Commodity replenishment data recommendation method and system, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a commodity replenishment data recommendation method and system, electronic equipment and a storage medium, belongs to the technical field of artificial intelligence, and is applied to a commodity sales scene. The method comprises the steps of firstly obtaining commodity information of candidate commodities, extracting commodity classification data from the commodity information according to preset commodity classification parameters, and then determining a commodity classification visualization graph according to the commodity classification parameters and the commodity classification data, and according to a preset commodity sales volume grade prediction model, performing sales volume grade prediction on the commodity classification visualization graph to obtain a target sales volume grade. And finally, performing replenishment prediction according to the candidate on-sale data, the target sales level and the commodity classification visualization graph in the commodity information to obtain replenishment data, and performing replenishment recommendation according to the replenishment data. Therefore, according to the embodiment of the invention, the accuracy of the replenishment data prediction can be improved, and the real-time performance of the replenishment data recommendation can be improved, so that merchants can grasp the sales trends of products in real time to make correct replenishment decisions.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology and is applied to the commodity sales scenario, and particularly relates to a method and system for recommending commodity replenishment data, an electronic device, and a storage medium. Background Art

[0002] In the commodity sales scenario, to ensure the rationality of inventory, merchants need to analyze the products they sell to predict in advance the products to be replenished and the replenishment quantity of the products. Currently, merchants usually manually count and evaluate the historical sales data of the products on sale at regular intervals to determine the product replenishment data such as the products to be replenished next time. However, the sales situation of products is affected by various factors such as market demand, marketing effect, competition of similar products, and potential users, resulting in low accuracy of manually counting and evaluating products and their replenishment quantities, and further leading to a series of inventory management problems such as product out-of-stock or overstock. Therefore, how to intelligently and accurately predict and timely recommend the replenishment data of products, so that merchants can grasp the sales dynamics of products in real time to make correct replenishment decisions, has become an urgent problem to be solved. Summary of the Invention

[0003] The main purpose of the embodiments of the present application is to propose a method and system for recommending commodity replenishment data, an electronic device, and a storage medium, aiming to improve the prediction accuracy of replenishment data and the real-time performance of replenishment data recommendation, so that merchants can grasp the sales dynamics of products in real time to make correct replenishment decisions.

[0004] To achieve the above purpose, in the first aspect of the embodiments of the present application, a method for recommending commodity replenishment data is proposed, and the method includes:

[0005] Obtain the commodity information of candidate commodities; wherein, each of the candidate commodities has candidate on-sale data;

[0006] Extract commodity classification data from the commodity information according to preset commodity classification parameters;

[0007] Visualize according to the commodity classification parameters and the commodity classification data to obtain a commodity classification visualization diagram;

[0008] Predict the sales volume level of the commodity classification visualization diagram according to a preset commodity sales volume level prediction model to obtain a target sales volume level;

[0009] Perform replenishment prediction according to the candidate on-sale data, the target sales volume level, and the commodity classification visualization diagram to obtain replenishment data;

[0010] Perform replenishment recommendation according to the replenishment data.

[0011] In some embodiments, visualizing according to the commodity classification parameter and the commodity classification data to obtain a commodity classification visualization diagram includes:

[0012] Search for an index value range from a preset quantization data table according to the commodity classification parameter;

[0013] Convert the commodity classification data into line segment visualization data according to the index value range;

[0014] Visualize according to the commodity classification parameter and the line segment visualization data to obtain the commodity classification visualization diagram.

[0015] In some embodiments, the line segment visualization data includes: the original length of the line segment and the multiple of the interval length. The index value range has a maximum value and a minimum value of the interval. Converting the commodity classification data into line segment visualization data according to the index value range includes:

[0016] Determine the interval length of the index value range according to the maximum value and the minimum value of the interval;

[0017] Determine the total data length according to the commodity classification data and the minimum value of the interval;

[0018] Determine the multiple of the interval length and the remaining data length according to the interval length and the total data length; wherein, the multiple of the interval length is the quotient of the total data length divided by the interval length, and the remaining data length is the remainder of the total data length divided by the interval length;

[0019] Normalize the remaining data length according to the interval length to obtain the original length of the line segment.

[0020] In some embodiments, visualizing according to the commodity classification parameter and the line segment visualization data to obtain the commodity classification visualization diagram includes:

[0021] Select a commodity classification visualization template according to the commodity classification parameter; wherein, the commodity classification visualization template includes: picture size, central black point, direction feature points, and the line segment direction of each commodity classification parameter. The direction feature points are used to determine the direction of the commodity classification visualization diagram;

[0022] Determine each index line segment according to the commodity classification parameter, the multiple of the interval length, the interval length, the line segment direction, the original length of the line segment, and the central black point; wherein, the central black point is the starting point of the index line segment;

[0023] Determine the boundary line according to the original length of the line segment;

[0024] Construct the commodity classification visualization graph according to the picture size, the direction feature points, the index line segments, and the boundary lines.

[0025] In some embodiments, determining each of the index line segments according to the commodity classification parameters, the interval length multiple, the interval length, the line segment direction, the original line segment length, and the central black dot includes:

[0026] Select a target line segment length from the interval length and the original line segment length according to the interval length multiple;

[0027] Determine an original line segment according to the commodity classification parameters, the target line segment length, the line segment direction, and the central black dot;

[0028] Mark a length multiple icon at the end of the original line segment according to the interval length multiple to obtain the index line segment.

[0029] In some embodiments, performing replenishment prediction according to the candidate on-sale data, the target sales level, and the commodity classification visualization graph to obtain replenishment data includes:

[0030] Select a target classification visualization graph from the commodity classification visualization graph according to the target sales level;

[0031] Determine a target commodity according to the target classification visualization graph; wherein, the target commodity has a target identifier;

[0032] Use the candidate on-sale data of the target commodity as the target on-sale data;

[0033] Perform replenishment prediction according to the target identifier, the target on-sale data, and the target sales level to obtain the replenishment data.

[0034] In some embodiments, before obtaining the commodity information of the candidate commodity, the method further includes:

[0035] Obtain the historical information of the training commodity from a preset commodity training set;

[0036] Extract historical index data from the historical information according to the commodity classification parameters;

[0037] Classify the historical information according to the commodity classification parameters and the historical index data to obtain commodity replenishment categories;

[0038] Determine the commodity sales data of the commodity replenishment category according to the total number of historical information in the commodity replenishment category;

[0039] Visualize according to the commodity classification parameters, the historical index data, and the commodity sales volume data to obtain an annotated classification visualization graph;

[0040] Pre-train the commodity sales volume level prediction model according to the annotated classification visualization graph.

[0041] To achieve the above object, a second aspect of the embodiments of the present application proposes a recommendation system for commodity replenishment data, and the system includes:

[0042] An information acquisition module, configured to acquire commodity information of candidate commodities; wherein, each of the candidate commodities has candidate on-sale data;

[0043] An index quantification module, configured to extract commodity classification data from the commodity information according to preset commodity classification parameters;

[0044] An image construction module, configured to perform visualization according to the commodity classification parameters and the commodity classification data to obtain a commodity classification visualization graph;

[0045] A sales volume prediction module, configured to predict the sales volume level of the commodity classification visualization graph according to a preset commodity sales volume level prediction model to obtain a target sales volume level;

[0046] A replenishment data prediction module, configured to perform replenishment prediction according to the candidate on-sale data, the target sales volume level, and the commodity classification visualization graph to obtain replenishment data;

[0047] A replenishment recommendation module, configured to perform replenishment recommendation according to the replenishment data.

