A mall personalized service method and system based on S2B2C model

Abstract

The present invention discloses a method and system for personalized service of a shopping mall based on the S2B2C model, which relates to the technical field of business data processing, including analyzing the behavioral data of target consumers in the shopping mall platform to construct a consumption tag set of the target consumers, to determine the situation of each tag in the consumption tag set, and to provide a basis for subsequent personalized services. Consumer groups similar or close to the target consumers are screened by means of consumer screening indicators and the consumption tag sets of other consumers, so as to supplement the tag situation of the target consumers according to similar consumer groups, thereby improving the reliability and adaptability of the recommended commodity push service and improving the personalized service effect between merchants and consumers. Supply chain adjustment suggestions between target merchants and suppliers are generated in combination with commodity supply demand and evaluation indicators, providing a basis for supply chain adjustments between merchants and suppliers, thereby improving the synergy effect of the supply chain and improving consumer satisfaction.

Description

Technical Field

[0001] The present invention relates to the technical field of business data processing, and in particular to a mall personalized service method and system based on the S2B2C model. Background Art

[0002] In the digital age, as consumers' shopping habits become increasingly diversified and personalized, traditional retail models are facing tremendous pressure for change. The S2B2C (Supplier to Business to Consumer) model came into being. It integrates supply chain resources and uses big data and artificial intelligence technologies to build an efficient and convenient bridge between merchants and consumers. The core of this model is to use the power of the platform to empower the B-end (merchants or distributors) so that these small B-ends can better serve the C-end (consumers). Through functions such as intelligent customer service and automated order processing, the S2B2C model not only optimizes supply chain management and improves operational efficiency, but also provides consumers with more accurate and convenient services, thereby improving the overall user experience and business competitiveness. This innovative business model is gradually becoming an important engine for the development of modern retail.

[0003] In the prior art, the services of the mall are often targeted at all objects, and no customized services are provided based on individual personalization, resulting in poor effects and low adaptability of the mall services, which cannot meet changing needs.

[0004] Therefore, how to improve the effectiveness and adaptability of mall services is a technical problem that needs to be solved. Summary of the invention

[0005] The purpose of the present invention is to solve the problems of poor mall service effect and low adaptability in the prior art, and propose a mall personalized service method based on the S2B2C model, which is applied to a mall platform including suppliers, merchants and consumers working together. The method includes:

[0006] Obtain the behavior data of target consumers in the shopping mall platform and the basic information of commodities on the commodity platform, define all commodity tags of commodities according to the basic information of commodities on the commodity platform, analyze the behavior data of target consumers in the shopping mall platform to construct the consumption tag set of target consumers, thereby determining the consumption tag set of all consumers, which includes commodity preference tags and consumer information tags, and is marked with the label weights of the commodity preference tags and consumer information tags;

[0007] Generate consumer screening indicators through the product preference tags in the consumption tag set and the consumer's own information tags. Use the consumer screening indicators and the consumption tag set of other consumers to screen consumer groups that are similar or close to the target consumer, record them as consumer proximity groups, and determine the preferred product tag set of the consumer proximity group based on the consumption tag set of the consumer proximity group;

[0008] Update the weights of product preference tags based on the product preference tags in the target consumer's consumption tag set and the preferred product tag set of the consumer's proximity group, so as to find recommended products, push the recommended products to the target consumers, and adjust the consumption tag set based on the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers;

[0009] Obtain the target merchant's product order data, and integrate the target merchant's consumer audience based on the product order data, build the target merchant's product market tag set based on the consumer audience's consumption tag set, formulate product supply demand based on the product market tag set, calculate the supply chain evaluation index between the target merchant and the supplier, combine the product supply demand and the evaluation index to generate the supply chain adjustment suggestions between the target merchant and the supplier, and send the supply chain adjustment suggestions to the target merchant for analysis, so as to optimize the service between the merchant and the supplier.

