A commodity management method, device, and storage medium based on identification resolution

Abstract

The present application discloses a method, a device, and a storage medium, which include: collecting identification codes of commodities of the same type, obtaining attribute information of the commodities based on the identification codes, obtaining a sample set, and generating multidimensional variables based on the attribute information; receiving user selection to determine a target variable, determining a sample distribution based on the target variable and the number of multidimensional variables, and determining the form of a multiple regression equation based on the properties of the sample distribution; using other variables in the multidimensional variables except the target variable as feature variables, inputting the feature variables into the multiple regression equation to obtain an output value of each sample; determining a cost function of the multiple regression equation based on the difference between the output value and the actual value of the target variable; determining a minimum value of the cost function, determining parameters in the multiple regression equation based on the minimum value, and determining an expression of the multiple regression equation based on the parameters; and obtaining a predicted value of the output target variable based on the corresponding multiple regression equation, so that the user can manage the commodities.

Description

Technical Field

[0001] The present application relates to the field of machine learning, and in particular to a commodity management method, device, and storage medium based on identity resolution. Background Art

[0002] The Industrial Internet's identification resolution system is a crucial component of the Industrial Internet's network layer. Like the Industrial Internet's neural network, it uses identification codes to give devices and products an identity. The information recorded in these codes serves as a profile for the objects, and identification resolution technology enables the storage and access of this profile information.

[0003] Although product information can be checked through identification codes, it is difficult for product sellers to quickly predict product profits, prices, expected inventory and other characteristics that affect the merchant's interests based on the product type, cost, size, logistics time, etc., so that merchants can better adjust and replenish products. Summary of the Invention

[0004] This application provides a commodity management method, device, and storage medium based on identity resolution, which solves the technical problem of difficulty in predicting the risks and profits of commodities based on commodity attributes.

[0005] A commodity management method based on identity resolution, applied to a commodity management system, includes:

[0006] Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information;

[0007] Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution;

[0008] Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample;

[0009] determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable;

[0010] Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters;

[0011] The predicted value of the output target variable is obtained according to the corresponding multivariate regression equation, so that the user can manage the product.

[0012] In one embodiment of the present application, the method of receiving user selection to determine the target variable, determining the sample distribution according to the target variable and the number of the multidimensional variables, and determining the form of the multivariate regression equation according to the properties of the sample distribution specifically includes: receiving user selection to determine the target variable; determining the domain and range of the sample distribution under the condition of the target variable; determining the change of the target variable of the sample distribution within the range, and determining the properties of the sample distribution according to the change; the properties of the sample distribution include monotonicity and parity in the domain; matching the properties of the sample distribution with the properties of the multivariate regression equation; and determining the form of the multivariate regression equation with the highest degree of matching.

[0013] In one embodiment of the present application, determining the cost function of the multiple regression equation based on the difference between the output value and the actual value of the target variable specifically includes: determining the difference between the output value after the characteristic variable of each sample is input into the multiple regression equation and the actual value of the target variable, and determining the square of the difference of each sample; adding the squares of the difference of each sample to obtain a sum of squares; dividing the sum of squares by the number of samples to obtain a mean of the sum of squares; determining an expression for the mean of the sum of squares, and determining a cost function; the cost function includes the parameters of the multiple regression equation.

[0014] In one embodiment of the present application, determining the minimum value of the cost function, determining the parameters in the multiple regression equation based on the minimum value, and determining the expression of the multiple regression equation based on the parameters specifically include: determining the partial derivative of the cost function with respect to each parameter in the multiple regression equation; updating each parameter in the multiple regression equation based on the partial derivative; when the partial derivative of any parameter is determined to be 0, no longer updating the parameter with a partial derivative of 0; when the partial derivatives of all parameters in the cost function are 0, determining each parameter value, substituting the parameter value into the multiple regression equation, and obtaining the expression of the multiple regression equation.

[0015] In one embodiment of the present application, a predicted value of the output target variable is obtained according to the corresponding multivariate regression equation so that the user can manage the goods, specifically including: accepting the price attributes of the goods queried by the user, and determining the predicted price of the goods; determining the cost of the goods according to the purchase price, logistics, and storage of the goods; determining the profit of the goods according to the cost of the goods and the predicted price of the goods; and screening the suppliers of the goods according to a preset profit value.

