags, determines whether to perform downward adjustment processing on the evaluation tags, marks the evaluation tags that require downward adjustment, obtains tag attributes of the marked tags, calculates downward adjustment coefficients of different marked tags based on the tag attributes of the marked tags and the tag demand coefficients, sets skip time for the marked tags according to the downward adjustment coefficients, thereby reducing the recommendation of invalid tags or keywords and improving the accuracy of marketing recommendations.

[0019] Example: A precision marketing recommendation system based on an AI big model, such as Figure 1 and Figure 2 As shown, it includes a user data acquisition module, a frequency modulation potential analysis module, a tag demand evaluation module and an information frequency modulation processing module, and the signals between the modules are connected; The functions of each module are as follows: The user data collection module is used to collect user data and transfer the user data to the frequency modulation potential analysis module. After receiving the user classification result returned by the frequency modulation potential analysis module, the user data collection module sets the detection time to collect the recommended information data and transfer it to the label demand evaluation module; The frequency modulation potential analysis module analyzes the user's information frequency modulation potential using the K nearest neighbor algorithm based on the user data transmitted by the user data collection module and classifies the user, and transmits the user classification result back to the user data collection module; The tag demand evaluation module is used to receive recommended information data, screen out key tags according to the recommended information data and calculate the corresponding tag demand coefficient, detect the user's clearing behavior and perform secondary screening on the key tags to obtain evaluation tags, use the modified geometric mean method to determine whether to adjust the demand of the evaluation tags downward, mark the evaluation tags that are to be adjusted downward, and send the tag demand coefficient of the marked tag to the information frequency modulation processing module; After receiving the marking tag, the information frequency modulation processing module obtains the tag attribute of the marking tag, calculates the down-adjustment coefficient of the corresponding marking tag based on the tag attribute of the marking tag and the tag requirement coefficient, and sets the skip time for the marking tag according to the down-adjustment coefficient.

[0020] It should be noted that the skip return mechanism is used to perform secondary calibration on the skip time of some marked tags to improve the fault tolerance of marketing recommendations.

[0021] The user data in the user data collection module includes the number of tag categories occupied by the user's search keywords and the browsing time of tags in different categories; The user enters the search keyword through the search bar in the marketing platform to obtain the information of his own needs. The relational database of the marketing platform stores the tag categories of all item information and all search keywords corresponding to different tags. Every time the user enters a search keyword through the search bar, it will be recorded in the search history. In the search history, the search keyword entered by the user is identified and the tag category to which it belongs is determined. The total number of different tag categories is counted to obtain the number of tag categories occupied by the user's search keyword; When a user browses in the marketing platform, the tag where the user's browsing information is located is identified and the browsing time is recorded to obtain the browsing time of different categories of tags.

[0022] The frequency modulation potential analysis module uses the K nearest neighbor algorithm to analyze the user's information frequency modulation potential and classify the user based on the user's search keywords, the number of tag categories, and the browsing time of different categories. The specific steps are as follows: Collect the number of tag categories and browsing time of different categories of search keywords of multiple users, and perform standardization processing respectively, merge the number of tag categories of search keywords into a search data set, merge the browsing time of different categories into a browsing data set, take the ratio of the data in the search data set or the browsing data set to the maximum value in the search data set or the browsing data set as the standardization result of the corresponding data, randomly select a data in the search data set and the browsing data set as the reference data, set the nearest neighbor coefficient K, select K data with the smallest absolute value of the difference with the reference data in each data set as the marked data, calculate the average value of the marked data in the search data set and the browsing data set, and sum them to obtain the user classification threshold; The user's information frequency tuning potential is obtained by summing up the number of search keyword categories and the standardized browsing time of different categories, and the information frequency tuning potential is compared with the user classification threshold; when the user's information frequency tuning potential exceeds the user classification threshold, the user is judged to be a user with high information frequency tuning potential; when the user's information frequency tuning potential is lower than the user classification threshold, the user is judged to be a user with low information frequency tuning potential.

[0023] It should be noted that, the higher the number of search keywords in the label categories, the richer the user recommended content, the more need for information frequency tuning, and the higher the potential for information frequency tuning; the longer the browsing time of different categories, the richer the user recommended content, the more need for information frequency tuning, and the higher the potential for information frequency tuning. Information frequency tuning refers to removing or postponing some recommended information tags when there are too many recommended information tags; the neighbor coefficient K can be set according to the actual situation. For example, the neighbor coefficient K is set to 6, which will not be elaborated here.