[0010] In some embodiments of the present application, the behavior data of target consumers in the shopping mall platform is analyzed to construct a consumption tag set of the target consumers, including:

[0011] The behavior data includes direct behavior data and indirect behavior data. The indirect behavior data is analyzed to obtain a potential commodity list. The potential commodity list is marked with the behavior strength of each commodity. The direct behavior data is analyzed to obtain a purchase commodity list. The average behavior strength is determined according to the behavior strength of the commodities in the purchase commodity list. The commodities in the potential commodity list and the purchase commodity list are compared to determine the conversion rate. The behavior strength threshold is set in combination with the average behavior strength and the conversion rate.

[0012] Generate a first set of products according to the purchase product list, determine the difference set of the two lists through the potential product list and the purchase product list, filter products in the difference set through the behavior intensity threshold, generate a second set of products, assign different weights to the first set of products and the second set of products, and generate a product preference label according to the product label of each product in the first set of products and the second set of products;

[0013] Obtain the target consumer's own basic data, generate their own information tags based on their own basic data, calculate the correlation between product preference tags and their own information tags, quantify the respective label weights of product preference tags and consumer's own information tags, and thus form the target consumer's consumption label set.

[0014] In some embodiments of the present application, consumer screening indicators are generated by using product preference tags in the consumer tag set and consumer information tags, including:

[0015] Generate a highly correlated tag combination based on the correlation between the product preference tag and the product's own information tag, and generate a consumer screening index based on the highly correlated tag combination;

[0016] ;

[0017] in, For target consumers and Consumer screening indicators among consumers, , are the screening coefficients of highly correlated label combinations and overall labels, respectively. is the number of highly correlated label combinations, For target consumers and The first The similarity of highly correlated tag combinations, The tag set representing the target consumer and the The number of intersections of the tag sets of consumers is , Indicates the target consumers and The first The similarity of the tags, , Represents the target consumer’s tag set and The target consumer part and the first The combination coefficient of the consumer part, , are the target consumer part and the The sum of the label weights of the consumer parts, is the preset constant, The tag set representing the target consumer and the The number of labels of the target consumer in the difference of the label sets of consumers is , The tag set for the target consumer and The difference between the label sets of consumers in the target consumer part The tag weight of the tag represents the tag set of the target consumer and the The difference between the label sets of consumers The number of tags in the consumer part is , The tag set for the target consumer and The difference between the label sets of consumers Consumers in the The label weight of each label.

[0018] In some embodiments of the present application, determining the preferred product tag set of the consumer proximity group according to the consumer tag set of the consumer proximity group includes:

[0019] Conduct horizontal and vertical comparisons of each consumer's consumption tag set within the consumer proximity group;

[0020] The horizontal comparison is to sort the label weights of each consumer's preferred product labels and divide the preferred product labels into multiple sections;

[0021] The vertical comparison is to compare the preferred product label types among consumers to obtain the first common preferred product label type and the second common preferred product label type;

[0022] The label weights of the consumer's preferred product labels are normalized, and the preferred product label set of the consumer proximity group is determined by combining the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels under the two types belong.

[0023] In some embodiments of the present application, the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels of the two types belong are combined to determine the preferred product label set of the consumer proximity group, including:

[0024] The first common preferred product label type and the second common preferred product label type each correspond to a different label type coefficient, and the segment to which the preferred product label belongs corresponds to a different label segment coefficient;

[0025] For the same preferred product label, the label weight of the preferred product label is determined according to the label type coefficient and multiple label segment coefficients, thereby determining the preferred product label set of the consumer proximity group.

[0026] In some embodiments of the present application, the consumer audience group of the target merchant is integrated according to the commodity order data, and the commodity market tag set of the target merchant is constructed in combination with the consumption tag set of the consumer audience group, including:

[0027] Determine each consumer's consumption tag set and market demand in the product order data, use statistical analysis methods to determine the commonalities of the tags in these consumers' consumption tag sets, and build the target merchant's product market tag set based on market demand.

[0028] In some embodiments of the present application, the evaluation index of the supply chain between the target merchant and the supplier is calculated, including:

[0029] The evaluation index is obtained according to the stability index and the customization capability index. The stability index is composed of multiple supply chain stability parameters.

[0030] ;

[0031] in, As the evaluation index, , are the evaluation coefficients corresponding to the stability index and the customization capability index, is the number of supply chain stability parameters involved in the supply chain, For the The stability weight corresponding to the supply chain stability parameter is , Respectively The standard deviation and mean of the supply chain stability parameters, For customized capability indicators, For consumer satisfaction with the supply chain, is a preset constant.

