Product and service pushing method and system based on payment value chain model

Abstract

The invention discloses a product and service pushing method and system based on a payment value chain model, and belongs to the technical field of Internet product pushing, and the method specifically comprises the steps: collecting user payment value chain data in real time through closed-loop applications, such as digital catering, digital campus, and even digital business district, and carrying out the preprocessing; features are extracted, a mixed data model is constructed, user portraits are constructed based on the model, and user groups are divided through a clustering strategy; user derivative product demand information is identified, a value threshold value is set, a derivative product recommendation list meeting the demand is generated, and a collaborative filtering method is used for sorting; and according to the user portrait and the sorted recommendation list, a pushing strategy is formulated, the effect is tracked, and the strategy is optimized. According to the method, the recommendation accuracy and the user satisfaction degree are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of Internet product push, and more specifically, to a method and system for pushing products and services based on a payment value chain model. Background Art

[0002] With the development of mobile payments, payment data has become an important resource for companies to understand user consumption habits and predict consumption trends. However, most existing payment systems only focus on payment itself and lack in-depth mining and utilization of payment data, resulting in a lack of accuracy and personalization in product and service push.

[0003] For example, a Chinese patent application with publication number CN110689406A discloses a product push system based on the Internet, including: a data acquisition module, a server, a database, a product analysis module, a push analysis module, a product push module, a sharing task module, a sharing calculation module and a push degree sharing module; the product push module obtains the consumer's selected push product through the server and sends the push product information of the product to the consumer's corresponding push shopping platform within the push time period; thereby achieving accurate product push; through the push degree calculation, the consumer's acceptance of the push product information can be understood, and when the push degree is less than the set threshold, the product is stopped from being pushed to the consumer.

[0004] The above existing technologies have the following problems: lack of detailed data processing and analysis process; lack of real-time tracking and optimization of push effects; relatively single functional modules; push strategies mainly rely on consumers' selected push products and push time periods, without considering users' personalized needs and interest preferences. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a method and system for pushing products and services based on a payment value chain model. Through closed-loop applications such as digital catering, digital campus, and one-quarter digital shopping district, user payment value chain data is collected in real time and preprocessed; features are extracted and a hybrid data model is constructed, a user portrait is constructed based on the model, and user groups are divided through a clustering strategy; user derivative product demand information is identified, a value threshold is set, a recommended list of derivative products that meet the demand is generated, and the list is sorted using a collaborative filtering method; a push strategy is formulated based on the user portrait and the sorted recommendation list, the effect is tracked and the strategy is optimized. The present invention improves the accuracy of recommendations and user satisfaction.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] Methods for pushing products and services based on the payment value chain model include:

[0008] Step S1: collect the user's payment value chain data in real time and perform pre-processing;

[0009] Step S2: extract features from the preprocessed payment value chain data, and build a hybrid data model based on the characteristics of the payment value chain;

[0010] Step S3: Based on the user data in the hybrid data model, construct a user portrait and divide the users into different groups through a clustering strategy;

[0011] Step S4: Based on the user portraits after user group division, identify the user's demand information for derivative products, and at the same time, set a value threshold. According to the demand identification and value threshold, generate a derivative product recommendation list that meets the user's needs, and use the item-based collaborative filtering method to sort the derivative product recommendation list;

[0012] Step S5: Develop and implement a push strategy based on the divided user portraits and the sorted derivative product recommendation list. At the same time, track the push effect and adjust and optimize the push strategy based on the tracking results.

