Adaptive construction method of product portrait

Abstract

The present invention relates to the field of data analysis technology, and specifically to a method for adaptively constructing product portraits, comprising the following steps: one, recording an analyzed supermarket as a target supermarket, and screening out a plurality of adjacent supermarkets that meet the reference requirements of the target supermarket; two, weightedly calculating a similarity evaluation index between the target supermarket and any adjacent supermarket based on multiple similarity value analyses; using the product portrait of the adjacent supermarket with the largest similarity evaluation index as the initial portrait of the target supermarket; three, constructing a product category set and a price set of the target supermarket based on the initial portrait, marking the products as flexible products or fixed products, and dividing the flexible products into low-end products, mid-end products, and high-end products according to the prices; and four, adjusting the initial portrait based on the customer's purchase order data, wherein the adjustment is divided into a product category set adjustment and a price set adjustment.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a method for adaptively constructing product portraits. Background Art

[0002] Profiling is a method that integrates information about a target group's characteristics, behaviors, needs, and preferences. Through data collection, analysis, and integration, it forms one or more representative virtual representations. It aims to help companies gain a deeper understanding of the target audience for their services or products, enabling more accurate decisions in product design, marketing strategies, and user experience optimization. Both user and product profiling are crucial tools for companies in digital transformation and market competition, aiming to improve business efficiency and effectiveness through a data-driven approach.

[0003] Product portraits are usually used in product planning, product marketing and other fields, and mainly serve the early stage of product development. However, the existing technology lacks a product portrait construction method that can be applied in the field of shopping malls and supermarkets. It is also impossible to comprehensively consider users' consumption levels and consumption habits based on multiple data when constructing product portraits. It is difficult to clearly define the product category positioning and pricing strategy, resulting in the procurement and shelving of goods easily deviating from the consumption levels and consumption habits of users in the region. Summary of the Invention

[0004] In response to the above-mentioned shortcomings of the existing technology, the present invention provides a method for adaptively constructing product portraits, which can effectively solve the problem in the existing technology that it is difficult to integrate multiple data to consider user consumption levels and consumption habits when constructing product portraits in shopping malls and supermarkets.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0006] The present invention provides a method for adaptively constructing a product portrait, comprising the following steps:

[0007] Step 1: The supermarket being analyzed is designated as the target supermarket. Based on the target supermarket's geographic region and the per capita GDP of each statistical jurisdiction, multiple adjacent supermarkets with operating hours that meet the threshold requirements are selected.

[0008] Step 2: The target supermarket and all adjacent supermarkets are collectively designated as the analysis supermarket. Based on the total business area of ​​the analysis supermarket, a radiation area is drawn for the analysis supermarket. The impact value of the analysis supermarket is calculated based on the population and per capita income of different circular areas within the radiation area.

[0009] Compare multiple adjacent supermarkets with the target supermarket one by one:

[0010] Calculate area similarity based on total business area, type similarity based on planned area of ​​different commodity categories, impact similarity based on scope impact, and customer group similarity based on the proportion of different types of ground buildings.

[0011] Calculate the similarity evaluation index between the target supermarket and any adjacent supermarket based on multiple similarity value analyses and weighting;

[0012] The product profile of the neighboring supermarket with the largest similarity evaluation index is used as the initial profile of the target supermarket;

[0013] Step 3: Based on the initial profile, a set of product categories and price points for the target supermarket is constructed. The prices of different products in the target supermarket and adjacent supermarkets are analyzed. The target products are marked as either flexible or fixed. Flexible products are then categorized into low-end, mid-end, and high-end products based on their prices.

[0014] Step 4: Adjust the initial profile based on the customer's purchase order data. The adjustments are divided into product category set adjustment and price set adjustment;

[0015] Product category aggregation adjustment adjusts product inventory based on the profit share and value share of each product category;

[0016] The price set adjustment adjusts the flexible product prices based on the sales proportion of low-end products, mid-end products, and high-end products.