[0032] Correspondingly, the present application also provides a personalized service system for a shopping mall based on the S2B2C model, which is applied to a shopping mall platform in which suppliers, merchants and consumers collaborate, including:

[0033] The analysis module is used to obtain the behavior data of the target consumers in the shopping mall platform and the basic information of the commodities on the commodity platform, define all the commodity tags of the commodities according to the basic information of the commodities on the commodity platform, analyze the behavior data of the target consumers in the shopping mall platform to construct the consumption tag set of the target consumers, thereby determining the consumption tag set of all consumers, which includes the commodity preference tags and the consumer's own information tags, and is annotated with the label weights of the commodity preference tags and the consumer's own information tags;

[0034] A screening module is used to generate a consumer screening index through the product preference tags in the consumption tag set and the consumer's own information tags, and to screen a consumer group similar or close to the target consumer by means of the consumer screening index and the consumption tag set of other consumers, and record them as consumer proximity groups, and determine the preferred product tag set of the consumer proximity group according to the consumption tag set of the consumer proximity group;

[0035] The push module is used to update the weights of product preference tags according to the product preference tags in the consumption tag set of the target consumer and the preferred product tag set of the consumer proximity group, so as to find recommended products, push the recommended products to the target consumer, and adjust the consumption tag set according to the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers;

[0036] The adjustment module is used to obtain the product order data of the target merchant, and integrate the consumer audience group of the target merchant based on the product order data, build the product market tag set of the target merchant in combination with the consumer tag set of the consumer audience group, formulate the product supply demand based on the product market tag set, calculate the evaluation index of the supply chain between the target merchant and the supplier, combine the product supply demand and the evaluation index to generate the supply chain adjustment suggestion between the target merchant and the supplier, and send the supply chain adjustment suggestion to the target merchant for analysis, so as to optimize the service between the merchant and the supplier.

[0037] By applying the above technical solutions, the behavior data of target consumers in the mall platform is analyzed to construct the consumption tag set of target consumers, so as to determine the situation of each tag in the consumption tag set and provide a basis for subsequent personalized services. Consumer groups similar or close to the target consumers are screened by relying on consumer screening indicators and the consumption tag sets of other consumers, so as to supplement the target consumers' tags according to similar consumer groups, improve the reliability and adaptability of the recommended product push service, and improve the personalized service effect between merchants and consumers. Supply chain adjustment suggestions between target merchants and suppliers are generated by combining product supply demand and evaluation indicators, providing a basis for supply chain adjustments between merchants and suppliers, thereby improving the synergy of the supply chain and improving consumer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flowchart of a personalized service method for a shopping mall based on the S2B2C model proposed by the present invention;

[0039] Figure 2 This is a structural diagram of a personalized service system for a shopping mall based on the S2B2C model proposed by the present invention. DETAILED DESCRIPTION

[0040] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0041] Reference Figure 1, a personalized service method for shopping malls based on the S2B2C model, applied to shopping mall platforms that include suppliers, merchants and consumers. The S2B2C model (Supply to Business to Customer) is a new e-commerce model that closely connects suppliers (Supplier), distributors / merchants (Business) and consumers (Customer) to form an efficient and flexible supply chain system. The core of this model is to use the power of the platform to empower the B-end (merchants or distributors) so that these small B-ends can better serve the C-end (consumers), while emphasizing the collaborative cooperation between the platform and merchants to achieve higher efficiency and lower costs through sharing resources, data and technology.

[0042] The method comprises the following steps:

[0043] Step S101, obtain the behavior data of the target consumer in the shopping mall platform and the basic information of the goods on the commodity platform, define all the commodity tags belonging to the goods according to the basic information of the goods on the commodity platform, analyze the behavior data of the target consumer in the shopping mall platform to construct the consumption tag set of the target consumer, so as to determine the consumption tag set of all consumers, the consumption tag set includes the commodity preference tag and the consumer's own information tag, and is marked with the label weights of the commodity preference tag and the consumer's own information tag.