[0013] Specifically, the specific steps of step S4 include:

[0014] S4.1: Obtain the user portraits after being divided into user groups in step S3, wherein each user portrait belongs to a corresponding user group;

[0015] S4.2: Use machine learning algorithms to comprehensively analyze the behavior patterns and consumption habits of each user group in the user portrait after user group division, and predict the demand for derivative products of each user group;

[0016] S4.3: Collect the value and production cost data of derivative products, and evaluate the value of derivative products based on the derivative product demand information of the user group. The formula for calculating the value of derivative products is:

[0017] ;

[0018] in, represents the value of the derivative product, Indicates the value positioning of derivative products, represents the penetration rate of derivative products, represents the premium of the derivative product, Indicates the innovation value of derivative products;

[0019] S4.4: Set value thresholds based on value assessment results , and compare the value of the derivative product with the value threshold, where cost represents the raw material cost, production cost, and transportation cost of the derivative product, and profit represents the profit level obtained from the derivative product. It represents the profitability coefficient of the derivative product;

[0020] If the value of a derivative product exceeds the value threshold h, the derivative product is included in the derivative product recommendation list.

[0021] Specifically, the specific steps of step S4 also include:

[0022] S4.5: Based on demand identification and value threshold, filter derivative products, sort the filtered derivative products by relevance, and generate a preliminary derivative product recommendation list. The relevance formula is:

[0023] ;

[0024] Among them, c represents the correlation score between derivative products and user needs, Indicates the importance of the i-th keyword in user needs, Indicates the matching degree between derivative products and user demand keywords. Indicates the degree of match between the derivative product description and user needs, Indicates the user's evaluation score on the derivative product;

[0025] S4.6: Use a product-based collaborative filtering method to sort the products in the preliminary derivative product recommendation list to generate an optimized and sorted derivative product recommendation list.

[0026] Specifically, the specific steps of S4.6 include:

[0027] S4.61: Obtain pre-processed payment value chain data and a preliminary list of derivative product recommendations;

[0028] S4.62: Use the Pearson correlation coefficient to calculate the similarity between derivative products in the preliminary derivative product recommendation list , generate the derivative product similarity matrix, and adjust the similarity based on product popularity and user activity. The formula is:

[0029] ;

[0030] in, represents the adjusted derivative similarity, represents the popularity of derivative product j, represents the user activity related to derivative product j, represents the correction parameter, and is a positive number;

[0031] S4.63: Find the product set that the user likes based on the user's historical behavior, and find the product set that is most similar to the user's favorite products from the adjusted derivative product similarity matrix to obtain recommended products;

[0032] S4.64: Calculate the user's interest in the recommended products, sort the recommended products according to the interest, and generate an optimized sorted derivative product recommendation list.

[0033] Specifically, the interest calculation formula in S4.64 is:

[0034] ;

[0035] in, represents the interest of user u in the recommended product l, represents the association weight between recommended product l and recommended product k, represents the interest of user u in the recommended product k, represents the set of recommended products that user u likes, represents the set of m recommended products that are most similar to the recommended product l, Represents the synergy parameter between recommended product l and recommended product k.

[0036] Specifically, the specific steps of step S5 include:

[0037] S5.1: Determine the target user group for push notifications based on the user portrait segmentation results, and obtain a recommended list of derivative products with optimized sorting;

[0038] S5.2: Select products for the target user group from the optimized and ranked derivative product recommendation list and push them according to the preset push method;

[0039] S5.3: Configure the push address and API interface parameters of the push system, and send push messages to target users through the push system;

[0040] S5.4: Collect user behavior data after push notifications, and evaluate the push notification effect based on the collected user data;

[0041] S5.5: Based on the evaluation results, determine whether there are any problems with the push content and push time;

[0042] S5.6: If there are any problems, optimize the process of S5.2-S5.3 to obtain the optimized push strategy, and re-implement the optimized push strategy, while continuing to track the effect.

[0043] The product and service push system based on the payment value chain model includes: data processing module, model building module, user portrait building module, demand identification module, and push strategy formulation module;

[0044] The data processing module is used to collect the user's payment value chain data in real time and perform pre-processing operations;

[0045] The model building module is used to extract features from the preprocessed payment value chain data and build a hybrid data model based on the characteristics of the payment value chain;

[0046] The user portrait construction module constructs user portraits and divides user groups according to user data in the hybrid data model;

[0047] The demand identification module identifies the demand information of users for derivative products based on the user portraits after user group division, generates a derivative product recommendation list, and sorts the derivative product recommendation list;

[0048] The push strategy formulation module formulates a personalized push strategy based on the divided user portraits and the sorted derivative product recommendation list.