[0017] Furthermore, the screening process of adjacent supermarkets is as follows:

[0018] Divide the map into multiple geographical regions based on geographical location and natural conditions, divide the geographical regions into multiple statistical jurisdictions, and obtain the per capita GDP in each statistical jurisdiction;

[0019] Obtain the geographical area and statistical jurisdiction to which the target supermarket belongs, record them as the target area and target jurisdiction respectively. Filter out multiple statistical jurisdictions within the target area that are less than the preset distance threshold from the target jurisdiction and record them as adjacent jurisdictions. Obtain the per capita GDP of any adjacent jurisdiction and the target jurisdiction, record them as GDP1 and GDP0 respectively, and substitute them into the formula Calculation is performed, and when the calculation result is within a preset range, the adjacent jurisdiction is recorded as a similar jurisdiction;

[0020] The business hours of each supermarket are obtained, and multiple supermarkets with business hours greater than or equal to a preset threshold in similar jurisdictions are selected as adjacent supermarkets.

[0021] Furthermore, the range impact value calculation process is as follows:

[0022] Record the supermarket with the calculated impact value as the analysis supermarket, obtain the total business area of ​​the analysis supermarket, and substitute it into the formula Calculate the radiation index G of the analyzed supermarket, where:

[0023] S is the total business area of ​​the analyzed supermarket;

[0024] e is a natural constant;

[0025] S′ is the preset area standard value;

[0026] k is a preset constant coefficient;

[0027] a is a preset constant index and the value range of a is;

[0028] Indicates that the value is not less than The smallest integer;

[0029] Draw a circle with a radius of Z*R with the location of the supermarket as the center, and record it as the radiation area. R is the preset length value. Split the circle into multiple circular areas with a width of R and record them as AREA. j , where j = 1, 2, ..., G, to obtain the population number people in different circular areas j and per capita annual income j , substitute into the formula Calculate the range impact value DF of the analyzed supermarket area , where λ1 and λ2 are preset weight coefficients.

[0030] Furthermore, the customer group similarity value calculation process is as follows:

[0031] Obtain the radiation area of ​​adjacent supermarkets and target supermarkets, divide the surface buildings into commercial buildings, residential buildings and office buildings, and obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the radiation area of ​​adjacent supermarkets, which are recorded as P A ′1, P A ′2, P A ′3, obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the target supermarket radiation area and record them as P B ′1, P B ′2、P B ′3, substitute into the formula Calculation is performed to obtain the customer group similarity value Similar4, where h = 1, 2, 3.

[0032] Furthermore, the target product labeling process is as follows:

[0033] Get the selling price of the target product in each adjacent supermarket and take the maximum and minimum values ​​to construct the selling price range [Sale0, Sale1];

[0034] Obtain the official suggested retail price of the target product and record it as the anchor price. If the target product does not have an official suggested retail price, obtain the midpoint of the price range and record it as the anchor price.

[0035] Substitute into the formula Calculate the price fluctuation index β sale , where Sale′ represents the anchored selling price and a floating threshold is preset. When the price floating index is greater than or equal to the preset floating threshold, the target product is marked as a flexible product. When the price floating index is less than the preset floating threshold, the product is marked as a fixed product.

[0036] When the target product is a flexible product, split the target product's price range [Sale0, Sale1] into the first interval Second interval and the third interval, When the price of the target product in the price set is in the first interval, the second interval, or the third interval, it is recorded as a low-order product, a mid-order product, or a high-order product.

[0037] Furthermore, different customers are marked as member customers and non-member customers. Each member customer has a unique member ID. An adjustment period is set, and all customer purchase orders are recorded within the adjustment period. The purchase orders include the purchased goods and the sales price of the purchased goods. The profit per item purchased is calculated based on the sales price of the purchased goods.

[0038] The total sales volume of each commodity during the adjustment period is calculated and multiplied by the single-unit profit to obtain the total cycle profit of each commodity. The total cycle profit of all commodities during the adjustment period is calculated to obtain the total profit.

[0039] Furthermore, the process of adjusting the product category set is as follows:

[0040] Obtain the inventory quantity of different products in the product category set and multiply it by the profit per unit to obtain the inventory value of each product. Calculate the sum of the inventory values ​​of all products in the product category set to obtain the total value. Divide the products in the product category set into different categories and calculate the sum of the inventory values ​​of products in each category as the category value. Calculate the category value of each category and divide it by the total value to obtain the value share of each category.