[0044] In this embodiment, the behavior data includes browsing history, purchase history, search keywords, click behavior, dwell time, etc. Basic information of commodities on the commodity platform: commodity name, price, category, brand, description, picture, evaluation, etc. Commodity preference label: Generate commodity preference label based on consumer behavior data, such as commodity category, brand, price, etc. Consumer's own information label: including basic information such as age, gender, geographic location, occupation, income level, etc.

[0045] In some embodiments of the present application, the behavior data of target consumers in the shopping mall platform is analyzed to construct a consumption tag set of the target consumers, including:

[0046] The behavior data includes direct behavior data and indirect behavior data. The indirect behavior data is analyzed to obtain a potential commodity list. The potential commodity list is marked with the behavior strength of each commodity. The direct behavior data is analyzed to obtain a purchase commodity list. The average behavior strength is determined according to the behavior strength of the commodities in the purchase commodity list. The commodities in the potential commodity list and the purchase commodity list are compared to determine the conversion rate. The behavior strength threshold is set in combination with the average behavior strength and the conversion rate.

[0047] Generate a first set of products according to the purchase product list, determine the difference set of the two lists through the potential product list and the purchase product list, filter products in the difference set through the behavior intensity threshold, generate a second set of products, assign different weights to the first set of products and the second set of products, and generate a product preference label according to the product label of each product in the first set of products and the second set of products;

[0048] Obtain the target consumer's own basic data, generate their own information tags based on their own basic data, calculate the correlation between product preference tags and their own information tags, quantify the respective label weights of product preference tags and consumer's own information tags, and thus form the target consumer's consumption label set.

[0049] In this embodiment, direct behavior data refers to data related to purchase or subsequent data after purchase, and indirect behavior data refers to relevant data before purchase (clicks, browsing, favorites, etc.). The behavior intensity is a combination of clicks, browsing, favorites, etc. to indicate the interest before purchase. The behavior intensity threshold is set by combining the average behavior intensity and conversion rate (this threshold is used to filter out products that have not been purchased but are of great interest). Different conversion rates correspond to different adjustment coefficients, and the average behavior intensity * adjustment coefficient = behavior intensity threshold.

[0050] In this embodiment, different weights are assigned to the first commodity set and the second commodity set. Generally speaking, the weight of the first commodity set is greater than that of the second commodity set, because the first commodity has been purchased, while the second commodity has not been purchased but is of greater interest. The correlation between the commodity preference label and the own information label can be expressed by the Pearson correlation coefficient. For example, young women may prefer to buy fashionable and light luxury commodities, while middle-aged men may pay more attention to the practicality and quality of commodities.

[0051] Step S102, generate consumer screening indicators through product preference tags in the consumption tag set and consumer information tags, and use the consumer screening indicators and other consumers' consumption tag sets to screen consumer groups that are similar or close to the target consumer, recorded as consumer proximity groups, and determine the preferred product tag set of the consumer proximity group based on the consumption tag set of the consumer proximity group.

[0052] In this embodiment, the sample size may be small and insufficient for personalized recommendation service, depending only on the consumption tag set of the target consumer. Here, the situation of similar consumer groups is supplemented to help realize the service.

[0053] In some embodiments of the present application, consumer screening indicators are generated by using product preference tags in the consumer tag set and consumer information tags, including:

[0054] Generate a highly correlated tag combination based on the correlation between the product preference tag and the product's own information tag, and generate a consumer screening index based on the highly correlated tag combination;

[0055] ;

[0056] in, For target consumers and Consumer screening indicators among consumers, , are the screening coefficients of highly correlated label combinations and overall labels, respectively. is the number of highly correlated label combinations, For target consumers and The first The similarity of highly correlated tag combinations, The tag set representing the target consumer and the The number of intersections of the tag sets of consumers is , Indicates the target consumers and The first The similarity of the tags, , Represents the target consumer’s tag set and The target consumer part and the first The combination coefficient of the consumer part, , are the target consumer part and the The sum of the label weights of the consumer parts, is the preset constant, The tag set representing the target consumer and the The number of labels of the target consumer in the difference of the label sets of consumers is , The tag set for the target consumer and The difference between the label sets of consumers in the target consumer part The label weight of the label, The tag set representing the target consumer and the The difference between the label sets of consumers The number of tags in the consumer part is , The tag set for the target consumer and The difference between the label sets of consumers Consumers in the The label weight of each label.

