Commodity correlation mining method, device, electronic device and readable storage medium
By vectorizing sales information, calculating co-occurrence and sales correlation between product categories, the problem of multi-dimensional product correlation calculation in the existing technology is solved, and more flexible and targeted product correlation is achieved. Sexual mining effect.
Patent Information
- Application Number
- CN202110812093.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-07-19
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-07-19
AI Technical Summary
In the prior art, the calculation of product category correlation is mainly based on the order model, and multi-dimensional correlation calculation cannot be performed, and commodity correlation cannot be mined from other directions except orders.
By determining the category name and sales information in the sales information, vectorized processing, calculate the co-occurrence correlation and sales correlation between product categories, comprehensively consider the co-occurrence correlation and sales correlation, and achieve multi-dimensional product correlation in product correlation in multiple dimensions. Dig.
It realizes the analysis of product correlation from different dimensions, improves the flexibility of mining calculations and the pertinence of results, and can more comprehensively explore the correlation between products.
Smart Images

Figure CN113610563B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of commodity correlation, and in particular to a commodity correlation mining method, device, electronic device and readable storage medium. Background Art
[0002] In the relevant technology, the calculation of product category correlation is currently mainly based on the order model, but does not support multi-dimensional correlation calculation, and cannot mine product correlation from other directions besides orders. Summary of the invention
[0003] The present invention aims to solve at least one of the technical problems existing in the prior art.
[0004] To this end, the first aspect of the present invention provides a commodity correlation mining method.
[0005] A second aspect of the present invention provides a commodity correlation mining device.
[0006] A third aspect of the present invention provides an electronic device.
[0007] A fourth aspect of the present invention provides a readable storage medium.
[0008] In view of this, the first aspect of the present invention proposes a commodity correlation mining method, comprising: determining sales information, wherein the sales information includes category name information and category sales information corresponding to each commodity category in N commodity categories, wherein N is an integer greater than 1; vectorizing the sales information to obtain a text vector containing the category name information and the category sales information; determining the sales volume of each commodity category in the N commodity categories and the number of occurrences of each commodity category in the sales information based on the text vector; determining the co-occurrence correlation between any two commodity categories based on the number of occurrences; determining the sales volume correlation between any two commodity categories based on the sales volume; determining the total correlation between any two commodity categories based on the co-occurrence correlation and the sales volume correlation.
[0009] The commodity correlation mining method provided by the present invention first determines the required sales information and processes the sales information to obtain a text vector containing the commodity category name and category sales information, then determines the co-occurrence correlation and sales volume correlation between different commodity categories based on the text vector, and then determines the total correlation between commodity categories based on the co-occurrence correlation and sales volume correlation.
[0010] Specifically, the sales volume of each product category and the number of times each product category appears in the sales information can be determined based on the text vector. The co-occurrence correlation between different product categories can be determined based on the number of occurrences. The sales volume correlation between different product categories can be determined based on the sales volume. By comprehensively considering the co-occurrence correlation and sales volume correlation, the total correlation of different product categories can be obtained.
[0011] Furthermore, the total correlation of different product categories can be determined based on the ratio between the co-occurrence correlation and the sales volume correlation.
[0012] Commodities can be divided into different commodity categories at the macro-category level, or at the stock keeping unit (SKU) level, which is the smallest classification level of commodities.
[0013] Through comprehensive evaluation of co-occurrence correlation and sales volume correlation, when importing different sales information, the correlation of products can be analyzed from different dimensions, thereby realizing multi-dimensional product correlation mining.
[0014] The commodity correlation mining method according to the above technical solution of the present invention may also have the following additional technical features:
[0015] In one possible design, determining sales information specifically includes: obtaining order data related to N commodity categories, wherein the order data includes an ordering account, an ordering time, and a sales volume corresponding to the order; determining first order data, second order data, and third order data in the order data, and determining sales information according to at least one of the first order data, the second order data, or the third order data; wherein the first order data is order data corresponding to each ordering account, the second order data is order data whose corresponding ordering time is within a preset time period, and the third order data is order data whose corresponding sales volume is within a preset sales volume range.