[0016] In one embodiment of the present application, when a user queries a product, the method further includes: obtaining a product identification code, obtaining attribute information of the product according to the product identification code as a multi-dimensional variable, the multi-dimensional variable including price attribute, cost attribute, sales attribute, and profit attribute; obtaining sample data of similar products according to the prefix of the product identification code to form a sample set; receiving the profit attribute of the product queried by the user, and generating a target variable y according to the profit attribute, and determining the distribution of the sample attribute in the model; generating corresponding feature variables x for the attributes other than the profit attribute in the multi-dimensional variable; inputting the sample set into the model for training, determining the multivariate regression equation of the queried product, and deriving the predicted profit value according to the multivariate regression equation; comparing different categories of products according to the profit value, and determining the type of product with the highest profit value; determining the effective inventory of the product with the highest profit value; calculating the saleable time according to the effective inventory; determining a replenishment plan when it is judged that the saleable time is less than the sales time; and determining a call-out plan when it is judged that the saleable time is greater than the sales time.

[0017] In one embodiment of the present application, after collecting product identification codes of the same type, the method further includes: obtaining the product identification code, obtaining the corresponding URL identifier based on the product identification code; accessing the secondary node based on the URL identifier, determining the information server address recorded in the secondary node, accessing the information server address and establishing communication with the information server to obtain relevant product information.

[0018] In one embodiment of the present application, before collecting identification codes of goods of the same type, the method further includes: identifying the identification code information on the scanned goods; receiving basic information of the goods input by the staff; binding the identification code with the basic information of the goods; classifying the identification code according to preset classification rules; and storing the identification code and the basic information corresponding to the identification code in a database according to the classification results.

[0019] A commodity management device based on identity resolution, comprising:

[0020] at least one processor; and,

[0021] a memory communicatively connected to the at least one processor; wherein,

[0022] The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0023] Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information;

[0024] Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution;

[0025] Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample;

[0026] determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable;

[0027] Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters;

[0028] The predicted value of the output target variable is obtained according to the corresponding multivariate regression equation, so that the user can manage the product.

[0029] A non-volatile storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to:

[0030] Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information;

[0031] Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution;

[0032] Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample;

[0033] determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable;

[0034] Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters;

[0035] The predicted value of the output target variable is obtained according to the corresponding multivariate regression equation, so that the user can manage the product.

[0036] The present application provides a commodity management method, device, and storage medium based on identity resolution, which at least include the following beneficial effects: by using machine learning, the processing process can be automated, the calculation of data can be more accurate, and large-scale data can be calculated, so that the project can be supported by data and help people make better decisions. According to the various attributes of the commodity, various variables are generated, and a prediction model is obtained by training the model. The data is predicted by the machine, which reduces labor costs and can quickly predict prices, sales, profits, etc., providing sellers with better purchasing and shipping plans, helping merchants to achieve better management of stock adjustment and replenishment. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0038] Figure 1 A schematic diagram of steps of a product management method based on identity resolution provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of the composition of a commodity management device based on identity resolution provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solutions and advantages of this application more clear, the following will be a clear and complete description of this application in conjunction with the specific embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0041] An identifier can be understood as a name tag used to identify different objects, entities, or IoT objects. It can be a string of numbers, letters, symbols, and text that follows a certain pattern. Identity resolution is the technical process of "information transformation" that accurately and efficiently obtains the corresponding information of an object identifier in a complex network environment.

[0042] When merchants sell goods, they often hope to make more profits. However, since goods are priced differently in different regions, and even different merchants have different prices for the same goods, if merchants want to know at what pricing conditions they can make more profits and how to allocate goods in order to maximize profits, this requires professional knowledge to calculate. However, ordinary merchants generally do not hire such personnel, and it is also difficult to find such people.

[0043] This application provides a commodity management method, device, and storage medium based on identity resolution, which are described in detail below.