[0048] To achieve the above object, a third aspect of the embodiments of the present application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect above is implemented.

[0049] To achieve the above object, a fourth aspect of the embodiments of the present application proposes a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect above is implemented.

[0050] The recommended method and system for merchandise replenishment data, electronic device, and storage medium proposed in this application first obtain the merchandise information of candidate merchandise; each candidate merchandise has candidate on-sale data. Then, merchandise classification data is extracted from the merchandise information according to preset merchandise classification parameters, and visualization is performed based on the merchandise classification parameters and the merchandise classification data to obtain a merchandise classification visualization diagram, which can quantify and visualize the factors affecting the sales of candidate merchandise, and intuitively reflect the sales situation and potential demand of candidate merchandise in the market. Thereafter, a sales volume level prediction is performed on the merchandise classification visualization diagram according to a preset merchandise sales volume level prediction model to obtain a target sales volume level, and replenishment prediction is performed based on the candidate on-sale data, the target sales volume level, and the merchandise classification visualization diagram to obtain replenishment data. Finally, replenishment recommendations are made based on the replenishment data. Therefore, the recommended method and system, electronic device, and storage medium illustrated in the embodiments of this application can, after visualizing the merchandise classification parameters and the merchandise classification data, perform a sales volume level prediction on the merchandise classification visualization diagram through the merchandise sales volume level prediction model to obtain a target sales volume level, and then determine the replenishment data based on the target sales volume level, the candidate on-sale data, and the merchandise classification visualization diagram, which can improve the accuracy of replenishment data prediction; moreover, the replenishment recommendation plan recommended to merchants based on the replenishment data can improve the real-time nature of replenishment data recommendation, enabling merchants to grasp the sales dynamics of products in real time for correct replenishment decisions, and further enabling merchants to replenish inventory in a timely manner to avoid economic losses caused by out-of-stock or overstocking. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is an alternative flowchart of the recommended method for merchandise replenishment data provided by an embodiment of this application;

[0052] Figure 2 is Figure 1 a flowchart of step S103 in

[0053] Figure 3 is Figure 2 a flowchart of step S202 in

[0054] Figure 4 is Figure 2 a flowchart of step S203 in

[0055] Figure 5 is a specific implementation schematic diagram of the merchandise classification visualization template provided by an embodiment of this application;

[0056] Figure 6 is Figure 4 a flowchart of step S402 in

[0057] Figure 7 is another alternative specific implementation schematic diagram of the merchandise classification visualization diagram provided by an embodiment of this application;

[0058] Figure 8 It is another optional specific implementation schematic diagram of the commodity classification visualization diagram provided by the embodiments of the present application;

[0059] Figure 9 is Figure 1 the flowchart of step S105 in

[0060] Figure 10 It is another optional flowchart of the method for recommending commodity replenishment data provided by the embodiments of the present application;

[0061] Figure 11 It is another optional specific implementation schematic diagram of the commodity classification visualization diagram provided by the embodiments of the present application;

[0062] Figure 12 It is the specific implementation schematic diagram of the annotation classification visualization diagram provided by the embodiments of the present application;

[0063] Figure 13 It is the structural schematic diagram of the recommendation system for commodity replenishment data provided by the embodiments of the present application;

[0064] Figure 14 It is the hardware structural schematic diagram of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0065] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0066] It should be noted that although the functional modules are divided in the system schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the system or the order in the flowchart. The terms "first", "second", etc. in the description and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and do not have to be used to describe a specific order or sequence.

[0067] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0068] First, several nouns involved in the present application are analyzed:

[0069] Artificial Intelligence (AI): It is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence; artificial intelligence is a branch of computer science. Artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a way similar to human intelligence. The research in this field includes robots, speech recognition, image recognition, natural language processing, and expert systems, etc. Artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence also refers to the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0070] Image Recognition (IR): It is a technology for understanding and interpreting the content of images applied to computers, capable of recognizing various contents such as scenes, animals, human expressions, and text. For example, when a computer obtains a picture containing an animal, image recognition technology can analyze the pixel information in the image through complex algorithms and models, and then identify whether the animal in the picture is a cat, a dog, or other animals. In practical applications, image recognition is widely used in the analysis of data statistical charts to help users better understand the distribution and characteristics of image data. Image recognition can also provide references for the training and optimization of data analysis models. For example, in the field of autonomous driving, image recognition technology can analyze statistical charts of road scenes, such as vehicle density maps and pedestrian flow maps, and can adjust the driving strategy of the vehicle in real time to ensure driving safety. In agricultural production, image recognition technology can analyze the distribution map of crop pests and diseases to help farmers take timely control measures.

[0071] In the process of e-commerce operation, commodity information such as the basic attributes of commodities, market demand conditions, and marketing effects will all affect the sales volume of commodities. For example, basic attributes such as the color and size of commodities will directly affect consumers' purchase choices, while market demand conditions such as the usage season of commodities will affect the sales cycle of commodities. In addition, the number of times of commodity advertising will affect the exposure times of commodities on e-commerce platforms. If the sales volume of commodities is inaccurately evaluated, it may lead to a mismatch between the actual sales situation and the expected sales situation after the merchant replenishes the goods, thus causing problems such as overstocking or out-of-stock.

[0072] Due to the huge amount of information contained in product information, it is difficult for traditional product replenishment data recommendation methods to accurately analyze the relationship between massive product information and product sales volume. Therefore, the current prediction of product replenishment volume mainly relies on the experience of operators, the comparison of the advantages of similar products, and the comparison of sales volume. This not only requires operators to have rich product analysis experience but also requires them to invest a large amount of time cost in analyzing the product information of similar products. That is, the existing methods for predicting product replenishment volume are inefficient and difficult to provide effective replenishment decision support for merchants. Therefore, how to improve the prediction accuracy of replenishment data and the real-time nature of recommendations, so that merchants can grasp the sales dynamics of products in real time to make correct replenishment decisions, has become an urgent problem to be solved.

[0073] Based on this, the embodiments of this application provide a method and system for recommending product replenishment data, an electronic device, and a storage medium, aiming to improve the prediction accuracy of replenishment data and the real-time nature of replenishment data recommendations, so that merchants can grasp the sales dynamics of products in real time to make correct replenishment decisions.

[0074] The method and system for recommending product replenishment data, the electronic device, and the storage medium provided by the embodiments of this application are specifically described through the following embodiments. First, the method for recommending product replenishment data in the embodiments of this application is described.

[0075] The embodiments of this application can obtain and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0076] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0077] The recommended method for merchandise replenishment data provided by the embodiments of this application relates to the field of artificial intelligence technology and is applied to the merchandise sales scenario. The recommended method for merchandise replenishment data provided by the embodiments of this application can be applied to a terminal, or to a server side, or can also be software running on a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, or can also be configured as a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the recommended method for merchandise replenishment data, etc., but is not limited to the above forms.

[0078] This application can be used in many general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This application can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0079] Figure 1 is an optional flowchart of the recommended method for merchandise replenishment data provided by the embodiments of this application, Figure 1 The method in [the figure] may include, but is not limited to, steps S101 to S106.