[0049] Specifically, the user portrait construction module includes: a user portrait construction unit and a group division unit;

[0050] The user portrait construction unit constructs a user portrait according to the user data in the hybrid data model;

[0051] The group division unit divides users into different groups using a clustering strategy.

[0052] Specifically, the demand identification module includes: a demand identification unit, a recommendation list generation unit, and a recommendation sorting unit;

[0053] The demand identification unit identifies the potential demand of each user group for the derivative product based on the user portrait and group segmentation results;

[0054] The recommendation list generating unit is used to set a value threshold, screen out derivative products that meet the user's needs according to the demand identification result and the value threshold, and generate a derivative product recommendation list;

[0055] The recommendation ranking unit uses an item-based collaborative filtering method to sort the derivative product recommendation list according to the user's historical evaluation and similarity of the derivative products.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention proposes a product and service push system based on the payment value chain model, and optimizes and improves the architecture, operation steps and processes. The system has the advantages of simple process, low investment and operation costs, and low production work costs.

[0058] 2. The present invention proposes a method for pushing products and services based on the payment value chain model. By collecting and analyzing the user's payment value chain data in real time, the method can accurately construct a user portrait and further divide the users into different groups. This detailed division helps to understand user needs more deeply, thereby generating a more personalized derivative product recommendation list. The collaborative filtering method based on items further improves the accuracy and relevance of the recommendations, allowing users to more easily find products that meet their needs.

[0059] 3. The present invention proposes a method for pushing products and services based on the payment value chain model. The method also focuses on the formulation and implementation of push strategies, as well as the tracking and optimization of effects. By continuously adjusting and optimizing push strategies, the method can ensure that users receive product recommendations that best meet their needs, thereby improving user experience and satisfaction. At the same time, real-time tracking of push effects also provides strong data support for subsequent improvements and optimizations, forming a virtuous circle of continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of a method for pushing products and services based on a payment value chain model according to the present invention;

[0061] Figure 2 This is a principle flow chart of a method for pushing products and services based on a payment value chain model of the present invention;

[0062] Figure 3 A flow chart for constructing a product recommendation list derived from the method for pushing products and services based on the payment value chain model of the present invention;

[0063] Figure 4 This is a system architecture diagram for pushing products and services based on the payment value chain model of the present invention. DETAILED DESCRIPTION

[0064] Example 1

[0065] See also Figure 1-Figure 3 An embodiment of the present invention provides a method for pushing products and services based on a payment value chain model, comprising the following steps:

[0066] Step S1: collect the user's payment value chain data in real time and perform pre-processing;

[0067] Among them, the collection of payment value chain data is based on the SaaS service provided by its own platform, that is, obtaining information such as consumer goods through its own closed-loop application. Among them, its own closed-loop applications include but are not limited to digital catering, digital campus, and one-minute digital business district.

[0068] Furthermore, the specific steps of step S1 include:

[0069] (1) Data collection: obtaining users’ payment data in real time, including but not limited to transaction time, transaction amount, transaction category, and transaction status, and storing the collected data in a database for subsequent processing and analysis;

[0070] (2) Data preprocessing: Clean the collected data, remove duplicate data and invalid data, format and normalize the data, and convert the data to the same dimension for subsequent analysis.

[0071] Step S2: extract features from the preprocessed payment value chain data, and build a hybrid data model based on the characteristics of the payment value chain;

[0072] Furthermore, the specific steps of step S2 include:

[0073] (1) Obtain preprocessed payment value chain data and divide it into training sets and test sets for subsequent model training and evaluation;

[0074] (2) Based on the characteristics of the payment value chain, principal component analysis is used to extract features from the pre-processed payment value chain data, including but not limited to transaction amount, transaction frequency, transaction time, transaction location, and user type. Principal component analysis is a prior art in this field and is not an inventive solution of the present application, and is not elaborated here.