[0044] Figure 1 A schematic diagram of a method for managing products based on identity resolution provided in an embodiment of the present application may include the following steps:

[0045] S101: Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information.

[0046] Specifically, for clothing and footwear, after scanning a product identification code and parsing the code, we can obtain product attributes such as brand, size, price, sales volume of the same style, material, cost, and profit. Based on this attribute information, we can generate corresponding variables. Based on the product identifier, we can query the prefix to obtain the attribute information of similar products and obtain a sample set.

[0047] In one embodiment of the present application, before collecting identification codes of goods of the same type, the identification code information on the scanned goods is identified; the basic information of the goods entered by the staff is received; the identification code is bound to the basic information of the goods; the identification code is classified according to preset classification rules; and the identification code and the basic information corresponding to the identification code are stored in a database according to the classification results.

[0048] Specifically, before a product is identified, its basic information needs to be bound to the identification code. This requires the staff to input the basic information of the product into the device and distinguish the types of products based on the prefix of the identification code string, such as using the prefix character A- to represent food, character F- to represent clothing, etc., and then bind it together with the identification code and store it in the database.

[0049] In one embodiment of the present application, after collecting product identification codes of the same type, the product identification code is obtained, and the corresponding URL identifier is obtained based on the product identification code; the secondary node is accessed based on the URL identifier, the information server address recorded in the secondary node is determined, the information server address is accessed and communication is established with the information server to obtain relevant product information.

[0050] Specifically, based on the URL identifier, the server address storing the product identification code can be obtained. First, based on the URL information in the identifier query request, an identifier resolution query is initiated to the national top-level node resolution server, and the URL address information of the second-level node resolution server returned by the national top-level node resolution server is received. Then, an identifier query request is sent to the URL address information of the second-level node resolution server, and the URL address information of the enterprise resolution system sent by the second-level node resolution server is received. Then, an identifier query request is sent to the URL address information of the enterprise resolution system, and the product identification data sent by the enterprise resolution system is received.

[0051] S102: Receive user selection to determine the target variable, determine the sample distribution according to the target variable and the number of multi-dimensional variables, and determine the form of the multivariate regression equation according to the properties of the sample distribution.

[0052] In one embodiment of the present application, a user selection is received to determine a target variable; the domain and range of the sample distribution in the case of the target variable are determined; the change of the target variable of the sample distribution within the range is determined, and the properties of the sample distribution are determined based on the change; the properties of the sample distribution include monotonicity and parity in the domain; the properties of the sample distribution are matched with the properties of the multiple regression equation; and the form of the multiple regression equation with the highest degree of matching is determined.

[0053] Specifically, when a user searches for a product, the product identification code is obtained, and the attribute information of the product is obtained based on the product identification code as a multi-dimensional variable. The multi-dimensional variable includes price attributes, cost attributes, sales attributes, and profit attributes.

[0054] For example, for clothing products, by scanning the clothing identification code to be queried (for example, a solid color style of a women's short-sleeved shirt), the price, cost, sales volume, profit and other attributes of the clothing can be obtained.

[0055] The sample data of similar products are obtained according to the prefix of the product identification code to form a sample set; the profit attribute of the product is received when the user queries it, and the target variable y is generated according to the profit attribute to determine the distribution of the sample attribute in the model.

[0056] According to the distribution of samples in multi-dimensional coordinates, determine the monotonicity, parity and other properties of the samples in each dimension, and select a suitable function model. For example, if the sample distribution is monotonically increasing, then the relationship between the target variable and the feature variable can be a univariate function. If it increases first and then decreases, then it can be a binary function.

[0057] For example, we can assume that the form of the multiple regression equation of the feature variable and the target variable is:

[0058] h(x)=θ1x1+θ3x3+θ3x3.

[0059] S103: Variables other than the target variable in the multidimensional variables are used as feature variables, and the feature variables are input into a multiple regression equation to obtain an output value for each sample.