[0080] Step S101, obtain the merchandise information of candidate merchandise; wherein, each candidate merchandise has candidate on-sale data;

[0081] Step S102, extract merchandise classification data from the merchandise information according to preset merchandise classification parameters;

[0082] Step S103, perform visualization according to the merchandise classification parameters and the merchandise classification data to obtain a merchandise classification visualization graph;

[0083] Step S104: Predict the sales volume level of the visualized product classification graph according to the preset product sales volume level prediction model to obtain the target sales volume level;

[0084] Step S105: Conduct replenishment prediction based on the candidate on-sale data, the target sales volume level, and the visualized product classification graph to obtain replenishment data;

[0085] Step S106: Make replenishment recommendations based on the replenishment data.

[0086] Steps S101 to S106 shown in the embodiments of the present application first obtain the product information of candidate products; among them, each candidate product has candidate on-sale data. Then, extract the product classification data from the product information according to the preset product classification parameters, and perform visualization based on the product classification parameters and the product classification data to obtain the visualized product classification graph, which can quantify and visualize the factors affecting the sales of candidate products, and intuitively reflect the sales situation and potential demand of candidate products in the market. Thereafter, predict the sales volume level of the visualized product classification graph according to the preset product sales volume level prediction model to obtain the target sales volume level, and conduct replenishment prediction based on the candidate on-sale data, the target sales volume level, and the visualized product classification graph to obtain replenishment data. Finally, make replenishment recommendations based on the replenishment data. Therefore, the recommendation method shown in the embodiments of the present application can, after visualizing the product classification parameters and the product classification data, predict the sales volume level of the visualized product classification graph through the product sales volume level prediction model to obtain the target sales volume level, and then determine the replenishment data according to the target sales volume level, the candidate on-sale data, and the visualized product classification graph, which can improve the accuracy of replenishment data prediction; moreover, the replenishment recommendation plan recommended to merchants based on the replenishment data can improve the timeliness of replenishment data recommendation, so that merchants can timely grasp the sales dynamics of products to make correct replenishment decisions, and then enable merchants to replenish inventory in time, avoiding economic losses caused by out-of-stock or overstock.

[0087] In step S101 of some embodiments, the candidate product refers to the product that the target object plans to sell, which can be a product on sale, a historically sold product, or a product not yet on the market, and can be a physical product or a virtual product. Among them, the target object refers to a merchant, who is the core entity of commercial activities, provides products and services to the market, and can be an individual enterprise, a professional retailer, a general retailer, a chain enterprise, or a multinational company. The merchant can sell products offline or online and engage in real economy activities or virtual economy activities. Product information refers to various detailed data and descriptions related to the candidate product, which can reflect the market competitiveness and potential sales trends of the candidate product. Product information can be directly stored in the e-commerce platform or local database in the form of data and crawled from the e-commerce platform or called from the local database through JAVA technology when product information needs to be obtained, or it can be recorded or saved in the form of paper documents and input into the computer in the form of pictures or manual input when product information needs to be obtained. On this basis, through the comprehensive collection of product information, basic data support can be provided for subsequent product replenishment recommendations, ensuring the accuracy and effectiveness of product replenishment recommendations, thereby helping merchants achieve accurate inventory management and efficient sales operations.

[0088] In some embodiments, the form of product information can be one or several of transaction records, data tables, and product description texts, or other forms. Product information can be the basic attributes of products such as brand, model, specification, and color, the on-sale data such as inventory quantity, sales volume, sales frequency, and inventory turnover rate, the promotion data such as the number of advertisement placements, the user behavior analysis data such as the number of previews, favorite data, purchase times, and repurchase times, the time data such as season, year, month, and date, the product use classification data, customer group classification data, and other data, or any combination of the above data. On this basis, the candidate on-sale data refers to the on-sale data of the candidate product, which can be used to predict replenishment data such as the time point for replenishment recommendations for the merchant.

[0089] In step S102 of some embodiments, the preset product classification parameters refer to the parameters for classifying candidate products, which are used to extract product classification data from the product information of candidate products, and then classify the candidate products according to the product classification parameters and separation index data to obtain product replenishment categories, enabling merchants to clearly understand the attributes and characteristics of each product, and thus more effectively carry out inventory management, sales strategy formulation, market positioning and other tasks. For example, clothing sellers classify clothing according to the applicable seasons, materials, colors, sizes and user groups of the clothing. In the scenario of electronic product sales, the product classification parameters may include product use classification, brand, model, function and selling price, etc. Product classification data refers to the specific data and information on which the product classification parameters are based when classifying products, and can accurately reflect the characteristics and attributes of the products. For example, in a comprehensive e-commerce platform, for products in the category of books, the product classification data may include: the author of the book is "Zhang San", the publisher is "ABC University Press", the publication time is "December 2015", the ISBN number is "978-7-111-56789-0", the number of pages of the book is "135 pages", the binding type is "paperback", and the literary category to which it belongs is "novel", etc. On this basis, product classification data can be extracted from the product information according to the preset product classification parameters to ensure that the extracted product classification data is accurate and targeted, and then the candidate products can be accurately classified. Specifically, when the product information is stored in the form of a data table, the product classification data can be directly retrieved according to the product classification parameters; when the product information is text, the product classification data of the product classification parameters can be extracted from the product information through natural language processing technology; when the product information is a picture, first extract the text information of the product from the picture through image recognition technology, and then extract the product classification data of the product classification parameters from the text information.

[0090] In step S103 of some embodiments, the product classification visualization diagram refers to an image obtained by visualizing the product replenishment categories. The product classification visualization diagram can be in the form of a radar chart, bar chart, pie chart or line chart, etc. Since the image recognition algorithm based on artificial intelligence is more sensitive to images than to data, by visualizing according to the product classification parameters and product classification data to obtain the product classification visualization diagram, the complex product classification parameters and product classification data can be intuitively displayed in the form of an image, enabling the image recognition algorithm based on artificial intelligence to more accurately analyze the potential relationship between the product classification parameters, product classification data and the candidate sales level of the product.

[0091] Please refer to Figure 2 , in some embodiments, step S103 may include but is not limited to steps S201 to S203:

[0092] Step S201: Find the index value range from the preset quantization data table according to the product classification parameters;

[0093] Step S202: Convert the product classification data into line segment visualization data according to the index value range;

[0094] Step S203: Perform visualization according to the product classification parameters and the line segment visualization data to obtain a product classification visualization graph.

[0095] In step S201 of some embodiments, the preset quantization data table is used to quantify the product classification data to obtain classified quantization data. For example, if the specific product classification data is the product sales time "20110708", then for the product classification parameter "season", the "season" corresponding to "20110708" can be found from the quantization data table as "the 2nd season", and "2" is used as the classified quantization data; for the product classification parameter "month", the "month" corresponding to "20110708" can be found from the quantization data table as "7", and "7" is used as the classified quantization data. If the specific product classification data is that the user group is "female at 20 years old", then for the product classification parameter "user group", the "user group" corresponding to "20 years old" and "female" can be found from the quantization data table as "-20", and "-20" is used as the classified quantization data; if the specific product classification data is that the user group is "male at 40 years old", then for the product classification parameter "user group", the "user group" corresponding to "40 years old" and "male" can be found from the quantization data table as "40", and "40" is used as the classified quantization data. For each product classification parameter, there is an index value range in the quantization data table, and the index value range refers to the value range of the classified quantization data. For example, if the product classification data is "color", then the index value range can be [0, 9], where each value in the index value range represents the following meanings: 0 = "black", 1 = "red", 2 = "orange", 3 = "yellow", 4 = "green", 5 = "blue", 6 = "indigo", 7 = "violet", 8 = "white", 9 = "transparent". For the product classification parameter "user group", the index value range can be [-100, 100], where [-100, -1] respectively correspond to female users aged 1 to 100 years old, and [1, 100] respectively correspond to male users aged 1 to 100 years old.