[0041] Obtain the total period profit of each product in the same product category and sum them up to get the category profit of the product category. Divide the category profit by the total profit in the adjustment period to get the profit share of each product category. Adjust the product inventory in each product category based on the profit share and value share. Specifically:

[0042] When the profit share minus the value share is lower than the preset share threshold, the product category is recorded as a reduced category, and the total period profit profit1 of each product in the reduced category and the total period profit profit2 of similar products are obtained and substituted into the formula Calculation is performed to obtain the reduction reference value Re1 of each product. The inventory of the products is reduced in descending order according to the reduction reference value, where φ is a preset value. When the product is a flexible product, the value is 1, and when the product is a fixed product, the value is 2.

[0043] Furthermore, the price set adjustment process is as follows:

[0044] Obtain all purchase orders within the adjustment period and record them as order Oeder n , n is the order number;

[0045] When ordering Oeder n When the corresponding customer is a member customer, get the order Oeder n The quantity of low-end goods, mid-end goods, and high-end goods is multiplied by a preset expansion coefficient to obtain the order demand of low-end goods, mid-end goods, and high-end goods, respectively, which are recorded as Oeder n x 、Oeder n y 、Oeder n z , where x, y, and z represent low-order goods, mid-order goods, and high-order goods, respectively;

[0046] When ordering Oeder n When the corresponding customer is a non-member customer, get the order Oeder n The quantity of low-end goods, mid-end goods, and high-end goods is recorded as the order demand of low-end goods, mid-end goods, and high-end goods as Oeder n x 、Oeder n y 、Oeder n z ;

[0047] Substitute into the formula Calculate in and get the demand NUM for low-level goods, mid-level goods, and high-level goods during the adjustment period. X 、NUM Y 、NUM Z ;

[0048] When the demand for low-level goods reaches its maximum, the selling prices of mid-level and high-level goods are reduced;

[0049] When the demand for mid-level products reaches its maximum, increase the price of low-level products and reduce the price of high-level products;

[0050] When the demand for high-end products reaches its maximum, increase the selling prices of low-end and high-end products.

[0051] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0052] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0053] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0054] 1. The adjacent supermarkets screened out by the present invention through the time threshold and the distance threshold can be used as the source of horizontal comparison data for the target supermarket. The similar supermarkets obtained through multiple comparisons and screening have similar scale, operating environment and regional characteristics as the target supermarket, and the product portraits of the two are highly overlapped. By screening out similar supermarkets and citing their product portraits, the target supermarket can determine the positioning and pricing of its own products when it first opens for business, so as to suit its own scale and the local consumption environment.

[0055] 2. The commodity category set adjustment process of the present invention can compare the value share and profit share to screen out commodity categories with low profits. By reducing the reference value to adjust the inventory of commodities in the commodity category, the inventory quantity of low-profit commodities on sale in supermarkets can be reduced, and the product portrait can be optimized to improve inventory circulation efficiency. The price set adjustment process can adjust the prices of different flexible commodities based on the demand for low-end commodities, mid-end commodities, and high-end commodities within the adjustment period, and can improve the degree of consistency between product portraits and customer consumption levels under the premise of customer acceptance, thereby further improving the construction effect of product portraits. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0057] Figure 1 Schematic diagram of the overall method of the present invention. DETAILED DESCRIPTION

[0058] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0059] The present invention will be further described below with reference to the embodiments.

[0060] See Figure 1 , the adaptive construction method of product portraits at least includes:

[0061] Step 1: The supermarket being analyzed is recorded as the target supermarket. Based on the data characteristics of the target supermarket, the surrounding adjacent supermarkets are selected, where:

[0062] Divide the map into multiple geographical regions based on geographical location and natural conditions (taking China as an example, it is divided into Northeast China, East China, Southwest China, and South China), and then divide the geographical regions into multiple statistical jurisdictions to obtain the per capita GDP in each statistical jurisdiction;

[0063] The target supermarket's geographical area and statistical jurisdiction are obtained, recorded as the target area and target jurisdiction respectively. Within the target area, multiple statistical jurisdictions that are less than a preset distance threshold from the target jurisdiction are selected and recorded as adjacent jurisdictions (in a specific embodiment, the distance threshold is 100km). The per capita GDP of any adjacent jurisdiction and the target jurisdiction is obtained, recorded as GDP1 and GDP0 respectively, and substituted into the formula Calculation is performed, and when the calculation result is within a preset range, the adjacent jurisdiction is recorded as a similar jurisdiction;