[0075] (3) Based on the selected features and the characteristics of the payment value chain, a hybrid data model is constructed. The hybrid data model can combine the advantages of the relational model and the dimensional model to improve the performance and accuracy of the model. In addition, when constructing the model, it is necessary to select appropriate algorithms and parameters, and perform model training and tuning.

[0076] (4) Use the training set and test set to evaluate the mixed data model to assess its fit and prediction performance, and based on the evaluation results, optimize and adjust the mixed data model to improve its accuracy and generalization ability.

[0077] Step S3: Based on the user data in the hybrid data model, construct a user portrait and divide the users into different groups through a clustering strategy;

[0078] Furthermore, the specific steps of step S3 include:

[0079] (1) Extract user data from the hybrid data model, including user behavior data, transaction data, and personal information, and clean the data to remove duplicate, missing, or outlier values;

[0080] (2) According to business needs, determine the label system of user portraits, such as user age, gender, occupation, and consumption habits, use association rule mining to extract features related to the label system from user data, and build user portraits for users based on the extracted features, that is, generate a portrait containing multiple labels for each user. Among them, association rule mining is the existing technical content in this field and is not an inventive solution of this application, so it will not be elaborated here;

[0081] (3) Determine the number of clusters or thresholds based on the user portrait label system and data characteristics in order to divide users into different groups;

[0082] (4) Clustering the user portraits using the selected clustering strategy, and dividing the users into different groups based on the clustering results, each group having similar characteristics and behavior patterns. In the present invention, the clustering strategy adopts the K-means clustering algorithm, which is a prior art in the art and is not an inventive solution of the present application, and is not described in detail herein;

[0083] (5) Verify the clustering results to ensure the internal similarity and external differences of each group. Based on the verification results, optimize the clustering algorithm and parameters to improve the accuracy and stability of clustering.

[0084] Step S4: Based on the user portraits after user group division, identify the user's demand information for derivative products, and at the same time, set a value threshold. According to the demand identification and value threshold, generate a derivative product recommendation list that meets the user's needs, and use the item-based collaborative filtering method to sort the derivative product recommendation list;

[0085] For example, it is assumed that the following output has been obtained through step S3:

[0086] 1) User profile: User A likes sports and often buys sports shoes and fitness equipment;

[0087] 2) User group classification: User A belongs to the group of sports enthusiasts;

[0088] Then in step S4:

[0089] 1) Analyze purchase history and behavior patterns: It is found that users in the sports enthusiast group often buy sports shoes, fitness equipment and healthy food;

[0090] 2) Identify potential needs: Based on user portraits, user A may have a need for the latest sports shoes and smart bracelets;

[0091] 3) Set a value threshold: Set a medium-price threshold for sports enthusiasts to filter out reasonably priced sports shoes and smart bracelets;

[0092] 4) Generate a personalized recommendation list: Use collaborative filtering methods to recommend several of the latest sports shoes and smart bracelets to user A.

[0093] Step S5: Develop and implement a push strategy based on the divided user portraits and the sorted derivative product recommendation list. At the same time, track the push effect and adjust and optimize the push strategy based on the tracking results.