[0096] It should be noted that if the product classification data is the number of ad placements, the number of product collections, the number of previews, etc., the product classification data can be directly used as the classified quantization data.

[0097] In step S202 of some embodiments, the index value range has a maximum value and a minimum value of the range. Therefore, the classified quantization data can be converted into line segment visualization data according to the numerical relationship between the maximum value, the minimum value of the index value range, and the commodity classification data. The line segment visualization data is used to generate line segments corresponding to the commodity classification parameters and the commodity classification data in the commodity classification visualization graph. Specifically, for each classified quantization data in the index value range, it can be mapped to a line segment in the commodity classification visualization graph, and the length of the line segment represents the commodity classification parameters and the commodity classification data corresponding to the classified quantization data.

[0098] Steps S201 to S203 illustrated in the embodiments of the present application first find the index value range from the preset quantization data table according to the commodity classification parameters, and convert the commodity classification data into line segment visualization data according to the index value range. Then, visualization is performed according to the commodity classification parameters and the line segment visualization data to obtain the commodity classification visualization graph. Therefore, the recommended method for commodity replenishment data shown in the embodiments of the present application can visually display the commodity classification parameters and the commodity classification data by performing data visualization on the commodity classification parameters according to the commodity classification parameters and the preset quantization data table, facilitating the commodity sales volume level prediction model to understand and analyze the commodity classification parameters and the commodity classification data for accurate sales volume level prediction, reducing the difficulty of commodity sales volume level prediction, and improving the accuracy of commodity sales volume level prediction.

[0099] Please refer to Figure 3 , in some embodiments, the line segment visualization data may include but is not limited to: the original length of the line segment and the multiple of the interval length. The original length of the line segment is used to describe the relative position or size of the commodity classification data within the index value range, and the multiple of the interval length is used to describe the relative position or size of the commodity classification data exceeding the interval; step S202 may include but is not limited to steps S301 to S304:

[0100] Step S301, determine the interval length of the index value range according to the maximum value and the minimum value of the range;

[0101] Step S302, determine the total data length according to the commodity classification data and the minimum value of the interval;

[0102] Step S303, determine the multiple of the interval length and the remaining data length according to the interval length and the total data length; where the multiple of the interval length is the quotient of the total data length divided by the interval length, and the remaining data length is the remainder of the total data length divided by the interval length;

[0103] Step S304, normalize the remaining data length according to the interval length to obtain the original length of the line segment.

[0104] In step S301 of some embodiments, the interval length of the index value range is determined according to the maximum value and the minimum value of the interval. The specific way to determine the interval length of the index value range can be: subtracting the minimum value of the interval from the maximum value of the interval. For example, if the index value range is used to describe the price of a commodity, the maximum value of the interval can be 1000, indicating that the highest price of the commodity is one thousand yuan, and the minimum value of the interval can be 100, indicating that the lowest price of the commodity is one hundred yuan, then the interval length can be determined as 900 by subtracting the minimum value of the interval from the maximum value of the interval.

[0105] In step S302 of some embodiments, the total data length is used to describe the relationship between the commodity classification data and the index value range. After determining the classified quantization data according to the commodity classification data, the total data length can be determined according to the classified quantization data and the minimum value of the interval. Specifically, the total data length is obtained by subtracting the minimum value of the interval from the classified quantization data. For example, if the index value range is used to describe the number of advertisements placed for a commodity, the minimum value of the interval can be 0, indicating that no advertisements are placed for the commodity, the maximum value of the interval can be 1000, indicating that the number of advertisements placed for the commodity is one thousand, and the classified quantization data is 500, indicating that the actual number of advertisements placed for the commodity is five hundred, then the total data length can be determined as 500 by subtracting the minimum value of the interval from the classified quantization data.

[0106] In step S303 of some embodiments, during the actual application process of the recommended method for commodity replenishment data, after determining the classified quantization data according to the commodity classification data, the classified quantization data may exceed the maximum value or the minimum value of the index value range, that is, the classified quantization data cannot be directly found within the index value range. At this time, it is necessary to first determine the total data length according to the classified quantization data and the minimum value of the interval, and then determine the interval length multiple and the data remainder length according to the total data length and the interval length. Specifically, first divide the total data length by the interval length, then use the obtained quotient as the interval length multiple, and use the obtained remainder as the data remainder length. For example, if the total data length is 500 and the interval length is 900, then the interval length multiple is 0 and the data remainder length is 500. If the total data length is 1200 and the interval length is 900, then the interval length multiple is 1 and the data remainder length is 300.

[0107] In step S304 of some embodiments, the data margin length may be normalized according to the interval length to obtain the original line segment length. Normalization is a method of scaling data proportionally within a preset standard interval, and the standard interval is usually [0,1]. Normalizing the data margin length can avoid the drawn line segment exceeding the range of the product classification visualization graph during the process of visualizing according to the product classification parameters and the line segment visualization data. For example, if the data margin length is 50 and the interval length is 90, then the original line segment length may be 50 divided by 90, which is approximately 0.56.

[0108] Steps S301 to S304 illustrated in the embodiments of the present application first determine the interval length of the index value range according to the interval maximum value and the interval minimum value. Then, the total data length is determined according to the product classification data and the interval minimum value, and the interval length multiple and the data margin length are determined according to the interval length and the total data length; where the interval length multiple is the quotient of the total data length divided by the interval length, and the data margin length is the remainder of the total data length divided by the interval length. Finally, the data margin length is normalized according to the interval length to obtain the original line segment length. Therefore, the recommended method for product replenishment data illustrated in the embodiments of the present application is convenient for visualizing the product classification parameters and the line segment visualization data by converting the product classification data into line segment visualization data. Moreover, by splitting the total data length into an interval length multiple and a data margin length, it is possible to avoid modifying the range of the index value range when the total data length exceeds the range of the index value range, thereby improving the accuracy of the sales level prediction model for predicting the sales level. Normalizing the data margin length can standardize the line segment length and avoid the drawn line segment exceeding the image range according to the total data length, thereby improving the accuracy of the product sales level prediction model for analyzing the product classification visualization graph.

[0109] Please refer to Figure 4 , in some embodiments, step S203 may include but is not limited to steps S401 to S404:

[0110] Step S401, select a product classification visualization template according to the product classification parameters; where the product classification visualization template includes: picture size, central black point, direction feature points, and the line segment direction of each product classification parameter, and the direction feature points are used to determine the direction of the product classification visualization graph;

[0111] Step S402, determine each index line segment according to the product classification parameters, interval length multiple, interval length, line segment direction, original line segment length, and central black point; where the central black point is the starting point of the index line segment;

[0112] Step S403, determine the boundary line according to the original line segment length;

[0113] Step S404: Construct a visualization graph for product classification based on the picture size, direction feature points, index line segments, and boundary lines.