[0057] In this embodiment, the highly correlated tag combination is a combination of the product preference tag and the self-information tag, and the consumer screening index is determined by partial highly correlated tag combinations, overall tag similarities, and overall tag differences. It represents the correction of the difference between two consumers to the combination of highly correlated labels and the overall label similarity.

[0058] In some embodiments of the present application, determining the preferred product tag set of the consumer proximity group according to the consumer tag set of the consumer proximity group includes:

[0059] Conduct horizontal and vertical comparisons of each consumer's consumption tag set within the consumer proximity group;

[0060] The horizontal comparison is to sort the label weights of each consumer's preferred product labels and divide the preferred product labels into multiple sections;

[0061] The vertical comparison is to compare the preferred product label types among consumers to obtain the first common preferred product label type and the second common preferred product label type;

[0062] The label weights of the consumer's preferred product labels are normalized, and the preferred product label set of the consumer proximity group is determined by combining the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels under the two types belong.

[0063] In this embodiment, the horizontal comparison is the comparison between consumers' own labels, and they are segmented according to the weight size. The vertical comparison is the comparison between multiple consumer label types. The first common preference product label type is that every consumer in the consumer close group has this label (preferred product label), and the second common preference product label type is that multiple consumers in the consumer close group have this label.

[0064] In some embodiments of the present application, the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels of the two types belong are combined to determine the preferred product label set of the consumer proximity group, including:

[0065] The first common preferred product label type and the second common preferred product label type each correspond to a different label type coefficient, and the segment to which the preferred product label belongs corresponds to a different label segment coefficient;

[0066] For the same preferred product label, the label weight of the preferred product label is determined according to the label type coefficient and multiple label segment coefficients, thereby determining the preferred product label set of the consumer proximity group.

[0067] In this embodiment, in order to better consider the label situation of consumers in the consumer proximity group, the segment to which the preferred product label belongs represents the proportion of intervals with different degrees of weight. For the same preferred product label, the segments to which different consumers belong may be different. An average segment coefficient is obtained by combining multiple label segment coefficients, and then combined with the label type coefficient to determine the label weight corresponding to each label under a unified standard.

[0068] Step S103, update the weight of the product preference tag according to the product preference tags in the consumption tag set of the target consumer and the preferred product tag set of the consumer proximity group, so as to find recommendable products, push the recommendable products to the target consumers, and adjust the consumption tag set according to the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers.

[0069] In this embodiment, the weight of the product preference label is dynamically adjusted by combining the product preference label of the target consumer and the product label set of the consumer's close group. Then, the recommended products are found based on the updated consumption label set of the target consumer. The target consumer's feedback on the pushed products is collected, such as click-through rate, purchase rate, evaluation, etc. Based on the feedback results, the labels and their weights in the consumption label set are adjusted to optimize the recommendation effect.

[0070] Step S104, obtaining the product order data of the target merchant, and integrating the consumer audience group of the target merchant based on the product order data, building the product market tag set of the target merchant in combination with the consumer tag set of the consumer audience group, formulating the product supply demand based on the product market tag set, calculating the evaluation index of the supply chain between the target merchant and the supplier, generating the supply chain adjustment suggestion between the target merchant and the supplier in combination with the product supply demand and the evaluation index, and sending the supply chain adjustment suggestion to the target merchant for analysis, thereby optimizing the service between the merchant and the supplier.

[0071] In this embodiment, the commodity order data includes commodity sales volume, sales amount, purchaser information, etc. Based on the order data, the consumer audience of the target merchant is analyzed and integrated. Combined with the consumer tag set of the consumer audience, the commodity market tag set of the target merchant is constructed to reflect market demand and trend. According to the commodity market tag set, a commodity supply demand plan is formulated.

[0072] In some embodiments of the present application, the consumer audience group of the target merchant is integrated according to the commodity order data, and the commodity market tag set of the target merchant is constructed in combination with the consumption tag set of the consumer audience group, including:

[0073] Determine each consumer's consumption tag set and market demand in the product order data, use statistical analysis methods to determine the commonalities of the tags in these consumers' consumption tag sets, and build the target merchant's product market tag set based on market demand.