[0064] Obtain the business hours of each supermarket (i.e., the length of time the supermarket has been in operation since its opening), and select multiple supermarkets in similar jurisdictions whose business hours are greater than or equal to a preset time threshold as neighboring supermarkets (in a specific embodiment, the distance threshold is 1 year);

[0065] It should be noted that, on the one hand, the distance threshold screening determines that the adjacent supermarkets and the target supermarket have similar regional characteristics (affecting regional consumption habits), and the GDP comparison determines that the customer groups served have similar consumption habits. On the other hand, the duration threshold determines that the adjacent supermarkets have stable operations and can be used as a comparison reference. Therefore, the adjacent supermarkets screened by the duration threshold and distance threshold can be used as a source of horizontal comparison data for the target supermarket, which is convenient for the target supermarket to determine the initial product portrait.

[0066] Step 2: Further analyze the adjacent supermarkets and compare them with the target supermarket to identify similar supermarkets with similar business conditions as the target supermarket. Based on the similar supermarkets, determine the initial portrait of the target supermarket's product portrait, where

[0067] Multiple adjacent supermarkets are compared with the target supermarket one by one. The comparison process is as follows (for data before the target supermarket opens):

[0068] Obtain the total business area of ​​the adjacent supermarket and the target supermarket, record them as S1 and S0 respectively, and substitute them into the formula Calculation is performed to obtain the area similarity value Similar1, which reflects to a certain extent the similarity of the overall business scale between the adjacent supermarkets and the target supermarket;

[0069] There are multiple commodity categories in advance. The planned area ratio of different commodity categories in adjacent supermarkets is obtained and recorded as P 1i , obtain the planned area proportion of different commodity categories in the target supermarket and record it as P 0i , where i represents different types of serial numbers (i.e. P 1i represents the ratio of the planned sales area of ​​the i-th commodity category in the adjacent supermarket to the total area of ​​the supermarket), and substitute it into the formula The calculation is performed to obtain the type similarity value Similar2. The planned area ratio of different product categories reflects the category structure in the supermarket product portrait. The type similarity value is used to compare the similarity of the product categories sold between adjacent supermarkets and the target supermarket.

[0070] It should be noted that the determination of major commodity categories is based on the commodity divisions within supermarkets, such as the fresh food area, dry goods area, daily necessities area, wine and beverage area, and home appliance area.

[0071] Calculate the range impact value of the target supermarket and each adjacent supermarket separately. The calculation process is as follows:

[0072] Record the supermarket with the calculated impact value as the analysis supermarket, obtain the total business area of ​​the analysis supermarket, and substitute it into the formula Calculate the radiation index G of the analyzed supermarket, where:

[0073] S is the total business area of ​​the analyzed supermarket;

[0074] e is a natural constant;

[0075] S′ is a preset standard area value (in a specific embodiment, the value is 100 square meters);

[0076] k is a preset constant coefficient (in a specific embodiment, the value is 2);

[0077] a is a preset constant exponent and the value range of a is (0.5, 1);

[0078] Indicates that the value is not less than The smallest integer (i.e. the radiation index G is a positive integer greater than 0);

[0079] Draw a circle with a radius of Z*R with the location of the supermarket being analyzed as the center, and record it as the radiation area. R is a preset length value (in a specific embodiment, the value is 500 meters). The circle is divided into multiple ring areas with a width of R (the difference between the inner and outer radii is equal to R) (a circle with a radius of R is regarded as a ring with an inner radius of 0 and an outer radius of R) and recorded as AREA. j , where j = 1, 2, ..., G, to obtain the population number people in different circular areas j and per capita annual income j , substitute into the formula Calculate the range impact value DF of the analyzed supermarket area , where λ1 and λ2 are preset weight coefficients (in a specific embodiment, the values ​​are 0.5 and 0.5);

[0080] It should be noted that, in a specific embodiment, the value of the population size adopts the number of permanent residents, and the per capita annual income can be obtained based on statistical data or questionnaire surveys.