[0114] In step S401 of some embodiments, a visualization template for product classification can be selected according to the product classification parameters. Among them, the visualization template for product classification is the basic framework for constructing the visualization graph, which defines the overall layout of the graph and is used to construct the visualization graph for product classification. Specifically, the visualization template for product classification includes, but is not limited to: picture size, central black point, direction feature points, and the line segment directions of each product classification parameter. Among them, the picture size is used to set the size of the visualization graph for product classification; the central black point is the common starting point of all index line segments; the direction feature points are used to determine the direction of the visualization graph for product classification to prevent inputting a visualization graph for product classification with the wrong direction into the product sales level prediction model, thereby causing the product sales level prediction model to confuse the index line segments corresponding to each product classification parameter; the line segment directions corresponding to each product classification parameter are used to determine the extension direction of each index line segment in the graph.

[0115] Please refer to Figure 5 , for example, if the product classification parameters are: product use, customer group, season, selling price, advertising investment, and preview times, then the visualization template for product classification shown in Figure 5 can be selected. Among them, the picture size is "Resolution: 608*608", the central black point is located at the center of the visualization template for product classification, the direction feature point is set at the upper right corner of the visualization template for product classification, and the line segment directions of the index line segments can be determined respectively according to the product classification parameters of product use, customer group, season, selling price, advertising investment, and preview times from the central black point.

[0116] In step S402 of some embodiments, each index line segment can be determined according to the product classification parameters, interval length multiple, interval length, line segment direction, line segment original length, and central black point. Among them, the index line segments are used to visualize the product classification parameters and product classification data on the visualization graph for product classification.

[0117] In step S403 of some embodiments, the boundary lines can be determined according to the line segment original length. The boundary lines are used to connect each index line segment and intuitively display the overall distribution and characteristics of the product classification data. Specifically, first, the boundary points of each index line segment are determined according to the line segment original length, and then the boundary points of all index line segments are connected to obtain the boundary lines. Among them, the boundary points are used to describe the position corresponding to the line segment original length on the index line segment. It should be noted that in order to enhance the characteristics of the overall distribution and characteristics of the product classification data, the boundary lines can be prominently set. For example: set the color of the boundary lines to red lines and thicken the boundary lines.

[0118] In step S404 of some embodiments, a commodity classification visualization graph can be constructed based on the picture size, direction feature points, index line segments, and boundary lines. Specifically, first, an original visualization image is generated according to the picture size, then the index line segments and boundary lines are filled on the original visualization image, and the direction feature points are marked on the original visualization image to obtain the commodity classification visualization graph.

[0119] In steps S401 to S404 illustrated in the embodiments of the present application, first, a commodity classification visualization template is selected according to the commodity classification parameters; wherein, the commodity classification visualization template includes: picture size, central black point, direction feature points, and the line segment directions of each commodity classification parameter, and the direction feature points are used to determine the direction of the commodity classification visualization graph. Then, each index line segment is determined according to the commodity classification parameters, interval length multiple, interval length, line segment direction, original line segment length, and central black point; wherein, the central black point is the starting point of the index line segment. Finally, the boundary line is determined according to the original line segment length, and the commodity classification visualization graph is constructed based on the picture size, direction feature points, index line segments, and boundary lines. Therefore, in the method for recommending commodity replenishment data illustrated in the embodiments of the present application, by selecting a commodity classification visualization template according to the commodity classification parameters, determining each index line segment according to the commodity classification parameters, interval length multiple, interval length, line segment direction, original line segment length, and central black point, and then determining the boundary line through the original line segment length, so as to fill the index line segments and boundary lines on the commodity classification visualization template, a commodity classification visualization graph that can simply and clearly reflect the commodity classification parameters and commodity classification data is obtained, enabling the commodity sales volume level prediction model to accurately identify the index line segments and boundary lines in the commodity classification visualization graph for commodity sales volume level prediction.

[0120] Please refer to Figure 6 , in some embodiments, step S402 may further include but is not limited to steps S601 to S603:

[0121] Step S601, select the target line segment length from the interval length and the original line segment length according to the interval length multiple;

[0122] Step S602, determine the original line segment according to the commodity classification parameters, target line segment length, line segment direction, and central black point;

[0123] Step S603, mark a length multiple icon at the end of the original line segment according to the interval length multiple to obtain the index line segment.

[0124] In step S601 of some embodiments, the target line segment length can be selected from the interval length and the original line segment length according to the interval length multiple. The target line segment length refers to the line segment length for drawing the index line segment. Specifically, when the interval length multiple is not zero, it indicates that the total data length exceeds the interval length, and the interval length is determined as the target line segment length; when the interval length multiple is zero, it indicates that the total data length does not exceed the interval length, and the original line segment length is determined as the target line segment length. Therefore, the index line segment drawn based on the target line segment length can directly reflect the relationship between the classified quantitative data and the index interval.

[0125] In step S602 of some embodiments, after determining the target line segment length, the original line segment can be determined according to the commodity classification parameter, the target line segment length, the line segment direction, and the central black dot. Specifically, with the central black dot as the origin, the original line segment is drawn according to the target line segment length in the line segment direction of the commodity classification parameter.

[0126] In step S603 of some embodiments, by marking the length multiple icon at the end of the original line segment according to the interval length multiple, the index line segment can be obtained. The length multiple icon is used to visually display the multiple relationship between the classified quantitative data and the interval length. Specifically, when the interval length multiple is zero, no length multiple icon is marked at the end of the original line segment; when the interval length multiple is not zero, the length multiple icon is marked at the end of the original line segment according to the interval length multiple. For example, the length multiple icon can be a circular icon. Refer to Figure 7 If the interval length multiple is 1, one length multiple icon is marked at the end of the original line segment; refer to Figure 8 If the interval length multiple is 2, two length multiple icons are marked at the end of the original line segment. In addition, the length multiple icon can also be square, triangular or other shapes, and the length multiple icon can also be represented by "xN", where N represents the specific interval length multiple, such as "x1", "x2", and "x3".

[0127] Steps S601 to S603 illustrated in the embodiments of the present application first select the target line segment length from the interval length and the original line segment length according to the interval length multiple, determine the original line segment according to the commodity classification parameter, the target line segment length, the line segment direction, and the central black dot, and then mark the length multiple icon at the end of the original line segment according to the interval length multiple to obtain the index line segment, which can enable the index line segment to accurately reflect the commodity classification data of the commodity classification parameter and facilitate visually understanding the relationship between the commodity classification data and the index interval from the commodity classification visualization diagram.

[0128] In step S104 of some embodiments, the preset commodity sales volume level prediction model is an image recognition model, which is used to predict the sales volume level of candidate commodities according to the commodity classification visualization diagram, and obtain the target sales volume level of the candidate commodities. The commodity sales volume level prediction model can be constructed based on YOLO-V4, CNNs, SVMs, k-NN, decision trees or GANs. The target sales volume level refers to the level at which the total sales volume of the candidate commodity in a specific future time period is located, reflecting the market prospect of the candidate commodity. An accurate target sales volume level can help merchants plan the inventory of the candidate commodity in advance and adjust the marketing strategy of the candidate commodity. The target sales volume level can be any one of the following: hot sale, best-selling, average, unpopular or hard to sell, or can be set according to the actual needs of those skilled in the art. Among them, in the process of predicting the target sales volume level, the specific value of the future time period can be a quarter, a month or a year, or can be set according to the actual needs of those skilled in the art or merchants. The specific value of the future time period is not specifically limited in this application.