[0074] In some embodiments of the present application, the evaluation index of the supply chain between the target merchant and the supplier is calculated, including:

[0075] The evaluation index is obtained according to the stability index and the customization capability index. The stability index is composed of multiple supply chain stability parameters.

[0076] ;

[0077] in, As the evaluation index, , are the evaluation coefficients corresponding to the stability index and the customization capability index, is the number of supply chain stability parameters involved in the supply chain, For the The stability weight corresponding to the supply chain stability parameter is , Respectively The standard deviation and mean of the supply chain stability parameters, For customized capability indicators, For consumer satisfaction with the supply chain, is a preset constant.

[0078] In this embodiment, in order to meet the personalized needs of consumers, customized goods need to be specially provided in the supply chain. The stability of the service between suppliers and merchants is the basis for merchants to continuously and stably provide products and services to consumers. If there is a disruption in the supply chain, merchants may face out-of-stock or delayed delivery, which will directly affect the consumer's shopping experience. The customization ability of suppliers to merchants determines how many unique and personalized products and services merchants can provide. The stronger this ability is, the more merchants can meet the diverse needs of consumers.

[0079] In this embodiment, supply chain stability parameters such as delivery time, inventory level, etc., the stability index is described by the coefficient of variation, and the customization capability index is a combination of the sales proportion of customized products, the development speed and flexibility of customized products, and the quality and cost-effectiveness of customized products. It means that the sum of stability index and customization capability index is modified by consumers' satisfaction with the supply chain, so as to provide personalized collaborative services among suppliers, distributors / businesses and consumers.

[0080] By applying the above technical solutions, the behavior data of target consumers in the mall platform is analyzed to construct the consumption tag set of target consumers, so as to determine the situation of each tag in the consumption tag set and provide a basis for subsequent personalized services. Consumer groups similar or close to the target consumers are screened by relying on consumer screening indicators and the consumption tag sets of other consumers, so as to supplement the target consumers' tags according to similar consumer groups, improve the reliability and adaptability of the recommended product push service, and improve the personalized service effect between merchants and consumers. Supply chain adjustment suggestions between target merchants and suppliers are generated by combining product supply demand and evaluation indicators, providing a basis for supply chain adjustments between merchants and suppliers, thereby improving the synergy of the supply chain and improving consumer satisfaction.

[0081] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0082] Correspondingly, the present application also provides a personalized service system for a shopping mall based on the S2B2C model, which is applied to a shopping mall platform in which suppliers, merchants and consumers collaborate with each other. Figure 2 As shown, including,

[0083] The analysis module is used to obtain the behavior data of the target consumers in the shopping mall platform and the basic information of the commodities on the commodity platform, define all the commodity tags of the commodities according to the basic information of the commodities on the commodity platform, analyze the behavior data of the target consumers in the shopping mall platform to construct the consumption tag set of the target consumers, thereby determining the consumption tag set of all consumers, which includes the commodity preference tags and the consumer's own information tags, and is annotated with the label weights of the commodity preference tags and the consumer's own information tags;

[0084] A screening module is used to generate a consumer screening index through the product preference tags in the consumption tag set and the consumer's own information tags, and to screen a consumer group similar or close to the target consumer by means of the consumer screening index and the consumption tag set of other consumers, and record them as consumer proximity groups, and determine the preferred product tag set of the consumer proximity group according to the consumption tag set of the consumer proximity group;

[0085] The push module is used to update the weights of product preference tags according to the product preference tags in the consumption tag set of the target consumer and the preferred product tag set of the consumer proximity group, so as to find recommended products, push the recommended products to the target consumer, and adjust the consumption tag set according to the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers;

[0086] The adjustment module is used to obtain the product order data of the target merchant, and integrate the consumer audience group of the target merchant based on the product order data, build the product market tag set of the target merchant in combination with the consumer tag set of the consumer audience group, formulate the product supply demand based on the product market tag set, calculate the evaluation index of the supply chain between the target merchant and the supplier, combine the product supply demand and the evaluation index to generate the supply chain adjustment suggestion between the target merchant and the supplier, and send the supply chain adjustment suggestion to the target merchant for analysis, so as to optimize the service between the merchant and the supplier.