[0094] The specific steps of step S4 include:

[0095] S4.1: Obtain the user portraits after being divided into user groups in step S3, wherein each user portrait belongs to a corresponding user group;

[0096] S4.2: Use a machine learning algorithm to comprehensively analyze the behavior patterns and consumption habits of each user group in the user portrait after user group division, and predict the demand for derivative products of each user group, wherein the machine learning algorithm is the prior art content in this field, is not the inventive solution of this application, and is not described in detail here;

[0097] S4.3: Collect the value and production cost data of derivative products, and evaluate the value of derivative products based on the derivative product demand information of the user group. The formula for calculating the value of derivative products is:

[0098] ;

[0099] in, represents the value of the derivative product, Indicates the value positioning of derivative products, represents the penetration rate of derivative products, represents the premium of the derivative product, Indicates the innovation value of derivative products;

[0100] It should be noted that the formula in the present invention takes into account more comprehensively multiple key factors affecting the value of derivative products, including the value positioning of derivative products, the penetration rate of derivative products, the premium of derivative products and the innovative value of derivative products, thereby being able to more accurately evaluate the market competitiveness and long-term development potential of an enterprise.

[0101] S4.4: Set value thresholds based on value assessment results , and compare the value of the derivative product with the value threshold, where cost represents the raw material cost, production cost, and transportation cost of the derivative product, and profit represents the profit level obtained from the derivative product. It represents the profitability coefficient of the derivative product;

[0102] If the value of a derivative product exceeds the value threshold h, the derivative product is included in the derivative product recommendation list;

[0103] S4.5: Based on demand identification and value threshold, select qualified derivative products, sort qualified derivative products by relevance, and generate a preliminary derivative product recommendation list. The relevance formula is:

[0104] ;

[0105] Among them, c represents the correlation score between derivative products and user needs, Indicates the importance of the i-th keyword in user needs, Indicates the matching degree between derivative products and user demand keywords. Indicates the degree of match between the derivative product description and user needs, Indicates the user's evaluation score on the derivative product;

[0106] Furthermore, qualified derivative products refer to those that meet the user demand identification and value thresholds, where user demand identification means that the functions, characteristics or uses of the derivative products need to match the preset needs. The rules for determining qualified derivative products include:

[0107] (1) Demand matching: First, identify the specific needs that the derivative product needs to meet, and then compare these needs with the functions, characteristics or uses of the derivative product one by one to determine whether they match;

[0108] (2) Value assessment: Secondly, the value of the derivative product is assessed, which involves comprehensive consideration of multiple aspects such as the product's cost, price, performance, durability, etc. to determine whether it reaches the preset value threshold;

[0109] Comprehensive judgment: Finally, the results of demand matching and value assessment are comprehensively judged. If a derivative product can both meet the user's demand identification and reach the preset value threshold, then it can be regarded as a "qualified derivative product."

[0110] It should be noted that in the present invention, by comprehensively considering the matching degree and importance of keywords, the matching degree of product descriptions and user evaluation scores, it is possible to more comprehensively and accurately evaluate the correlation between content and target needs, and provide a more refined quantitative basis for scenarios such as information retrieval and product recommendations.

[0111] S4.6: Use a product-based collaborative filtering method to sort the products in the preliminary derivative product recommendation list to generate an optimized and sorted derivative product recommendation list.

[0112] The specific steps of S4.6 include:

[0113] S4.61: Obtain pre-processed payment value chain data and a preliminary list of derivative product recommendations;

[0114] S4.62: Use the Pearson correlation coefficient to calculate the similarity between derivative products in the preliminary derivative product recommendation list , generate the derivative product similarity matrix, and adjust the similarity based on product popularity and user activity. The formula is:

[0115] ;

[0116] in, represents the adjusted derivative similarity, represents the popularity of derivative product j, represents the user activity related to derivative product j, represents the correction parameter, and is a positive number. Meanwhile, the Pearson correlation coefficient is the prior art content in this field and is not the inventive solution of the present application, and will not be described in detail here.

[0117] S4.63: Find the product set that the user likes based on the user's historical behavior, and find the product set that is most similar to the user's favorite products from the adjusted derivative product similarity matrix to obtain recommended products;

[0118] Furthermore, the specific steps of S4.63 include:

[0119] (1) Based on the user's historical behavior, use the association rule mining method to find the user's favorite product set and analyze the user's preferences for products of different types, brands, prices, etc.;

[0120] (2) Obtain the derivative product similarity results and derivative product similarity matrix in S4.62;

[0121] (3) Find the derivative product set that is most similar to the user’s favorite derivative product from the derivative product similarity matrix;

[0122] (4) Generate recommended products.