[0129] Taking YOLO-V4 as an example, YOLO-V4 is an image recognition algorithm based on CNNs. It performs object recognition on the input image through a single neural network and outputs the category and confidence value of the object. The commodity sales volume level prediction model based on YOLO-V4 first predicts the sales volume level of the input commodity classification visualization diagram, and obtains the confidence value that the commodity classification visualization diagram is classified into each candidate sales volume level. Then, the target sales volume level is selected from the candidate sales volume levels according to the confidence value. For example, if the confidence value that the candidate sales volume level of the commodity classification visualization diagram is hot sale is 10%, the confidence value that the candidate sales volume level is best-selling is 85%, the confidence value that the candidate sales volume level is average is 92%, the confidence value that the candidate sales volume level is unpopular is 47.64%, and the confidence value that the candidate sales volume level is hard to sell is 5%, then the candidate sales volume level "average" with the highest confidence value is selected as the target sales volume level.

[0130] In step S105 of some embodiments, the replenishment data refers to the data required for replenishing goods, which may include, but is not limited to, at least one of the following data: product identification, replenishment time, replenishment quantity, replenishment priority level, recommended time point for replenishment, or recommended replenishment method. The replenishment data can also be set to other specific data according to the needs of those skilled in the art. Among them, the product identification is used to represent the product replenishment category to which the product belongs, and each product replenishment category has a unique product identification. The replenishment time is used to describe the time when the merchant replenishes the goods. The replenishment time can be a time point or a time period. The replenishment priority level is used to describe the replenishment priority degree of the goods, which may include: level one, level two, or other levels. Level one indicates that replenishment of this product is most recommended, level two indicates that replenishment of this product is relatively recommended, and other levels follow this analogy. The specific number of levels can be set according to the actual needs of those skilled in the art. The recommended time point for replenishment is used to describe the recommended replenishment information sent to the merchant at a specific time point. The recommended replenishment method is used to describe the method of sending replenishment recommendation information to the merchant, which can be a text message, an email, or a voice message, or can also be set to other specific recommended methods according to the needs of those skilled in the art. The specific replenishment prediction method can be implemented by one or more of time series analysis, regression analysis, ABC analysis method, and replenishment prediction model.

[0131] Please refer to Figure 9 , in some embodiments, step S105 includes, but is not limited to, steps S901 to S904:

[0132] Step S901, select the target classification visualization graph from the product classification visualization graph according to the target sales volume level;

[0133] Step S902, determine the target product according to the target classification visualization graph; among them, the target product has a target identification;

[0134] Step S903, use the candidate on-sale data of the target product as the target on-sale data;

[0135] Step S904, perform replenishment prediction based on the target identification, target on-sale data, and target sales volume level to obtain the replenishment data.

[0136] In step S901 of some embodiments, when there are multiple candidate products, it is necessary to select products with high sales potential from the candidate products according to the target sales volume level of each candidate product. Since each candidate product has a product classification visualization diagram, the candidate products can be determined based on the product classification visualization diagram. On this basis, the target classification visualization diagram is first selected from the product classification visualization diagram according to the target sales volume level. Specifically, first compare the preset recommended sales volume level with the target sales volume level. If the recommended sales volume level matches the target sales volume level, the product classification visualization diagram corresponding to the target sales volume level is selected as the target classification visualization diagram. Among them, the recommended sales volume level can be at least one of the following: hot sales, best-selling, average, unpopular, or hard to sell. For example, the recommended sales volume level can be set to "hot sales", or the recommended sales volume level can be set to "hot sales and best-selling".

[0137] In step S902 of some embodiments, after determining the target classification visualization diagram, the target product can be determined according to the target classification visualization diagram; among them, the target product has a target identifier, and the target identifier refers to the product identifier of the target product. Specifically, when generating the product classification visualization diagram according to the product information of the candidate product, visualization diagram identification labels will be assigned to the product information and the product classification visualization diagram respectively. Therefore, after determining the target classification visualization diagram, the visualization diagram identification label of the target classification visualization diagram can be extracted, and then the target product can be determined from the candidate products according to the visualization diagram identification label.

[0138] In step S903 of some embodiments, the target on-sale data refers to the on-sale data of the target product and can be used to predict replenishment data such as the time point for recommending replenishment to the merchant. The target on-sale data may include: inventory quantity, sales amount, sales frequency, and inventory turnover rate.

[0139] In step S904 of some embodiments, after using the candidate on-sale data of the target product as the target on-sale data, replenishment prediction can be performed according to the target identifier, the target on-sale data, and the target sales volume level to obtain replenishment data. Specifically, first determine the inventory clearance duration according to the target on-sale data and the target sales volume level, then determine the replenishment time and replenishment quantity according to the inventory clearance duration, and determine the replenishment priority level according to the target sales volume level, replenishment time, and replenishment quantity. Finally, determine the recommended replenishment time point and the recommended replenishment method according to the replenishment time and replenishment quantity, and finally use the target identifier as the product identifier in the replenishment data.

[0140] Steps S901 to S904 illustrated in the embodiments of the present application first select a target classification visualization graph from the product classification visualization graph according to the target sales volume level, and determine the target product according to the target classification visualization graph; wherein, the target product has a target identifier. Then, use the candidate on-sale data of the target product as the target on-sale data. Finally, perform replenishment prediction based on the target identifier, the target on-sale data, and the target sales volume level to obtain replenishment data. Therefore, the recommended method for product replenishment data illustrated in the embodiments of the present application can, by determining the target product according to the target sales volume level and performing replenishment prediction based on the target identifier, the target on-sale data, and the target sales volume level of the target product, use a computer to perform product replenishment prediction to replace manual replenishment prediction, improving the accuracy and efficiency of replenishment prediction.

[0141] Please refer to Figure 10 , in some embodiments, before step S101, the recommended method for the above product replenishment data may further include, but is not limited to, steps S1001 to S1006:

[0142] Step S1001, obtain the historical information of the training product from a preset product training set;

[0143] Step S1002, extract historical index data from the historical information according to the product classification parameters;

[0144] Step S1003, classify the historical information according to the product classification parameters and the historical index data to obtain product replenishment categories;

[0145] Step S1004, determine the product sales volume data of the product replenishment category according to the total number of historical information in the product replenishment category;

[0146] Step S1005, perform visualization according to the product classification parameters, the historical index data, and the product sales volume data to obtain an annotated classification visualization graph;

[0147] Step S1006, pre-train the product sales volume level prediction model according to the annotated classification visualization graph.

[0148] In step S1001 of some embodiments, the preset product training set is used to store the historical information of the training product. The training product is used to pre-train the product sales volume level prediction model. The historical information of the candidate product is the product information of the training product. Specific embodiments of the historical information can refer to the product information of the candidate product, which will not be elaborated in this application.

[0149] In step S1002 of some embodiments, historical indicator data can be extracted from historical information according to product classification parameters. The specific extraction method of the historical indicator data is similar to the specific extraction method of the product classification data, and the specific embodiments of step S102 can be referred to, which will not be elaborated in this application.