[0016] In this design, different sales information can be determined based on the order data, and then commodity correlation mining in different dimensions can be achieved based on different sales information. Specifically, the order data may include the ordering account, ordering time, and sales volume corresponding to the order. The first order data corresponding to each ordering account can be determined based on the ordering account, and the commodity category correlation can be mined from the dimension of the relationship between commodities and people through the first order data. The second order data within a preset time period can be determined based on the ordering time, and the commodity category correlation can be mined from the time dimension through the second order data. The third order data within the preset sales range can be determined based on the sales volume, and the commodity category correlation can be mined from the sales volume dimension through the third order data.
[0017] Furthermore, other data information can be determined based on order data, thereby realizing the mining of product category correlation in other dimensions. By determining sales information, the analysis of product correlation in multiple different dimensions is realized, making the mining calculation more flexible and the mining results more targeted.
[0018] In a possible design, the co-occurrence correlation between any two commodity categories is determined based on the number of occurrences, specifically including: determining the first weight of any commodity category based on the number of occurrences of any commodity category and the number of text vectors; determining the second weight of any two commodity categories based on the sum of the first weights corresponding to each commodity category in any two commodity categories; determining the number of co-occurrences of any two commodity categories appearing at the same time; determining the ratio of the number of co-occurrences to the number of text vectors, and determining the weight of the number of co-occurrences between any two commodity categories based on the ratio; and determining the co-occurrence correlation based on the co-occurrence weight and the second weight.
[0019] In this design, the number of co-occurrences of any two commodity categories can be determined based on the number of occurrences, and then the weight of the number of co-occurrences of any two commodity categories, that is, the correlation of commodity sales opportunities, can be determined, and then the co-occurrence correlation can be determined based on the weight of the number of occurrences. Specifically, based on the number of occurrences of any commodity and the number of text vectors, the weight of the commodity category appearing in all text vectors, that is, the first weight, can be determined. The first weights of the two commodities are summed to obtain the second weights of the two commodities. The number of times the two commodities appear at the same time, that is, the number of co-occurrences, is determined, and the ratio of the number of co-occurrences to the number of text vectors is used as the weight of the number of co-occurrences of the two commodity categories, that is, the correlation of commodity sales opportunities. Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight.
[0020] For example, there are 89 text vectors, of which product category A appears in the 1st and 10th vectors respectively; product category B appears in the 1st, 10th, 17th and 81st vectors respectively. Product category C appears in the 1st, 10th, 67th, 83rd and 88th vectors respectively. At this time, the ratio of the number of occurrences of product category A to the number of text vectors, 2 / 89, is used as the first weight of product category A; similarly, the first weight of product category B can be determined to be 4 / 89; and the first weight of product category C is 5 / 89. The first weight of product category A and the first weight of product category B are summed, and the second weights of product category A and product category B are obtained as 2 / 89+4 / 89=6 / 89. Similarly, the second weights of product category A and product category C can be obtained as 2 / 89+5 / 89=7 / 89.
[0021] Since product category A and product category B appear in both the 1st vector and the 10th vector, the co-occurrence times of product category A and product category B are determined to be 2. The ratio of the co-occurrence times 2 to the number of text vectors 89 is used as the co-occurrence weight of the two product categories, that is, the co-occurrence weight (product sales opportunity correlation) is 2 / 89. Similarly, the co-occurrence times of product category A and product category C are determined to be 2, and the co-occurrence weight (product sales opportunity correlation) is 2 / 89.
[0022] Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight. The ratio between the co-occurrence weight 2 / 89 and the second weight 6 / 89 between product category A and product category B is taken as the co-occurrence correlation, that is, (2 / 89) ÷ (6 / 89) = 0.33. The co-occurrence correlation between product category A and product category B is 0.33.
[0023] In a possible design, the sales correlation between any two commodity categories is determined based on sales volume, specifically including: determining the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, and determining the total sales volume of all commodity categories in the sales information; determining the first sales volume weight of the first commodity category and the second sales volume weight of the second commodity category respectively based on the first sales volume, the second sales volume and the total sales volume; determining the sales correlation between any two commodity categories based on the first sales volume weight and the second sales volume weight.