[0024] The frequency tuning potential analysis module classifies the users and transmits the user classification results to the user data collection module. When the user is a high information frequency tuning potential user, the user data collection module sets a detection time to collect recommended information data. The recommended information data is the browsing frequency of the user under each tag; The user data collection module selects a period of time as the detection time, records and counts the number of tags to which the recommended content clicked by the user during the detection time, obtains the user's browsing frequency under each tag during the detection time, and passes the user's browsing frequency under each tag to the tag demand evaluation module.

[0025] It should be noted that the time interval of the detection time is set by professionals in this field and will not be analyzed here.

[0026] The tag demand evaluation module receives the browsing frequency of users under each tag, selects the median browsing frequency under each tag as the screening threshold, filters out tags with browsing frequencies exceeding the screening threshold as key tags, and uses the ratio of the browsing frequency of the key tag to the screening threshold as the tag demand coefficient of the corresponding tag; User clearing behavior refers to the user's active deletion of keywords in the search history within the detection time. The tags of each keyword in the clear list are identified and matched with the key tags. The successfully matched tags are retained as evaluation tags, and the clearing ratio of evaluation tags is calculated. The number of keywords in the clear list with the same evaluation tag is accumulated and divided by the total number of keywords in the clear list to obtain the tag clearing ratio of the corresponding evaluation tag.

[0027] It should be noted that the clear list is a temporary storage station for storing deleted information. Keywords deleted by users during the detection time can be obtained in the clear list. In the above, since the tags where some keywords in the clear list are located may not be in the key tags, they are matched and retained.

[0028] The label demand evaluation module comprehensively evaluates the label demand coefficient and label clearance ratio of the label and uses the modified geometric mean method to determine whether to adjust the demand of the evaluation label downward. The specific steps are as follows: Data conversion: Perform data conversion on the label requirement coefficient and label clearing ratio of the evaluation label: ,in, To evaluate the label requirement coefficient or label clearance ratio of the label, is the result of the conversion of the label requirement coefficient or label clearing ratio data corresponding to the evaluation label. It is a conversion parameter used to convert the label requirement coefficient and label clearing ratio into the same dimension; Repeated conversion: Set n conversion parameters to obtain conversion results of multiple label requirement coefficients or label clearing ratios; Data connection: multiple connection values ​​are obtained by subtracting the label requirement coefficient and label clearing ratio after data conversion using the same conversion parameters; Connection processing: perform logarithmic operations on multiple connection values ​​and take the arithmetic mean as the connection processing amount; Antilogarithmic processing: Take the antilogarithm of the connection processing amount to obtain the demand evaluation value.

[0029] When the demand evaluation value of an evaluation tag exceeds the preset evaluation threshold, the demand for the evaluation tag will not be adjusted downward; when the demand evaluation value of an evaluation tag is lower than the preset evaluation threshold, the demand for the evaluation tag will be adjusted downward.

[0030] The tag demand evaluation module marks the evaluation tags that are judged to be subject to demand reduction processing, and sends the tag demand coefficient of the marked tag to the information frequency modulation processing module.

[0031] It should be noted that when setting the conversion parameters, the conditions must be met to avoid the original data becoming zero or negative after data conversion. The smaller the label demand coefficient, the smaller the demand for evaluation labels, and the larger the label clearing ratio, which means that the user's demand for labels is reduced, and the smaller the demand for evaluation labels, the more demand reduction processing is needed; demand reduction processing is to reduce the appearance time of labels by delaying demand, thereby improving recommendation accuracy.

[0032] The tag attributes of the marking tag are the shelf time and evaluation star rating of the product under the corresponding tag. The lower the shelf time of the product, the worse the product timeliness, the easier it is for the marking tag to expire, and the longer the skip time needs to be extended. The evaluation star rating is the scoring system of the marketing platform. By setting different star ratings, different products can be evaluated. The lower the cumulative star rating of each product under the same tag, the worse the product timeliness, the easier it is for the marking tag to expire, and the longer the skip time needs to be extended.