[0150] In step S1003 of some embodiments, for each piece of historical indicator data of the product classification parameters, a product replenishment category is set. Therefore, after extracting the historical indicator data from the historical information, the historical information can be classified into specific product replenishment categories according to the product classification parameters and the historical indicator data. For example, if the product classification parameters include product color and product size, the historical indicator data of the product color are "white" and "black", and the historical indicator data of the product size are "one size fits all", then the product replenishment categories include "replenishment category 1" and "replenishment category 2", where the historical indicator data corresponding to "replenishment category 1" are "white" and "one size fits all", and the historical indicator data corresponding to "replenishment category 2" are "black" and "one size fits all".

[0151] In step S1004 of some embodiments, after the classification of all the historical information is completed, the total number of historical information in each product replenishment category can be determined. On this basis, the total number of historical information in each product replenishment category can be determined as the product sales volume data of the product replenishment category.

[0152] In step S1005 of some embodiments, visualization can be performed according to the product classification parameters, the historical indicator data, and the product sales volume data to obtain an annotated classification visualization diagram. Specifically, first, visualization is performed according to the product classification parameters and the historical indicator data to obtain a training classification visualization diagram as shown in Figure 11 . The specific drawing method of the training classification visualization diagram can refer to the specific embodiments of step S103. Then, the historical sales volume level is determined according to the product sales volume data, and the training classification visualization diagram is annotated according to the historical sales volume level to obtain an annotated classification visualization diagram. As shown in Figure 12 , if the historical sales volume level is a hot sale, then "0 - Hot Sale" can be marked in the upper left corner of the training classification visualization diagram, and specific colors, such as green, are set for the outer border of "0 - Hot Sale" and the training classification visualization diagram.

[0153] In some embodiments, the historical sales level can be any one of the following: hot sale, best - selling, average, unpopular, or hard - to - sell. For example, when the commodity sales data is 0 - 10, the historical sales level is determined to be hard - to - sell; when the commodity sales data is 11 - 100, the historical sales level is determined to be unpopular; when the commodity sales data is 101 - 500, the historical sales level is determined to be average; when the commodity sales data is 501 - 1000, the historical sales level is determined to be best - selling; when the commodity sales data is greater than or equal to 1001, the historical sales level is determined to be hot sale.

[0154] In step S1006 of some embodiments, the commodity sales level prediction model can be pre - trained according to the labeled classification visualization graph. Specifically, the labeled classification visualization graph is input into the commodity sales level prediction model, and the commodity sales level prediction model predicts the commodity sales level of the labeled classification visualization graph to obtain the predicted sales level, and the commodity sales level prediction model is optimized by the historical sales level and the predicted sales level in the labeled classification visualization graph. Taking the deep - learning - based image detection algorithm YOLO - V4 as an example, the core principle of YOLO - V4 is to use a CNN to construct an end - to - end object detection model and perform object detection by means of multi - scale feature and multi - level feature fusion. The commodity sales level prediction model constructed based on YOLO - V4 can automatically adjust the structure of the neural network according to the historical sales level and the predicted sales level during the pre - training process to optimize the commodity sales level prediction model.

[0155] The steps S1001 to S1006 illustrated in the embodiments of the present application first obtain the historical information of the training commodities from a preset commodity training set and extract the historical index data from the historical information according to the commodity classification parameters. Then, the historical information is classified according to the commodity classification parameters and the historical index data to obtain the commodity replenishment categories, and the commodity sales data of the commodity replenishment categories is determined according to the total number of historical information in the commodity replenishment categories. Finally, visualization is performed according to the commodity classification parameters, the historical index data, and the commodity sales data to obtain the labeled classification visualization graph, and the commodity sales level prediction model is pre - trained according to the labeled classification visualization graph. Therefore, the recommended method for commodity replenishment data shown in the embodiments of the present application can, by determining the labeled classification visualization graph according to the historical information of the training commodities and pre - training the commodity sales level prediction model according to the labeled classification visualization graph, perform associative learning between the graphic features of the labeled classification visualization graph and the sales level of the commodity, enable the neural network to recognize the sales levels corresponding to different graphic features to obtain the commodity sales level prediction model, and then provide an intelligent decision - making basis for merchants to formulate replenishment plans and optimize inventory management through the commodity sales level prediction model.

[0156] In step S106 of some embodiments, after obtaining the replenishment data, replenishment recommendations can be made based on the replenishment data. Specifically, replenishment recommendation information is generated according to the product identification, replenishment time, replenishment quantity, and replenishment priority level, and then the replenishment recommendation information is pushed to the merchant according to the recommended time point and recommended method of replenishment. For example, the replenishment data can be: the product identification is "SP00001", the replenishment time is "20200321", the replenishment quantity is "20-100", the replenishment priority level is "first level", the recommended time point of replenishment is "20200220", and the recommended method of replenishment is SMS notification. Then, on February 20, 2020, replenishment recommendations will be made by sending an SMS to the merchant. The information recommended to the merchant includes: replenishing the product "SP00001" on March 21, 2020, and prompting that the replenishment priority level of this product is "first level", and the replenishment quantity is 20 to 100.

[0157] Please refer to Figure 13 , the embodiment of the present application also provides a recommendation system for product replenishment data, which can implement the above-mentioned recommendation method for product replenishment data. The system includes:

[0158] An information acquisition module 1301, configured to acquire product information of candidate products; wherein each candidate product has candidate on-sale data;

[0159] An index quantification module 1302, configured to extract product classification data from the product information according to preset product classification parameters;

[0160] An image construction module 1303, configured to perform visualization according to the product classification parameters and the product classification data to obtain a product classification visualization diagram;

[0161] A sales volume prediction module 1304, configured to perform a sales volume level prediction on the product classification visualization diagram according to a preset product sales volume level prediction model to obtain a target sales volume level;

[0162] A replenishment data prediction module 1305, configured to perform a replenishment prediction according to the candidate on-sale data, the target sales volume level, and the product classification visualization diagram to obtain replenishment data;

[0163] A replenishment recommendation module 1306, configured to make a replenishment recommendation according to the replenishment data.

[0164] The specific implementation manner of the recommendation system for product replenishment data is basically the same as the specific embodiments of the above-mentioned recommendation method for product replenishment data, and will not be elaborated here.

[0165] An embodiment of the present application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the above-mentioned method for recommending product replenishment data is implemented. The electronic device can be any intelligent terminal including a tablet computer, a vehicle-mounted computer, etc.

[0166] Please refer to Figure 14 , Figure 14 which schematically shows the hardware structure of an electronic device in another embodiment. The electronic device includes:

[0167] A processor 1401, which can be implemented in ways such as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present application;

[0168] A memory 1402, which can be implemented in forms such as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1402 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 1402, and the processor 1401 is called to execute the method for recommending product replenishment data in the embodiments of the present application;

[0169] An input / output interface 1403, which is used to implement information input and output;

[0170] A communication interface 1404, which is used to implement communication and interaction between this device and other devices, and can achieve communication through a wired method (such as USB, network cable, etc.) or through a wireless method (such as a mobile network, WIFI, Bluetooth, etc.);

[0171] A bus 1405, which transmits information between various components of the device (such as the processor 1401, the memory 1402, the input / output interface 1403, and the communication interface 1404);

[0172] Among them, the processor 1401, the memory 1402, the input / output interface 1403, and the communication interface 1404 are communicatively connected to each other inside the device through the bus 1405.

[0173] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the above-mentioned method for recommending merchandise replenishment data.