[0060] Specifically, the feature variable x is generated for each attribute in the multidimensional variable except for the profit attribute. Based on the user-selected identification code prefix, all clothing categories within that prefix category are retrieved, such as other styles within the women's short-sleeved shirt category. The attribute information for each short-sleeved shirt is obtained, with the profit attribute serving as the target variable y. The other attributes of the clothing are used as feature variables: price as feature variable x1, cost as feature variable x2, and sales as feature variable x3. The feature variables are then entered into the function to obtain the predicted value for each sample.

[0061] S104: Determine the cost function of the multiple regression equation according to the difference between the output value and the actual value of the target variable.

[0062] In one embodiment of the present application, the difference h(x)-y between the output value after inputting the characteristic variable of each sample into the multiple regression equation and the actual value of the target variable is determined, and the square of the difference (h(x)-y) for each sample is determined. 2 ; Add the squares of the differences of each sample to get the sum of squares Divide the sum of squares by the sample size to get the mean of the sum of squares An expression for the mean of the sum of squares is determined, and a cost function J(θ) is determined; the cost function includes the parameter θ of the multiple regression equation.

[0063] S105: Determine the minimum value of the cost function, determine the parameters in the multiple regression equation according to the minimum value, and determine the expression of the multiple regression equation according to the parameters.

[0064] In one embodiment of the present application, the partial derivative of the cost function with respect to each parameter in the multiple regression equation is determined; each parameter in the multiple regression equation is updated according to the partial derivative; when the partial derivative of any parameter is determined to be 0, the parameter with a partial derivative of 0 is no longer updated; when the partial derivatives of all parameters in the cost function are 0, the value of each parameter is determined, and the parameter value is substituted into the multiple regression equation to obtain an expression of the multiple regression equation.

[0065] Specifically, by continuously updating the parameters of a variable through the gradient descent method until the partial derivative is equal to 0, the minimum value of the cost function in the dimension of the variable can be determined. Therefore, the value of the parameter when the partial derivative is equal to 0 is the optimal parameter value of the characteristic variable corresponding to the dimension. The optimal parameter value of each dimension is brought into the hypothetical multiple regression equation to determine the expression of the multiple regression equation.

[0066] S106: Obtain the predicted value of the output target variable according to the corresponding multiple regression equation, so that the user can manage the product.

[0067] Input the sample set into the model for training and determine the multiple regression equation for the query product

[0068] h(x)=θ1x1+θ3x3+θ3x3,

[0069] The predicted profit value of a product is derived from the multiple regression equation. Merchants can use this profit value to determine whether selling the product will satisfy them. They can also compare different product categories based on the profit value of the variable corresponding to the maximum value of the multiple regression equation to determine the product with the highest profit value.

[0070] In order to maximize profits, determine the cost, sales volume, price and other characteristic variables of the product when the profit value is the highest, and then determine the effective inventory of the product with the highest profit value (the inventory of the product that can be sold in the current season or the current holiday); calculate the saleable time (the number of days the product inventory can support sales) based on the effective inventory; when it is judged that the saleable time is less than the selling time (the time the product can be continuously sold, such as the time when New Year's goods can be sold before the New Year and during the New Year), it means that the product is not enough to sell, and the merchant determines a replenishment plan; when it is judged that the saleable time is greater than the selling time, it means that there will be surplus inventory, and the merchant determines the inventory transfer plan.

[0071] In one embodiment of the present application, the price attributes of the goods queried by the user are accepted to determine the predicted price of the goods; the cost of the goods is determined based on the purchase price, logistics, and storage of the goods; the profit of the goods is determined based on the cost of the goods and the predicted price of the goods; the supplier who can obtain the maximum profit value under the same conditions is determined, and the suppliers of goods are screened according to the preset profit value.

[0072] Specifically, if a merchant user wants to predict the local price of an item they sell, they can use the product identification code to query similar items as samples. After model training, they will generate a regression equation. Then, based on the product's attribute information, they will derive the predicted price of the item they want to query, which will serve as the selling price. Since each supplier's purchase price and logistics costs vary, merchants must minimize costs and maximize profits to save costs. Therefore, based on the product cost and predicted price, they can determine the profit value of purchasing from each supplier and identify the supplier with the highest profit value.