[0081] Obtain the range influence values ​​of the target supermarket and any adjacent supermarket, record them as DF0 and DF1 respectively, and substitute them into the formula Calculation is performed to obtain the influence similarity value Similar3, which reflects to some extent the similarity in the degree to which the overall business prospects of adjacent supermarkets and the target supermarket are affected by the population distribution and per capita income in the surrounding area.

[0082] Obtain the radiation area of ​​adjacent supermarkets and target supermarkets, divide the surface buildings into commercial buildings, residential buildings and office buildings, and obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the radiation area of ​​adjacent supermarkets based on the map, which are recorded as P. A ′1, P A ′2、P A ′3, obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the target supermarket radiation area and record them as P B ′1, P B ′2、P B ′3, substitute into the formula Calculate in and get the customer group (customer group) similarity value Similar4, where h = 1, 2, 3;

[0083] Commercial buildings, residential buildings and office buildings correspond to ground buildings mainly used for consumption, residence and office purposes respectively. The distinction between different types of ground buildings can be based on the proportion of internal space usage of the building. For example, if more than a certain proportion (60%) of the space in a building is used for residence, the ground building can be recorded as a residential building, such as a common apartment building. The different types of ground buildings used will determine the consumption habits of the people. For example, people in residential buildings will buy daily necessities, while people in office buildings will tend to buy office supplies and food and beverages.

[0084] Obtain the area similarity value Similar1, type similarity value Similar2, influence similarity value Similar3, and customer group similarity value Similar4 between the target supermarket and any adjacent supermarket, and substitute them into the formula Similar all =η1*Similar1+η2*Similar2+η3*Similar3+η4*Similar4 to obtain the similarity evaluation index Similar all , where η1, η2, η3, and η4 are all preset weight coefficients, and satisfy η1+η2+η3+η4=1. The adjacent supermarket with the largest similarity evaluation index is selected as the similar supermarket of the target supermarket, and the product portrait of the similar supermarket is used as the initial portrait of the target supermarket.

[0085] Generally speaking, similar supermarkets obtained through multiple comparisons and screening have similar scale, operating environment and regional characteristics as the target supermarket (these factors dominate the consumption habits and consumption levels of supermarket customers, and consumption habits and consumption levels are the key conditions for adaptation required for the construction of product portraits). Therefore, the product portraits of the two are highly overlapping. It should be noted that the product portrait in the present invention refers to the category composition and pricing standards of products in the supermarket. When implemented specifically, it refers to the category list and price list of each commodity in the supermarket. By screening out similar supermarkets and quoting their product portraits, the target supermarket can determine the positioning and pricing of the goods it sells when it initially opens for business, so as to suit its own scale and local consumption environment.

[0086] Step 3: Based on the initial profile, construct a product category set and a price set for the target supermarket. The product category set includes multiple different types of products, and the price set includes the price of each product. Perform price analysis for different products, including:

[0087] The analyzed product is recorded as the target product. The selling price of the target product in each adjacent supermarket is obtained and the maximum and minimum values ​​are taken to construct the selling price range [Sale0, Sale1].

[0088] Obtain the official suggested retail price of the target product and record it as the anchor price. If the target product does not have an official suggested retail price, obtain the midpoint of the price range and record it as the anchor price.

[0089] Substitute into the formula Calculate the price fluctuation index β sale , where Sale′ represents the anchored selling price and is preset with a floating threshold (in a specific embodiment, the value is 0.1). When the price floating index is greater than or equal to the preset floating threshold, the target product is marked as a flexible product. When the price floating index is less than the preset floating threshold, the product is marked as a fixed product.

[0090] Furthermore, when the target product is a flexible product, the price range of the target product [Sale0, Sale1] is split into the first interval Second interval and the third interval, When the price of the target product in the price set is in the first interval, it is recorded as a low-order product. When the price of the target product in the price set is in the second interval, it is recorded as a mid-order product. When the price of the target product in the price set is in the third interval, it is recorded as a high-order product.

[0091] It should be noted that the target product is specific to a certain brand, model, or batch, so that the target products of different supermarkets are comparable. When the target product is a flexible product, it means that the product has room for price adjustment, and customers are more accepting of price changes of this product. In other words, the selling price of the target product can be adjusted within a certain range according to the demand for the target product, thereby increasing profits without losing too much sales. When the target product is a fixed product, it means that the price of the product in the market has no room for adjustment, and customers are more sensitive to the price of this product.