[0033] The tag attributes and tag requirement coefficient of the tag are comprehensively calculated to calculate the tag down-adjustment coefficient and set the skip time. The specific steps are as follows: Randomly select multiple products under the marked label as target products, collect the shelf time and rating star of each target product, calculate the average shelf time of all target products as the label shelf time, and count the total number of rating stars of all target products as the label star value; After normalizing the label shelf time and label star value of the marked label, take the geometric mean as the demand adjustment amount, and calculate the downward adjustment coefficient of the corresponding marked label through the harmonic formula: , where B is the tag demand coefficient of the tag, s is the preset reconciliation parameter, h is the demand adjustment of the tag, and T is the down-regulation coefficient of the tag; The product of the default skip time of the mark tag and the downward adjustment coefficient is used as the skip time of the corresponding mark tag. By setting the skip time for different mark tags, the recommendation of a large number of invalid tags or keywords can be reduced, and the accuracy of marketing recommendations can be improved. At the same time, each mark tag is also recorded and saved for easy subsequent management analysis and call.

[0034] It should be noted that the reconciliation parameter is used to convert the downward adjustment coefficient into the skip time ratio of the marked label. The smaller the label demand coefficient of the marked label or the larger the demand adjustment amount, the larger the downward adjustment coefficient of the marked label, the lower the demand for the marked label, and the corresponding skip time of the marked label is extended. The default skip time is the time period when the marked label is not recommended as initially set in the marketing platform. The reconciliation parameter is set by professionals in this field, and the reconciliation parameter is greater than 1 to ensure the skip time delay effect of the marked label.

[0035] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0036] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0037] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0038] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0039] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A precision marketing recommendation system based on AI big model, characterized by: It includes user data collection module, frequency modulation potential analysis module, label demand evaluation module and information frequency modulation processing module; The user data collection module is used to collect user data and transfer the user data to the frequency modulation potential analysis module. After receiving the user classification result returned by the frequency modulation potential analysis module, the user data collection module sets the detection time to collect the recommended information data and transfer it to the label demand evaluation module; The frequency modulation potential analysis module analyzes the user's information frequency modulation potential using the K nearest neighbor algorithm based on the user data transmitted by the user data collection module and classifies the user, and transmits the user classification result back to the user data collection module; The tag demand evaluation module is used to receive recommended information data, screen out key tags according to the recommended information data and calculate the corresponding tag demand coefficient, detect the user's clearing behavior and perform secondary screening on the key tags to obtain evaluation tags, use the modified geometric mean method to determine whether to adjust the demand of the evaluation tags downward, mark the evaluation tags that are to be adjusted downward, and send the tag demand coefficient of the marked tag to the information frequency modulation processing module; After receiving the marking tag, the information frequency modulation processing module obtains the tag attribute of the marking tag, calculates the down-adjustment coefficient of the corresponding marking tag based on the tag attribute of the marking tag and the tag requirement coefficient, and sets the skip time for the marking tag according to the down-adjustment coefficient.

2. The precision marketing recommendation system based on AI big model according to claim 1 is characterized by: The user data in the user data collection module includes the number of tag categories occupied by the user's search keywords and the browsing time of tags in different categories; When a user enters a search keyword through the search bar, it will be recorded in the search history. In the search history, the search keyword entered by the user is identified and the tag category to which it belongs is determined. The total number of different tag categories is counted to obtain the number of tag categories occupied by the user's search keyword; When a user browses in the marketing platform, the tag where the user's browsing information is located is identified and the browsing time is recorded to obtain the browsing time of different categories of tags.

3. The precision marketing recommendation system based on AI big model according to claim 2 is characterized by: The frequency modulation potential analysis module uses the K nearest neighbor algorithm to analyze the user's information frequency modulation potential and classify the user based on the user's search keywords, the number of tag categories, and the browsing time of different categories. The specific steps are as follows: Collect the number of tag categories and browsing time of different categories of search keywords of multiple users, and perform standardization processing respectively, merge the number of tag categories of search keywords into a search data set, merge the browsing time of different categories into a browsing data set, take the ratio of the data in the search data set or the browsing data set to the maximum value in the search data set or the browsing data set as the standardization result of the corresponding data, randomly select a data in the search data set and the browsing data set as the reference data, set the nearest neighbor coefficient K, select K data with the smallest absolute value of the difference with the reference data in each data set as the marked data, calculate the average value of the marked data in the search data set and the browsing data set, and sum them to obtain the user classification threshold; The user's information tuning potential is obtained by summing up the number of search keyword categories and the normalized browsing time of different categories, and the information tuning potential is compared with the user classification threshold; When the user's information frequency tuning potential exceeds the user classification threshold, the user is judged as a user with high information frequency tuning potential; when the user's information frequency tuning potential is lower than the user classification threshold, the user is judged as a user with low information frequency tuning potential.