[0123] S4.64: Calculate the user's interest in the recommended products, sort the recommended products according to the interest, and generate an optimized sorted derivative product recommendation list.

[0124] The calculation formula for interest in S4.64 is:

[0125] ;

[0126] in, represents the interest of user u in the recommended product l, represents the association weight between recommended product l and recommended product k, represents the interest of user u in the recommended product k, represents the set of recommended products that user u likes, represents the set of m recommended products that are most similar to the recommended product l, The synergy parameter between recommended product l and recommended product k can be calculated based on the common users and common tags of the items, reflecting whether the two items are often liked or recommended by users together.

[0127] The specific steps of step S5 include:

[0128] S5.1: Determine the target user group for push notifications based on the user portrait segmentation results, and obtain a recommended list of derivative products with optimized sorting;

[0129] S5.2: Select products for the target user group from the optimized and ranked derivative product recommendation list and push them according to the preset push method;

[0130] Furthermore, the specific steps of S5.2 include:

[0131] (1) According to the profile and preferences of the target user group, select suitable products from the optimized and sorted list of recommended derivative products. When selecting, factors such as product price, quality, and brand should be considered to ensure that the pushed products meet the needs of the target user group;

[0132] (2) According to the user's historical behavior, usage habits, etc., set the push method, such as email push, SMS push, APP push;

[0133] (3) Determine the time, frequency, and content of push notifications to ensure that the push notifications attract users’ attention and stimulate their desire to buy;

[0134] (4) Push the selected products to the target user group according to the preset push method.

[0135] S5.3: Configure the push address and API interface parameters of the push system, and send push messages to target users through the push system;

[0136] S5.4: Collect user behavior data after push, and evaluate the push effect based on the collected user data. The evaluation adopts the method of analyzing click-through rate, conversion rate, and user evaluation, which can be achieved through data analysis tools;

[0137] S5.5: Based on the evaluation results, determine whether there are any problems with the push content and push time;

[0138] S5.6: If there are any problems, optimize the process of S5.2-S5.3 to obtain the optimized push strategy, and re-implement the optimized push strategy, while continuing to track the effect.

[0139] Example 2

[0140] See also Figure 4 Another embodiment provided by the present invention is a product and service push system based on a payment value chain model, comprising:

[0141] Data processing module, model building module, user portrait building module, demand identification module, and push strategy formulation module;

[0142] The data processing module is used to collect users' payment value chain data in real time and perform pre-processing operations such as cleaning, deduplication, and format unification to ensure the data quality for subsequent analysis;

[0143] The model building module is used to extract key features from the pre-processed payment value chain data and build a hybrid data model based on the characteristics of the payment value chain to reflect the user's payment behavior and search history for subsequent user portrait construction and group segmentation;

[0144] The user portrait construction module builds detailed user portraits based on the user data in the hybrid data model and divides the user groups, where each group has similar consumption characteristics and needs;

[0145] The demand identification module identifies the demand information of users for derivative products based on the user portraits after user group division, generates a recommended list of derivative products, and sorts the recommended list of derivative products;

[0146] The push strategy formulation module formulates personalized push strategies based on the divided user portraits and the sorted derivative product recommendation list, tracks the push effects, and adjusts and optimizes the push strategies based on the tracking results.

[0147] The user portrait construction module includes: a user portrait construction unit and a group division unit;

[0148] The user portrait construction unit constructs a detailed user portrait containing basic user information and consumption habits based on the user data in the hybrid data model;

[0149] The group division unit uses clustering strategy to divide users into different groups, providing a basis for subsequent derivative product recommendations.