[0024] In this design, the sales correlation between any two commodity categories can be determined based on the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories. Specifically, first determine the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories in the sales information. Then determine the first sales volume weight based on the first sales volume and the total sales volume, and determine the second sales volume weight based on the second sales volume and the total sales volume. Finally, determine the sales correlation between the two commodity categories based on the first sales volume weight and the second sales volume weight.
[0025] For example, there are 89 text vectors in total. Based on these 89 text information, it can be determined that the total sales of all product categories is 1180. In addition, based on the text vectors, it can be determined that the first sales of product category A is 16, the second sales of product category B is 12, and the second sales of product category C is 110. Then the ratio between the first sales of product category A and the total sales of all product categories is used as the first sales weight of product category A, that is, 12÷1180=0.010. Similarly, it can be determined that the second sales weight of product category B is 16÷1180=0.013.
[0026] The correlation between commodity category A and commodity category B is determined based on the first sales weight of commodity category A and the second sales weight of commodity category B. Specifically, the ratio between the first sales weight of commodity category A and the second sales weight of commodity category B is used as the sales correlation between commodity category A and commodity category B, that is, 0.010÷0.013=0.769×100%=76.9%.
[0027] Similarly, the weight of the sales volume of product category C is 110÷1180=0.093, and the sales volume correlation between product category A and product category C is determined to be 0.010÷0.093=0.1075×100%=10.75%.
[0028] The second aspect of the present invention proposes a commodity correlation mining device, comprising: an acquisition unit, used to determine sales information, wherein the sales information includes category name information and category sales information corresponding to each commodity category in N commodity categories, wherein N is an integer greater than 1; the acquisition unit is also used to vectorize the sales information to obtain a text vector containing the category name information and the category sales information; a calculation unit, used to determine the sales volume of each commodity category in the N commodity categories and the number of occurrences of each commodity category in the sales information based on the text vector; the calculation unit is also used to determine the co-occurrence correlation between any two commodity categories based on the number of occurrences; the calculation unit is also used to determine the sales volume correlation between any two commodity categories based on the sales volume; the calculation unit is also used to determine the total correlation between any two commodity categories based on the co-occurrence correlation and the sales volume correlation.
[0029] The commodity correlation mining device provided by the present invention includes an acquisition unit and a computing unit. First, the required sales information is determined by the acquisition unit, and the sales information is processed to obtain a text vector containing the category name and category sales information of the commodity. Then, the computing unit is used to determine the co-occurrence correlation and sales correlation between different commodity categories based on the text vector, and then the total correlation between the commodity categories is determined based on the co-occurrence correlation and sales correlation.
[0030] Specifically, the computing unit can determine the sales volume of each commodity category and the number of times each commodity category appears in the sales information based on the text vector, determine the co-occurrence correlation between different commodity categories based on the number of occurrences, and determine the sales volume correlation between different commodity categories based on the sales volume. By comprehensively considering the co-occurrence correlation and sales volume correlation, the total correlation of different commodity categories can be obtained.
[0031] Furthermore, the total correlation of different product categories can be determined based on the ratio between the co-occurrence correlation and the sales volume correlation.
[0032] Commodities can be divided into different commodity categories at the macro-category level, or at the stock keeping unit (SKU) level, which is the smallest classification level of commodities.
[0033] Through comprehensive evaluation of co-occurrence correlation and sales volume correlation, when importing different sales information, the correlation of products can be analyzed from different dimensions, thereby realizing multi-dimensional product correlation mining.
[0034] The commodity correlation mining device according to the above technical solution of the present invention may also have the following additional technical features:
[0035] In a possible design, the acquisition unit is specifically used to: acquire order data related to N commodity categories, wherein the order data includes the order account, order time and sales volume corresponding to the order; determine first order data, second order data and third order data in the order data, and determine sales information according to at least one of the first order data, the second order data or the third order data; wherein the first order data is the order data corresponding to each order account, the second order data is the order data whose corresponding order time is within a preset time period, and the third order data is the order data whose corresponding sales volume is within a preset sales volume range.