[0174] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0175] The method and system, electronic device, and storage medium for recommending merchandise replenishment data provided by the embodiments of the present application first obtain merchandise information of candidate merchandise; wherein each candidate merchandise has candidate on-sale data. Then, merchandise classification data is extracted from the merchandise information according to preset merchandise classification parameters, and visualization is performed based on the merchandise classification parameters and the merchandise classification data to obtain a merchandise classification visualization diagram, which can quantify and visualize the factors affecting the sales of candidate merchandise, and intuitively reflect the sales situation and potential demand of candidate merchandise in the market. Thereafter, a sales volume level prediction is performed on the merchandise classification visualization diagram according to a preset merchandise sales volume level prediction model to obtain a target sales volume level, and a replenishment prediction is performed based on the candidate on-sale data, the target sales volume level, and the merchandise classification visualization diagram to obtain replenishment data. Finally, replenishment recommendations are made based on the replenishment data. Therefore, the recommendation method and system, electronic device, and storage medium shown in the embodiments of the present application can, after visualizing the merchandise classification parameters and the merchandise classification data, perform a sales volume level prediction on the merchandise classification visualization diagram through the merchandise sales volume level prediction model to obtain a target sales volume level, and then determine the replenishment data based on the target sales volume level, the candidate on-sale data, and the merchandise classification visualization diagram, which can improve the accuracy of predicting replenishment data; moreover, the replenishment recommendation scheme recommended to merchants based on the replenishment data can improve the timeliness of replenishment data recommendation, so that merchants can grasp the sales dynamics of products in real time to make correct replenishment decisions, and then enable merchants to replenish inventory in a timely manner to avoid economic losses caused by out-of-stock or overstocking.

[0176] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0177] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation to the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0178] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0179] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and their appropriate combinations.

[0180] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above figures are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0181] It should be understood that in the present application, the expression "at least one (item)" or its similar expression refers to any combination of these items, including any combination of single item (s) or plural item (s). For example, at least one (item) of a, b, or c may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c may be single or multiple.

[0182] In several embodiments provided by the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the above division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of systems or units can be in electrical, mechanical or other forms.

[0183] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0184] In addition, each functional unit in various embodiments of the present application may be integrated into a processing unit, may exist separately as individual physical units, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0185] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present application. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs, etc., which can store programs.

[0186] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the rights of the embodiments of the present application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the rights of the embodiments of the present application.

Claims

1. A method for recommending product replenishment, characterized in that: The method comprises: Acquire product information of candidate products; wherein each of the candidate products has candidate sales data; Extracting commodity classification data from the commodity information according to preset commodity classification parameters; Perform visualization according to the commodity classification parameters and the commodity classification data to obtain a commodity classification visualization map; Predicting the sales level of the commodity classification visualization diagram according to a preset commodity sales level prediction model to obtain a target sales level; Perform replenishment prediction based on the candidate on-sale data, the target sales level, and the commodity classification visualization diagram to obtain replenishment data; A replenishment recommendation is made according to the replenishment data.

2. The method according to claim 1, characterized in that The step of performing visualization according to the commodity classification parameters and the commodity classification data to obtain a commodity classification visualization graph includes: According to the commodity classification parameters, searching for the index value range from a preset quantitative data table; Converting the commodity classification data into line segment visualization data according to the indicator value range; Visualization is performed according to the commodity classification parameters and the line segment visualization data to obtain the commodity classification visualization diagram.

3. The method according to claim 2, characterized in that The line segment visualization data includes: the original length of the line segment and the multiple of the interval length, the indicator value interval has an interval maximum value and an interval minimum value, and the converting of the commodity classification data into the line segment visualization data according to the indicator value interval includes: Determine the interval length of the indicator value interval according to the maximum value of the interval and the minimum value of the interval; Determine the total length of the data according to the commodity classification data and the minimum value of the interval; Determine the interval length multiple and the data margin length according to the interval length and the total data length; wherein the interval length multiple is the quotient of the total data length divided by the interval length, and the data margin length is the remainder of the total data length divided by the interval length; The data margin length is normalized according to the interval length to obtain the original length of the line segment.

4. The method according to claim 3, characterized in that The performing visualization according to the commodity classification parameters and the line segment visualization data to obtain the commodity classification visualization graph includes: A commodity classification visualization template is selected according to the commodity classification parameters; wherein the commodity classification visualization template includes: image size, central black dot, directional feature points and line segment direction of each commodity classification parameter, and the directional feature points are used to determine the direction of the commodity classification visualization image; Determine each indicator line segment according to the commodity classification parameter, the interval length multiple, the interval length, the line segment direction, the original length of the line segment and the central black dot; wherein the central black dot is the starting point of the indicator line segment; Determine the boundary line according to the original length of the line segment; The commodity classification visualization diagram is constructed according to the image size, the directional feature points, the indicator line segments and the boundary lines.

5. The method according to claim 4, characterized in that Determining each of the indicator line segments according to the commodity classification parameter, the interval length multiple, the interval length, the line segment direction, the original length of the line segment and the central black dot includes: Selecting a target length of the line segment from the interval length and the original length of the line segment according to the multiple of the interval length; Determine the original line segment according to the commodity classification parameter, the line segment target length, the line segment direction and the central black point; A length multiple icon is marked at the end of the original line segment according to the interval length multiple to obtain the indicator line segment.

6. The method according to claim 5, characterized in that The performing replenishment prediction according to the candidate on-sale data, the target sales level and the commodity classification visualization diagram to obtain replenishment data includes: Selecting a target classification visualization map from the commodity classification visualization maps according to the target sales level; Determining a target commodity according to the target classification visualization diagram; wherein the target commodity has a target identifier; using the candidate on-sale data of the target commodity as target on-sale data; A replenishment forecast is performed according to the target identifier, the target on-sale data and the target sales volume level to obtain the replenishment data.

7. The method according to any one of claims 1 to 6, characterized in that: Before obtaining the commodity information of the candidate commodity, the method further includes: Obtain historical information of training products from a preset product training set; Extracting historical indicator data from the historical information according to the commodity classification parameters; Classify the historical information according to the commodity classification parameter and the historical indicator data to obtain a commodity replenishment category; Determine the sales volume data of the commodity in the commodity replenishment category according to the total amount of historical information in the commodity replenishment category; Visualization is performed according to the commodity classification parameters, the historical indicator data, and the commodity sales data to obtain a labeled classification visualization graph; The commodity sales level prediction model is pre-trained according to the labeled classification visualization graph.

8. A recommendation system for commodity replenishment data, characterized in that: The system comprises: An information acquisition module, used to acquire product information of candidate products; wherein each of the candidate products has candidate sales data; An indicator quantification module, used to extract commodity classification data from the commodity information according to preset commodity classification parameters; An image construction module, used for performing visualization according to the commodity classification parameters and the commodity classification data to obtain a commodity classification visualization map; A sales volume prediction module, used to predict the sales volume level of the commodity classification visualization diagram according to a preset commodity sales volume level prediction model to obtain a target sales volume level; A replenishment data prediction module, used to perform replenishment prediction based on the candidate on-sale data, the target sales level and the commodity classification visualization diagram to obtain replenishment data; A replenishment recommendation module is used to make replenishment recommendations based on the replenishment data.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the method for recommending commodity replenishment data according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for recommending commodity replenishment data according to any one of claims 1 to 7 is implemented.

Images