[0150] The demand identification module includes: a demand identification unit, a recommendation list generation unit, and a recommendation sorting unit;

[0151] Demand identification unit, which identifies the potential demand for derivative products of each user group based on user portraits and group segmentation results;

[0152] A recommendation list generating unit is used to set a value threshold, screen out derivative products that meet the user's needs according to the demand identification result and the value threshold, and generate a derivative product recommendation list;

[0153] The recommendation ranking unit uses an item-based collaborative filtering method to sort the derivative product recommendation list according to the user's historical evaluation and similarity of the derivative products, thereby improving the accuracy and relevance of the recommendations.

[0154] The push strategy formulation module includes: push strategy formulation unit, push implementation unit, and optimization unit;

[0155] The push strategy formulation unit formulates personalized push strategies based on user portraits and recommendation lists, including push time, push method, push content, etc.

[0156] The push implementation unit is used to implement the formulated push strategy and recommend derivative products to users.

[0157] The optimization unit is used to track push effects, including user click-through rate, conversion rate and other indicators, and adjust and optimize push strategies based on tracking results to improve push efficiency and user satisfaction.

[0158] Example 3

[0159] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for pushing products and services based on a payment value chain model are implemented. For details, please refer to the above method embodiment, which will not be repeated here.

[0160] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

[0161] If the disclosed technical solution involves personal information, the product using the disclosed technical solution has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the disclosed technical solution involves sensitive personal information, the product using the disclosed technical solution has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A method for pushing products and services based on a payment value chain model, characterized in that: include: Step S1: collect the user's payment value chain data in real time and perform pre-processing; Step S2: extract features from the preprocessed payment value chain data, and build a hybrid data model based on the characteristics of the payment value chain; Step S3: Based on the user data in the hybrid data model, construct a user portrait and divide the users into different groups through a clustering strategy; Step S4: Based on the user portraits after user group division, identify the user's demand information for derivative products, and at the same time, set a value threshold. According to the demand identification and value threshold, generate a derivative product recommendation list that meets the user's needs, and use the item-based collaborative filtering method to sort the derivative product recommendation list; Step S5: Develop and implement a push strategy based on the divided user portraits and the sorted derivative product recommendation list. At the same time, track the push effect and adjust and optimize the push strategy based on the tracking results.

2. The method for pushing products and services based on the payment value chain model according to claim 1, characterized in that: The specific steps of step S4 include: S4.1: Obtain the user portraits after being divided into user groups in step S3, wherein each user portrait belongs to a corresponding user group; S4.2: Use machine learning algorithms to comprehensively analyze the behavior patterns and consumption habits of each user group in the user portrait after user group division, and predict the demand for derivative products of each user group; S4.3: Collect the value and production cost data of derivative products, and evaluate the value of derivative products based on the derivative product demand information of the user group. The formula for calculating the value of derivative products is: ； in, represents the value of the derivative product, Indicates the value positioning of derivative products, represents the penetration rate of derivative products, represents the premium of the derivative product, Indicates the innovation value of derivative products; S4.4: Set value thresholds based on value assessment results , and compare the value of the derivative product with the value threshold, where cost represents the raw material cost, production cost, and transportation cost of the derivative product, and profit represents the profit level obtained from the derivative product. It represents the profitability coefficient of the derivative product; If the value of a derivative product exceeds the value threshold h, the derivative product is included in the derivative product recommendation list.

3. The method for pushing products and services based on the payment value chain model according to claim 2, characterized in that: The specific steps of step S4 also include: S4.5: Based on demand identification and value threshold, filter derivative products, sort the filtered derivative products by relevance, and generate a preliminary derivative product recommendation list. The relevance formula is: ； Among them, c represents the correlation score between derivative products and user needs, Indicates the importance of the i-th keyword in user needs, Indicates the matching degree between derivative products and user demand keywords. Indicates the degree of match between the derivative product description and user needs, Indicates the user's evaluation score on the derivative product; S4.6: Use a product-based collaborative filtering method to sort the products in the preliminary derivative product recommendation list to generate an optimized and sorted derivative product recommendation list.