[0036] In this design, the acquisition unit can determine different sales information based on the order data, and then implement commodity correlation mining in different dimensions based on the different sales information. Specifically, the order data may include the ordering account, ordering time and sales volume corresponding to the order. The first order data corresponding to each ordering account can be determined according to the ordering account, and the commodity category correlation can be mined from the dimension of the relationship between commodities and people through the first order data. The second order data within the preset time period can be determined according to the ordering time, and the commodity category correlation can be mined from the time dimension through the second order data. The third order data within the preset sales range can be determined according to the sales volume, and the commodity category correlation can be mined from the sales volume dimension through the third order data.
[0037] Furthermore, the acquisition unit can also determine other data information based on the order data, thereby realizing the mining of commodity category correlation in other dimensions. By determining the sales information, the analysis of commodity correlation in multiple different dimensions is realized, making the mining calculation more flexible and the mining results more targeted.
[0038] In one possible design, the operation unit includes a first operation subunit, which is used to determine the co-occurrence correlation between any two commodity categories based on the number of occurrences, specifically including: determining the first weight of any commodity category based on the number of occurrences of any commodity category and the number of text vectors; determining the second weight of any two commodity categories based on the sum of the first weights corresponding to each commodity category in any two commodity categories; determining the number of co-occurrences of any two commodity categories appearing at the same time; determining the ratio of the number of co-occurrences to the number of text vectors, and determining the co-occurrence weight between any two commodity categories based on the ratio; determining the co-occurrence correlation based on the co-occurrence weight and the second weight.
[0039] In this design, the calculation unit includes a first calculation subunit, which can determine the number of co-occurrences of any two commodity categories based on the number of occurrences, and then determine the weight of the number of co-occurrences of any two commodity categories, that is, the correlation of commodity sales opportunities, and then determine the co-occurrence correlation based on the weight of the number of occurrences. Specifically, according to the number of occurrences of any commodity and the number of text vectors, the weight of the commodity category appearing in all text vectors can be determined, that is, the first weight. The first weights of the two commodities are summed to obtain the second weights of the two commodities. Determine the number of times the two commodities appear at the same time, that is, the number of co-occurrences, and use the ratio of the number of co-occurrences to the number of text vectors as the weight of the number of co-occurrences of the two commodity categories, that is, the correlation of commodity sales opportunities. Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight.
[0040] For example, there are 89 text vectors, of which product category A appears in the 1st and 10th vectors respectively; product category B appears in the 1st, 10th, 17th and 81st vectors respectively. Product category C appears in the 1st, 10th, 67th, 83rd and 88th vectors respectively. At this time, the ratio of the number of occurrences of product category A to the number of text vectors, 2 / 89, is used as the first weight of product category A; similarly, the first weight of product category B can be determined to be 4 / 89; and the first weight of product category C is 5 / 89. The first weight of product category A and the first weight of product category B are summed, and the second weights of product category A and product category B are obtained as 2 / 89+4 / 89=6 / 89. Similarly, the second weights of product category A and product category C can be obtained as 2 / 89+5 / 89=7 / 89.
[0041] Since product category A and product category B appear in both the 1st vector and the 10th vector, the co-occurrence times of product category A and product category B are determined to be 2. The ratio of the co-occurrence times 2 to the number of text vectors 89 is used as the co-occurrence weight of the two product categories, that is, the co-occurrence weight (product sales opportunity correlation) is 2 / 89. Similarly, the co-occurrence times of product category A and product category C are determined to be 2, and the co-occurrence weight (product sales opportunity correlation) is 2 / 89.
[0042] Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight. The ratio between the co-occurrence weight 2 / 89 and the second weight 6 / 89 between product category A and product category B is taken as the co-occurrence correlation, that is, (2 / 89) ÷ (6 / 89) = 0.33. The co-occurrence correlation between product category A and product category B is 0.33.
[0043] In one possible design, the operation unit also includes a second operation subunit, which is used to determine the sales correlation between any two commodity categories based on the sales volume, specifically including: determining the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, and determining the total sales volume of all commodity categories in the sales information; determining the first sales volume weight of the first commodity category and the second sales volume weight of the second commodity category respectively according to the first sales volume, the second sales volume and the total sales volume; determining the sales correlation between any two commodity categories according to the first sales volume weight and the second sales volume weight.