[0092] Furthermore, when it is impossible to construct a price range based on the prices of the target product in adjacent supermarkets, similar products of the target product are identified and the price range of the similar products is used as the price range of the target product. The similar products and the target product must meet the following conditions:

[0093] Condition 1: The similar product and the target product are of the same type (the same type is determined based on the product name, international design classification, and the product's classification and location on the shelf);

[0094] Condition 2: The price difference between similar products of the same type and the target product is minimal;

[0095] Condition 3: Similar products have a range of prices.

[0096] Step 4: Optimize and adjust the initial profile based on customer purchase data. The adjustments are divided into product category set adjustments and price set adjustments.

[0097] Different customers are marked as members and non-members. Each member has a unique membership ID to distinguish them at checkout. By distinguishing between non-members and members, the influence coefficients of different customers are adjusted. Because members have a stronger demand for supermarket products than non-members, an adjustment period is set. During this adjustment period, all customer purchase orders are recorded. The purchase orders include the purchased items and the purchase price. The profit per item is calculated based on the purchase price.

[0098] The total profit for each product during the adjustment period is calculated by summing the sales volume of each product and multiplying it by the profit per unit. The total profit for each product during the adjustment period is then calculated by summing the total profits of all products during the adjustment period.

[0099] The process of adjusting the product category set is as follows:

[0100] Obtain the inventory quantity of different products in the product category set and multiply it by the profit per unit to obtain the inventory value of each product. Calculate the sum of the inventory values ​​of all products in the product category set to obtain the total value. Divide the products in the product category set into different categories and calculate the sum of the inventory values ​​of products in each category as the category value. Calculate the category value of each category and divide it by the total value to obtain the value share of each category.

[0101] Obtain the total period profit of each product in the same product category and sum them up to get the category profit of the product category. Divide the category profit by the total profit in the adjustment period to get the profit share of each product category. Adjust the product inventory in each product category based on the profit share and value share. Specifically:

[0102] When the profit share minus the value share is lower than the preset share threshold, the product category is recorded as a reduced category, and the total period profit profit1 of each product in the reduced category and the total period profit profit2 of similar products are obtained and substituted into the formula Calculation is performed to obtain the reduction reference value Re1 of each product. The inventory of the products is reduced in descending order according to the reduction reference value, where φ is a preset value. When the product is a flexible product, the value is 1, and when the product is a fixed product, the value is 2.

[0103] It should be noted that by comparing the value share and profit share, we can screen out the major categories of goods with low profits. By reducing the reference value to adjust the inventory of goods in the major categories, we can reduce the inventory of low-profit goods on sale in supermarkets, optimize product portraits, and improve inventory turnover efficiency.

[0104] The price aggregate adjustment process is as follows:

[0105] Calculate the demand for low-end products, mid-end products, and high-end products during the adjustment period. The demand calculation process is as follows:

[0106] Obtain all purchase orders within the adjustment period and record them as order Oeder n , n is the order number;

[0107] When ordering Oeder n When the corresponding customer is a member customer, get the order Oeder n The quantity of low-end goods, mid-end goods, and high-end goods is multiplied by a preset expansion coefficient (in a specific embodiment, the value is 1.5) to obtain the order demand of low-end goods, mid-end goods, and high-end goods, respectively, which are recorded as Oeder n x 、Oeder n y 、Oeder n z , where x, y, and z represent low-order goods, mid-order goods, and high-order goods, respectively;

[0108] When ordering Oeder n When the corresponding customer is a non-member customer, get the order Oeder n The quantity of low-end goods, mid-end goods, and high-end goods is recorded as the order demand of low-end goods, mid-end goods, and high-end goods as Oeder n x 、Oeder n y 、Oeder n z ;

[0109] Substitute into the formula Calculate in and get the demand NUM for low-level goods, mid-level goods, and high-level goods during the adjustment period. X 、NUM Y 、NUM Z ;

[0110] When the demand for low-level goods reaches its maximum, the selling prices of mid-level and high-level goods are reduced;

[0111] When the demand for mid-level products reaches its maximum, increase the price of low-level products and reduce the price of high-level products;

[0112] When the demand for high-end products reaches its maximum, increase the selling prices of low-end and high-end products.