4. The method for pushing products and services based on the payment value chain model according to claim 3, characterized in that: The specific steps of S4.6 include: S4.61: Obtain pre-processed payment value chain data and a preliminary list of derivative product recommendations; S4.62: Use the Pearson correlation coefficient to calculate the similarity between derivative products in the preliminary derivative product recommendation list , generate the derivative product similarity matrix, and adjust the similarity based on product popularity and user activity. The formula is: ； in, represents the adjusted derivative similarity, represents the popularity of derivative product j, represents the user activity related to derivative product j, represents the correction parameter, and is a positive number; S4.63: Find the product set that the user likes based on the user's historical behavior, and find the product set that is most similar to the user's favorite products from the adjusted derivative product similarity matrix to obtain recommended products; S4.64: Calculate the user's interest in the recommended products, sort the recommended products according to the interest, and generate an optimized sorted derivative product recommendation list.

5. The method for pushing products and services based on the payment value chain model according to claim 4, characterized in that: The interest calculation formula in S4.64 is: ； in, represents the interest of user u in the recommended product l, represents the association weight between recommended product l and recommended product k, represents the interest of user u in the recommended product k, represents the set of recommended products that user u likes, represents the set of m recommended products that are most similar to the recommended product l, Represents the synergy parameter between recommended product l and recommended product k.

6. The method for pushing products and services based on the payment value chain model according to claim 5, characterized in that: The specific steps of step S5 include: S5.1: Determine the target user group for push notifications based on the user portrait segmentation results, and obtain a recommended list of derivative products with optimized sorting; S5.2: Select products for the target user group from the optimized and ranked derivative product recommendation list and push them according to the preset push method; S5.3: Configure the push address and API interface parameters of the push system, and send push messages to target users through the push system; S5.4: Collect user behavior data after push notifications, and evaluate the push notification effect based on the collected user data; S5.5: Based on the evaluation results, determine whether there are any problems with the push content and push time; S5.6: If there are any problems, optimize the process of S5.2-S5.3 to obtain the optimized push strategy, and re-implement the optimized push strategy, while continuing to track the effect.

7. A product and service push system based on a payment value chain model, which is used to implement a product and service push method based on a payment value chain model as described in any one of claims 1 to 6, characterized in that: include: Data processing module, model building module, user portrait building module, demand identification module, and push strategy formulation module; The data processing module is used to collect the user's payment value chain data in real time and perform pre-processing operations; The model building module is used to extract features from the preprocessed payment value chain data and build a hybrid data model based on the characteristics of the payment value chain; The user portrait construction module constructs user portraits and divides user groups according to user data in the hybrid data model; The demand identification module identifies the demand information of users for derivative products based on the user portraits after user group division, generates a derivative product recommendation list, and sorts the derivative product recommendation list; The push strategy formulation module formulates a personalized push strategy based on the divided user portraits and the sorted derivative product recommendation list.

8. The product and service push system based on the payment value chain model as claimed in claim 7, characterized in that: The user portrait construction module includes: a user portrait construction unit and a group division unit; The user portrait construction unit constructs a user portrait according to the user data in the hybrid data model; The group division unit divides users into different groups using a clustering strategy.

9. The product and service push system based on the payment value chain model as claimed in claim 8, characterized in that: The demand identification module includes: a demand identification unit, a recommendation list generation unit, and a recommendation sorting unit; The demand identification unit identifies the potential demand of each user group for the derivative product based on the user portrait and group segmentation results; The recommendation list generating unit is used to set a value threshold, screen out derivative products that meet the user's needs according to the demand identification result and the value threshold, and generate a derivative product recommendation list; The recommendation ranking unit uses an item-based collaborative filtering method to sort the derivative product recommendation list according to the user's historical evaluation and similarity of the derivative products.

Images