[0087] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required for implementing the present invention.

[0088] Those skilled in the art will appreciate that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more systems different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0089] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A personalized service method for a shopping mall based on the S2B2C model, applied to a shopping mall platform including suppliers, merchants and consumers, characterized in that: include, Obtain the behavior data of target consumers in the shopping mall platform and the basic information of commodities on the commodity platform, define all commodity tags of commodities according to the basic information of commodities on the commodity platform, analyze the behavior data of target consumers in the shopping mall platform to construct the consumption tag set of target consumers, thereby determining the consumption tag set of all consumers, which includes commodity preference tags and consumer information tags, and is marked with the label weights of the commodity preference tags and consumer information tags; Generate consumer screening indicators through product preference tags in the consumer tag set and consumer information tags. Use consumer screening indicators and other consumers' consumer tag sets to screen out consumer groups that are similar or close to the target consumer, and record them as consumer proximity groups. Determine the preferred product tag set of the consumer proximity group based on the consumer tag set of the consumer proximity group. Update the weights of product preference tags based on the product preference tags in the target consumer's consumption tag set and the preferred product tag set of the consumer's proximity group, so as to find recommended products, push the recommended products to the target consumers, and adjust the consumption tag set based on the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers; Obtain the target merchant's product order data, and integrate the target merchant's consumer audience based on the product order data, build the target merchant's product market tag set based on the consumer audience's consumption tag set, formulate product supply demand based on the product market tag set, calculate the supply chain evaluation index between the target merchant and the supplier, combine the product supply demand and the evaluation index to generate the supply chain adjustment suggestions between the target merchant and the supplier, and send the supply chain adjustment suggestions to the target merchant for analysis, so as to optimize the service between the merchant and the supplier.

2. The mall personalized service method based on the S2B2C model according to claim 1, characterized in that: Analyze the behavior data of target consumers in the mall platform to build a consumption tag set of target consumers, including: The behavior data includes direct behavior data and indirect behavior data. The indirect behavior data is analyzed to obtain a potential commodity list. The potential commodity list is marked with the behavior strength of each commodity. The direct behavior data is analyzed to obtain a purchase commodity list. The average behavior strength is determined according to the behavior strength of the commodities in the purchase commodity list. The commodities in the potential commodity list and the purchase commodity list are compared to determine the conversion rate. The behavior strength threshold is set in combination with the average behavior strength and the conversion rate. Generate a first set of products according to the purchase product list, determine the difference set of the two lists through the potential product list and the purchase product list, filter products in the difference set through the behavior intensity threshold, generate a second set of products, assign different weights to the first set of products and the second set of products, and generate a product preference label according to the product label of each product in the first set of products and the second set of products; Obtain the target consumer's own basic data, generate their own information tags based on their own basic data, calculate the correlation between product preference tags and their own information tags, quantify the respective label weights of product preference tags and consumer's own information tags, and thus form the target consumer's consumption label set.

3. The mall personalized service method based on the S2B2C model according to claim 2 is characterized in that: Generate consumer screening indicators through product preference tags in the consumer tag set and consumer information tags, including: Generate a highly correlated tag combination based on the correlation between the product preference tag and the product's own information tag, and generate a consumer screening index based on the highly correlated tag combination; ； in, For target consumers and Consumer screening indicators among consumers, , are the screening coefficients of highly correlated label combinations and overall labels, respectively. is the number of highly correlated label combinations, For target consumers and The first The similarity of highly correlated tag combinations, The tag set representing the target consumer and the The number of intersections of the tag sets of consumers is , Indicates the target consumers and The first The similarity of the tags, , Represents the target consumer’s tag set and The target consumer part and the first The combination coefficient of the consumer part, , are the target consumer part and the The sum of the label weights of the consumer parts, is the preset constant, The tag set representing the target consumer and the The number of labels of the target consumer in the difference of the label sets of consumers is , The tag set for the target consumer and The difference between the label sets of consumers in the target consumer part The label weight of the label, The tag set representing the target consumer and the The difference between the label sets of consumers The number of tags in the consumer part is , The tag set for the target consumer and The difference between the label sets of consumers Consumers in the The label weight of each label.