[0073] The above is a commodity management method based on identity resolution provided by the embodiment of the present application. Based on the same inventive idea, the embodiment of the present application also provides a corresponding commodity management device based on identity resolution, such as Figure 2 shown.

[0074] This embodiment provides a commodity management device based on identity resolution, including:

[0075] at least one processor; and,

[0076] a memory communicatively connected to at least one processor; wherein,

[0077] The memory stores instructions executable by at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0078] Collect identification codes of products of the same type, obtain attribute information of the products based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information;

[0079] Receive user selection to determine the target variable, determine the sample distribution based on the target variable and the number of multi-dimensional variables, and determine the form of the multiple regression equation based on the properties of the sample distribution;

[0080] The variables other than the target variable in the multidimensional variables are used as feature variables, and the feature variables are input into the multiple regression equation to obtain the output value of each sample;

[0081] Determine the cost function of the multiple regression equation based on the difference between the output value and the actual value of the target variable;

[0082] Determine the minimum value of the cost function, determine the parameters in the multiple regression equation based on the minimum value, and determine the expression of the multiple regression equation based on the parameters;

[0083] The predicted value of the output target variable is obtained according to the corresponding multiple regression equation, so that users can manage the products.

[0084] Based on the same idea, some embodiments of the present application also provide media corresponding to the above method.

[0085] Some embodiments of the present application provide a commodity management storage medium based on identity resolution, storing computer-executable instructions, wherein the computer-executable instructions are configured to:

[0086] Collect identification codes of products of the same type, obtain attribute information of the products based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information;

[0087] Receive user selection to determine the target variable, determine the sample distribution based on the target variable and the number of multi-dimensional variables, and determine the form of the multiple regression equation based on the properties of the sample distribution;

[0088] The variables other than the target variable in the multidimensional variables are used as feature variables, and the feature variables are input into the multiple regression equation to obtain the output value of each sample;

[0089] Determine the cost function of the multiple regression equation based on the difference between the output value and the actual value of the target variable;

[0090] Determine the minimum value of the cost function, determine the parameters in the multiple regression equation based on the minimum value, and determine the expression of the multiple regression equation based on the parameters;

[0091] The predicted value of the output target variable is obtained according to the corresponding multiple regression equation, so that users can manage the products.

[0092] The various embodiments in this application are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the method and medium embodiments are generally similar to the method embodiments, so their description is relatively simple. For relevant portions, refer to the description of the method embodiments.

[0093] The methods and media provided in the embodiments of the present application correspond one-to-one to the method, and therefore, the methods and media also have similar beneficial technical effects as the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the methods and media will not be repeated here.

[0094] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, commodity, or method comprising the element.

[0095] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A commodity management method based on identity resolution, characterized in that: Applied to commodity management systems, including: Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information; Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution; Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample; determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable; Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters; Obtaining the predicted value of the output target variable according to the corresponding multivariate regression equation to enable the user to manage the product; The receiving user selection to determine the target variable, determining the sample distribution according to the target variable and the number of the multi-dimensional variables, and determining the form of the multiple regression equation according to the properties of the sample distribution specifically includes: Receive user selection to determine target variable; Determine the domain and range of the sample distribution under the target variable; Determine the variation of the target variable of the sample distribution within the value range, and determine the properties of the sample distribution according to the variation; the properties of the sample distribution include monotonicity and parity in the definition domain; Matching the properties of the sample distribution with the properties of the multiple regression equation; Determine the form of the multiple regression equation that best fits; Determining the cost function of the multiple regression equation according to the difference between the output value and the actual value of the target variable specifically includes: Determine the difference between the output value of each sample after the characteristic variable is input into the multiple regression equation and the actual value of the target variable, and determine the square of the difference for each sample; Adding the squares of the difference values ​​of each sample to obtain a square sum; Divide the sum of squares by the sample size to obtain the mean of the sum of squares; Determine an expression for the mean of the sum of squares and determine a cost function; The cost function includes parameters of the multiple regression equation; When the user searches for a product, the method further includes: Obtaining a product identification code, and obtaining attribute information of the product based on the product identification code as a multi-dimensional variable, wherein the multi-dimensional variable includes price attribute, cost attribute, sales attribute, and profit attribute; obtaining sample data of similar products based on the prefix of the product identification code to form a sample set; Receive a user's query about the profit attribute of a product, generate a target variable y based on the profit attribute, and determine the distribution of the sample attribute in the model; Generate corresponding feature variables x for the attributes other than the profit attribute in the multi-dimensional variables; Inputting the sample set into the model for training, determining a multiple regression equation for the query product, and obtaining a predicted profit value based on the multiple regression equation; Comparing different categories of goods based on the profit values ​​to determine the category of goods with the highest profit value; Determine the effective inventory of the highest-profit merchandise; Calculate the available time based on the effective inventory; When it is determined that the available time is less than the selling time, a replenishment plan is determined; When it is determined that the saleable time is greater than the selling time, determining a call-out plan; Before collecting identification codes of commodities of the same type, the method further includes: Identify the scanned identification code information on the product; Receive basic information about the product input by the staff; Binding the identification code with the basic information of the product; Classifying the identification code according to preset classification rules; According to the classification result, the identification code and the basic information corresponding to the identification code are stored in a database.