[0113] It should be noted that the proportion of low-end, mid-end and high-end goods in purchase orders to a certain extent illustrates the user's acceptance of high-priced goods. Adjusting the prices of different flexible goods based on the demand for low-end, mid-end and high-end goods within the adjustment period can improve the degree of consistency between product portraits and customer consumption levels under the premise of customer acceptance, and further improve the effect of product portrait construction.

[0114] A computer device includes a memory and a processor, wherein the memory stores a computer program and the processor implements the steps of the above method when executing the computer program.

[0115] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0116] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for adaptively constructing a product portrait, characterized in that: The following steps are involved: Step 1: The supermarket being analyzed is designated as the target supermarket. Based on the target supermarket's geographic region and the per capita GDP of each statistical jurisdiction, multiple adjacent supermarkets with operating hours that meet the threshold requirements are selected. Step 2: The target supermarket and all adjacent supermarkets are collectively designated as the analysis supermarket. Based on the total business area of ​​the analysis supermarket, a radiation area is drawn for the analysis supermarket. The impact value of the analysis supermarket is calculated based on the population and per capita income of different circular areas within the radiation area. Compare multiple adjacent supermarkets with the target supermarket one by one: Calculate area similarity based on total business area, type similarity based on planned area of ​​different commodity categories, impact similarity based on scope impact, and customer group similarity based on the proportion of different types of ground buildings. Calculate the similarity evaluation index between the target supermarket and any adjacent supermarket based on multiple similarity value analyses and weighting; The product profile of the neighboring supermarket with the largest similarity evaluation index is used as the initial profile of the target supermarket; Step 3: Based on the initial profile, a set of product categories and price points for the target supermarket is constructed. The prices of different products in the target supermarket and adjacent supermarkets are analyzed. The target products are marked as either flexible or fixed. Flexible products are then categorized into low-end, mid-end, and high-end products based on their price. Step 4: Adjust the initial profile based on the customer's purchase order data. The adjustments are divided into product category set adjustment and price set adjustment; Product category aggregation adjustment adjusts product inventory based on the profit share and value share of each product category; The price set adjustment adjusts the flexible product prices based on the sales proportion of low-end products, mid-end products, and high-end products.

2. The method for adaptively constructing a product portrait according to claim 1, characterized in that: The screening process for adjacent supermarkets is as follows: Divide the map into multiple geographical areas based on geographical location and natural conditions, divide the geographical areas into multiple statistical jurisdictions, and obtain the per capita GDP in each statistical jurisdiction; The target supermarket's geographical area and statistical jurisdiction are recorded as the target area and target jurisdiction respectively. In the target area, multiple statistical jurisdictions whose distance from the target jurisdiction is less than the preset distance threshold are selected and recorded as adjacent jurisdictions. The per capita GDP of any adjacent jurisdiction and the target jurisdiction is recorded as , substitute into the formula Calculation is performed, and when the calculation result is within a preset range, the adjacent jurisdiction is recorded as a similar jurisdiction; Obtain the business hours of each supermarket, and filter out multiple supermarkets in similar areas whose business hours are greater than or equal to a preset business hour threshold and record them as adjacent supermarkets.

3. The method for adaptively constructing a product portrait according to claim 1, characterized in that: The scope impact value calculation process is as follows: Record the supermarket with the calculated impact value as the analysis supermarket, obtain the total business area of ​​the analysis supermarket, and substitute it into the formula Calculate the radiation index G of the analyzed supermarket, where: S is the total business area of ​​the analyzed supermarket; e is a natural constant; is the preset area standard value; k is a preset constant coefficient; a is a preset constant exponent and the value range of a is (0.5, 1); Indicates that the value is not less than The smallest integer; Draw a circle with a radius of Z*R with the location of the supermarket as the center, and record it as the radiation area. R is the preset length value. Split the circle into multiple ring areas with a width of R and record them as , where j = 1, 2, ..., G, to obtain the population in different circular areas and per capita annual income , substitute into the formula Calculate the impact value of the supermarket range analyzed ,in is the preset weight coefficient.