4. The mall personalized service method based on the S2B2C model according to claim 1, characterized in that: Determine the preferred product label set of the consumer proximity group based on the consumer label set of the consumer proximity group, including: Conduct horizontal and vertical comparisons of each consumer's consumption tag set within the consumer proximity group; The horizontal comparison is to sort the label weights of each consumer's preferred product labels and divide the preferred product labels into multiple sections; The vertical comparison is to compare the preferred product label types among consumers to obtain the first common preferred product label type and the second common preferred product label type; The label weights of the consumer's preferred product labels are normalized, and the preferred product label set of the consumer proximity group is determined by combining the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels under the two types belong.

5. The mall personalized service method based on the S2B2C model according to claim 4 is characterized in that: Combining the first common preferred product label type, the second common preferred product label type and the sections to which the preferred product labels under the two types belong, determining a preferred product label set of a consumer proximity group includes: The first common preferred product label type and the second common preferred product label type each correspond to a different label type coefficient, and the segment to which the preferred product label belongs corresponds to a different label segment coefficient; For the same preferred product label, the label weight of the preferred product label is determined according to the label type coefficient and multiple label segment coefficients, thereby determining the preferred product label set of the consumer proximity group.

6. The mall personalized service method based on the S2B2C model according to claim 1, characterized in that: And integrate the consumer audience of the target merchant based on the product order data, and build the product market label set of the target merchant based on the consumer label set of the consumer audience, including: Determine each consumer's consumption tag set and market demand in the product order data, use statistical analysis methods to determine the commonalities of the tags in these consumers' consumption tag sets, and build the target merchant's product market tag set based on market demand.

7. The mall personalized service method based on the S2B2C model according to claim 1, characterized in that: Calculate the evaluation index of the supply chain between the target merchant and the supplier, include, The evaluation index is obtained according to the stability index and the customization capability index. The stability index is composed of multiple supply chain stability parameters. ； in, As the evaluation index, , are the evaluation coefficients corresponding to the stability index and the customization capability index, is the number of supply chain stability parameters involved in the supply chain, For the The stability weight corresponding to the supply chain stability parameter is , Respectively The standard deviation and mean of the supply chain stability parameters, For customized capability indicators, For consumer satisfaction with the supply chain, is a preset constant.

8. A personalized service system for a shopping mall based on the S2B2C model, applied to a shopping mall platform involving suppliers, merchants and consumers, characterized in that: include, The analysis module is used to obtain the behavior data of the target consumers in the shopping mall platform and the basic information of the commodities on the commodity platform, define all the commodity tags of the commodities according to the basic information of the commodities on the commodity platform, analyze the behavior data of the target consumers in the shopping mall platform to construct the consumption tag set of the target consumers, thereby determining the consumption tag set of all consumers, which includes the commodity preference tags and the consumer's own information tags, and is annotated with the label weights of the commodity preference tags and the consumer's own information tags; A screening module is used to generate a consumer screening index through the product preference tags in the consumption tag set and the consumer's own information tags, and to screen a consumer group similar or close to the target consumer by means of the consumer screening index and the consumption tag set of other consumers, and record them as consumer proximity groups, and determine the preferred product tag set of the consumer proximity group according to the consumption tag set of the consumer proximity group; The push module is used to update the weights of product preference tags according to the product preference tags in the consumption tag set of the target consumer and the preferred product tag set of the consumer proximity group, so as to find recommended products, push the recommended products to the target consumer, and adjust the consumption tag set according to the subsequent monitoring results of the push service, so as to optimize the service between merchants and consumers; The adjustment module is used to obtain the product order data of the target merchant, and integrate the consumer audience group of the target merchant based on the product order data, build the product market tag set of the target merchant in combination with the consumer tag set of the consumer audience group, formulate the product supply demand based on the product market tag set, calculate the evaluation index of the supply chain between the target merchant and the supplier, combine the product supply demand and the evaluation index to generate the supply chain adjustment suggestion between the target merchant and the supplier, and send the supply chain adjustment suggestion to the target merchant for analysis, so as to optimize the service between the merchant and the supplier.

Images