2. The method according to claim 1, characterized in that Determining the minimum value of the cost function, determining the parameters in the multiple regression equation according to the minimum value, and determining the expression of the multiple regression equation according to the parameters specifically includes: determining a partial derivative of the cost function with respect to each parameter in a multiple regression equation; updating each parameter in the multiple regression equation according to the partial derivative; When it is determined that the partial derivative of any parameter is 0, the parameter whose partial derivative is 0 is no longer updated; When the partial derivatives of all parameters in the cost function are 0, each parameter value is determined, and the parameter value is substituted into the multiple regression equation to obtain an expression of the multiple regression equation.

3. The method according to claim 1, characterized in that The predicted value of the output target variable is obtained according to the corresponding multivariate regression equation to enable the user to manage the product, specifically including: Accepting the price attribute of the product queried by the user and determining a predicted price of the product; Determine the cost of the goods based on the purchase price, logistics, and storage of the goods; determining a profit for the commodity based on the cost of the commodity and the predicted price of the commodity; Filter suppliers of the goods according to the preset profit value.

4. The method according to claim 1, wherein After collecting product identification codes of the same type, the method further includes: Obtain a product identification code, and obtain a corresponding URL identifier based on the product identification code; Access the secondary node according to the URL identifier, determine the information server address recorded in the secondary node, access the information server address and establish communication with the information server to obtain relevant product information.

5. A commodity management device based on identity resolution, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information; Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution; Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample; determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable; Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters; Obtaining the predicted value of the output target variable according to the corresponding multivariate regression equation to enable the user to manage the product; The receiving user selection to determine a target variable, determining a sample distribution according to the target variable and the number of the multi-dimensional variables, and determining the form of a multiple regression equation according to the properties of the sample distribution specifically includes: Receive user selection to determine target variable; Determine the domain and range of the sample distribution under the target variable; Determine the variation of the target variable of the sample distribution within the value range, and determine the properties of the sample distribution according to the variation; the properties of the sample distribution include monotonicity and parity in the definition domain; Matching the properties of the sample distribution with the properties of the multiple regression equation; Determine the form of the multiple regression equation that best fits; Determining the cost function of the multiple regression equation according to the difference between the output value and the actual value of the target variable specifically includes: Determine the difference between the output value of each sample after the characteristic variable is input into the multiple regression equation and the actual value of the target variable, and determine the square of the difference for each sample; Adding the squares of the difference values ​​of each sample to obtain a square sum; Divide the sum of squares by the sample size to obtain the mean of the sum of squares; Determine an expression for the mean of the sum of squares and determine a cost function; The cost function includes parameters of the multiple regression equation; When a user searches for a product, Obtaining a product identification code, and obtaining attribute information of the product based on the product identification code as a multi-dimensional variable, wherein the multi-dimensional variable includes price attribute, cost attribute, sales attribute, and profit attribute; obtaining sample data of similar products based on the prefix of the product identification code to form a sample set; Receive a user's query about the profit attribute of a product, generate a target variable y based on the profit attribute, and determine the distribution of the sample attribute in the model; Generate corresponding feature variables x for the attributes other than the profit attribute in the multi-dimensional variables; Inputting the sample set into the model for training, determining a multiple regression equation for the query product, and obtaining a predicted profit value based on the multiple regression equation; Comparing different categories of goods based on the profit values ​​to determine the category of goods with the highest profit value; Determine the effective inventory of the highest-profit merchandise; Calculate the available time based on the effective inventory; When it is determined that the available time is less than the selling time, a replenishment plan is determined; When it is determined that the saleable time is greater than the selling time, determining a call-out plan; Before collecting the identification codes of the same type of products, Identify the scanned identification code information on the product; Receive basic information about the product input by the staff; Binding the identification code with the basic information of the product; Classifying the identification code according to preset classification rules; According to the classification result, the identification code and the basic information corresponding to the identification code are stored in a database.