4. The method for adaptively constructing a product portrait according to claim 1, characterized in that: The customer group similarity calculation process is as follows: Obtain the radiation area of ​​adjacent supermarkets and target supermarkets, divide the surface buildings into commercial buildings, residential buildings and office buildings, obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the radiation area of ​​adjacent supermarkets and record them as , obtain the floor area ratio of commercial buildings, residential buildings and office buildings in the target supermarket radiation area and record them as , substitute into the formula Calculate and get the customer group similarity value , where h=1,2,3.

5. The method for adaptively constructing a product portrait according to claim 1, characterized in that: The target product tagging process is as follows: Get the selling price of the target product in each adjacent supermarket and take the maximum and minimum values ​​to construct the selling price range ; Obtain the official suggested retail price of the target product and record it as the anchor price. If the target product does not have an official suggested retail price, obtain the midpoint of the price range and record it as the anchor price. Substitute into the formula Calculate the price fluctuation index ,in Indicates an anchored selling price with a preset floating threshold. When the price floating index is greater than or equal to the preset floating threshold, the target product is marked as a flexible product. When the price floating index is less than the preset floating threshold, the product is marked as a fixed product. When the target product is a flexible product, the price range of the target product Split into the first interval , the second interval and the third interval, ,When the price of the target product in the price set is in the first interval, the second interval, or the third interval, it is recorded as a low-order product, a mid-order product, or a high-order product.

6. The method for adaptively constructing a product portrait according to claim 1, characterized in that: Different customers are marked as member customers and non-member customers. Each member customer has a unique member ID. An adjustment period is set. All customer purchase orders are recorded during the adjustment period. The purchase orders include the purchased items and the sales price. The profit per item is calculated based on the sales price. The total sales volume of each commodity during the adjustment period is calculated and multiplied by the single-unit profit to obtain the total cycle profit of each commodity. The total cycle profit of all commodities during the adjustment period is calculated to obtain the total profit.

7. The method for adaptively constructing a product portrait according to claim 6, characterized in that: The process of adjusting the product category set is as follows: Obtain the inventory quantity of different products in the product category set and multiply it by the profit per unit to obtain the inventory value of each product. Calculate the sum of the inventory values ​​of all products in the product category set to obtain the total value. Divide the products in the product category set into different categories and calculate the sum of the inventory values ​​of products in each category as the category value. Calculate the category value of each category and divide it by the total value to obtain the value share of each category. Obtain the total period profit of each product in the same product category and sum them up to get the category profit of the product category. Divide the category profit by the total profit in the adjustment period to get the profit share of each product category. Adjust the product inventory in each product category based on the profit share and value share. Specifically: When the profit share minus the value share is lower than the preset share threshold, the commodity category is recorded as a reduced category, and the total period profit of each commodity in the reduced category is obtained. and the total profit of similar products during the period , substitute into the formula Calculate and obtain the reduction reference value of each commodity , reduce the inventory of goods in descending order according to the reduction reference value, where It is a preset value. When the product is a flexible product, the value is 1. When the product is a fixed product, the value is 2.

8. The method for adaptively constructing a product portrait according to claim 7, characterized in that: The price aggregate adjustment process is as follows: Get all purchase orders within the adjustment period and record them as orders , n is the order number; When the order Get the order when the corresponding customer is a member customer The quantity of low-end goods, mid-end goods, and high-end goods is multiplied by a preset expansion coefficient to obtain the order demand of low-end goods, mid-end goods, and high-end goods, respectively, which are recorded as , where x, y, and z represent low-order goods, mid-order goods, and high-order goods, respectively; When the order Get the order when the corresponding customer is a non-member customer The quantity of low-end goods, mid-end goods, and high-end goods is recorded as the order demand of low-end goods, mid-end goods, and high-end goods respectively. ; Substitute into the formula Calculate the demand for low-level goods, mid-level goods, and high-level goods during the adjustment period. ; When the demand for low-level goods reaches its maximum, the selling prices of mid-level and high-level goods are reduced; When the demand for mid-level products reaches its maximum, increase the price of low-level products and reduce the price of high-level products; When the demand for high-end products reaches its maximum, increase the selling prices of low-end and high-end products.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images