4. The precision marketing recommendation system based on AI big model according to claim 3 is characterized by: The frequency tuning potential analysis module classifies the users and transmits the user classification results to the user data collection module. When the user is a high information frequency tuning potential user, the user data collection module sets a detection time to collect recommended information data. The recommended information data is the browsing frequency of the user under each tag; The user data collection module selects a period of time as the detection time, records and counts the number of tags to which the recommended content clicked by the user during the detection time, obtains the user's browsing frequency under each tag during the detection time, and passes the user's browsing frequency under each tag to the tag demand evaluation module.

5. The precision marketing recommendation system based on AI big model according to claim 4 is characterized by: The tag demand evaluation module receives the browsing frequency of users under each tag, selects the median browsing frequency under each tag as the screening threshold, filters out tags with browsing frequencies exceeding the screening threshold as key tags, and uses the ratio of the browsing frequency of the key tag to the screening threshold as the tag demand coefficient of the corresponding tag; User clearing behavior refers to the user's active deletion of keywords in the search history within the detection time. The tags of each keyword in the clear list are identified and matched with the key tags. The successfully matched tags are retained as evaluation tags, and the clearing ratio of evaluation tags is calculated. The number of keywords in the clear list with the same evaluation tag is accumulated and divided by the total number of keywords in the clear list to obtain the tag clearing ratio of the corresponding evaluation tag.

6. The precision marketing recommendation system based on AI big model according to claim 5 is characterized by: The label demand evaluation module comprehensively evaluates the label demand coefficient and label clearance ratio of the label and uses the modified geometric mean method to determine whether to adjust the demand of the evaluation label downward. The specific steps are as follows: Data conversion: convert the label requirement coefficient and label clearing ratio of the evaluation label into the same dimension; Repeated conversion: Set n conversion parameters to obtain conversion results of multiple label requirement coefficients or label clearing ratios; Data connection: multiple connection values ​​are obtained by subtracting the label requirement coefficient and label clearing ratio after data conversion using the same conversion parameters; Connection processing: perform logarithmic operations on multiple connection values ​​and take the arithmetic mean as the connection processing amount; Antilogarithmic processing: Take the antilogarithm of the connection processing amount to obtain the demand evaluation value; When the demand evaluation value of an evaluation tag exceeds the preset evaluation threshold, the demand for the evaluation tag will not be adjusted downward; when the demand evaluation value of an evaluation tag is lower than the preset evaluation threshold, the demand for the evaluation tag will be adjusted downward.

7. The precision marketing recommendation system based on AI big model according to claim 1 is characterized by: The tag attributes of the marking tag are the shelf life of the product under the corresponding tag and the rating star; the rating star is a scoring system of the marketing platform, and different star ratings are set to achieve the purpose of evaluating different products.

8. The precision marketing recommendation system based on AI big model according to claim 7 is characterized by: When calculating the downward adjustment coefficient of the marked label, the information frequency modulation processing module randomly selects multiple products under the marked label as target products, collects the shelf time and evaluation star rating of each target product, calculates the average shelf time of all target products as the label shelf time, and counts the total number of evaluation stars of all target products as the label star value.

9. The precision marketing recommendation system based on AI big model according to claim 8 is characterized by: After normalizing the label shelf time and label star value of the marked label, take the geometric mean as the demand adjustment amount, and calculate the downward adjustment coefficient of the corresponding marked label through the harmonic formula: , where B is the tag demand coefficient of the tag, s is the preset reconciliation parameter, h is the demand adjustment of the tag, and T is the down-regulation coefficient of the tag; The product of the default skip time of the marker label and the down-adjustment coefficient is used as the skip time of the corresponding marker label.

Images