6. A non-volatile storage medium storing computer-executable instructions, characterized in that: The computer executable instructions are configured to: Collect identification codes of commodities of the same type, obtain attribute information of the commodities based on the identification codes, obtain a sample set, and generate multi-dimensional variables based on the attribute information; Receive a user's selection to determine a target variable, determine a sample distribution based on the target variable and the number of multidimensional variables, and determine a form of a multiple regression equation based on properties of the sample distribution; Using the other variables in the multidimensional variables except the target variable as feature variables, and inputting the feature variables into the multiple regression equation to obtain the output value of each sample; determining a cost function of the multiple regression equation based on a difference between the output value and an actual value of the target variable; Determining a minimum value of the cost function, determining parameters in the multiple regression equation according to the minimum value, and determining an expression of the multiple regression equation according to the parameters; Obtaining the predicted value of the output target variable according to the corresponding multivariate regression equation to enable the user to manage the product; The receiving user selection to determine the target variable, determining the sample distribution according to the target variable and the number of the multi-dimensional variables, and determining the form of the multiple regression equation according to the properties of the sample distribution specifically includes: Receive user selection to determine target variable; Determine the domain and range of the sample distribution under the target variable; Determine the variation of the target variable of the sample distribution within the value range, and determine the properties of the sample distribution according to the variation; the properties of the sample distribution include monotonicity and parity in the definition domain; Matching the properties of the sample distribution with the properties of the multiple regression equation; Determine the form of the multiple regression equation that best fits; Determining the cost function of the multiple regression equation according to the difference between the output value and the actual value of the target variable specifically includes: Determine the difference between the output value of each sample after the characteristic variable is input into the multiple regression equation and the actual value of the target variable, and determine the square of the difference for each sample; Adding the squares of the difference values ​​of each sample to obtain a square sum; Divide the sum of squares by the sample size to obtain the mean of the sum of squares; Determine an expression for the mean of the sum of squares and determine a cost function; The cost function includes parameters of the multiple regression equation; Obtaining a product identification code, and obtaining attribute information of the product based on the product identification code as a multi-dimensional variable, wherein the multi-dimensional variable includes price attribute, cost attribute, sales attribute, and profit attribute; obtaining sample data of similar products based on the prefix of the product identification code to form a sample set; Receive a user's query about the profit attribute of a product, generate a target variable y based on the profit attribute, and determine the distribution of the sample attribute in the model; Generate corresponding feature variables x for the attributes other than the profit attribute in the multi-dimensional variables; Inputting the sample set into the model for training, determining a multiple regression equation for the query product, and obtaining a predicted profit value based on the multiple regression equation; Comparing different categories of goods based on the profit values ​​to determine the category of goods with the highest profit value; Determine the effective inventory of the highest-profit merchandise; Calculate the available time based on the effective inventory; When it is determined that the available time is less than the selling time, a replenishment plan is determined; When it is determined that the saleable time is greater than the selling time, determining a call-out plan; Identify the scanned identification code information on the product; Receive basic information about the product input by the staff; Binding the identification code with the basic information of the product; Classifying the identification code according to preset classification rules; According to the classification result, the identification code and the basic information corresponding to the identification code are stored in a database.

Images