[0044] In this design, the calculation unit also includes a second calculation subunit, which can determine the sales correlation between any two commodity categories based on the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories. Specifically, first determine the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories in the sales information. Then determine the first sales volume weight based on the first sales volume and the total sales volume, and determine the second sales volume weight based on the second sales volume and the total sales volume. Finally, determine the sales correlation between the two commodity categories based on the first sales volume weight and the second sales volume weight.
[0045] For example, there are 89 text vectors in total. Based on these 89 text information, it can be determined that the total sales of all product categories is 1180. In addition, based on the text vectors, it can be determined that the first sales of product category A is 16, the second sales of product category B is 12, and the second sales of product category C is 110. Then the ratio between the first sales of product category A and the total sales of all product categories is used as the first sales weight of product category A, that is, 12÷1180=0.010. Similarly, it can be determined that the second sales weight of product category B is 16÷1180=0.013.
[0046] The correlation between commodity category A and commodity category B is determined based on the first sales weight of commodity category A and the second sales weight of commodity category B. Specifically, the ratio between the first sales weight of commodity category A and the second sales weight of commodity category B is used as the sales correlation between commodity category A and commodity category B, that is, 0.010÷0.013=0.769×100%=76.9%.
[0047] Similarly, the weight of the sales volume of product category C is 110÷1180=0.093, and the sales volume correlation between product category A and product category C is determined to be 0.010÷0.093=0.1075×100%=10.75%.
[0048] The third aspect of the present invention provides an electronic device, including a processor and a memory, wherein the processor is used to implement the steps of the commodity correlation mining method of any design of the first aspect of the present invention when executing the program or instruction stored in the memory.
[0049] The electronic device provided by the present invention includes a memory and a processor. The memory can store programs, and the processor can execute the programs stored in the memory to implement the steps of the commodity correlation mining method of any design of the first aspect above, so it has all the beneficial effects of the commodity correlation mining method, which will not be repeated here.
[0050] The fourth aspect of the present invention proposes a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the product correlation mining method as designed in any of the first aspects above are implemented, and thus the product correlation mining method has all the beneficial effects, which will not be repeated here.
[0051] Additional aspects and advantages of the present invention will become apparent from the following description or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:
[0053] Figure 1 A flowchart showing a method for mining commodity correlation according to an embodiment of the first aspect of the present invention is shown;
[0054] Figure 2 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0055] Figure 3 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0056] Figure 4 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0057] Figure 5 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0058] Figure 6 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0059] Figure 7 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0060] Figure 8 A flowchart showing a method for mining commodity correlation according to another embodiment of the first aspect of the present invention;
[0061] Fig. 9 A structural block diagram of a commodity correlation mining device according to an embodiment of the second aspect of the present invention is shown;
[0062] Fig.10 A structural block diagram of an electronic device according to an embodiment of the third aspect of the present invention is shown.
[0063] in, Fig. 9 and Fig.10 The corresponding relationship between the reference numerals and the component names is as follows:
[0064] 100 commodity correlation mining device, 110 acquisition unit; 120 operation unit; 122 first operation subunit; 124 second operation subunit; 200 electronic device, 210 processor, 220 memory. DETAILED DESCRIPTION
[0065] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.
[0066] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.
[0067] like Figure 1 As shown, an embodiment of the first aspect of the present invention provides a commodity correlation mining method, the method comprising:
[0068] S102, determining sales information, wherein the sales information includes category name information and category sales information corresponding to each of N commodity categories, wherein N is an integer greater than 1;
[0069] S104, performing vectorization processing on the sales information to obtain a text vector including category name information and category sales information;
[0070] S106, determining the sales volume of each of the N product categories and the number of occurrences of each product category in the sales information according to the text vector;
[0071] S108, determining the co-occurrence correlation between any two product categories according to the number of occurrences;
[0072] S110, determining the sales volume correlation between any two commodity categories according to the sales volume;
[0073] S112: Determine the total correlation between any two product categories based on the co-occurrence correlation and the sales volume correlation.
[0074] The commodity correlation mining method proposed in this embodiment first determines the required sales information and processes the sales information to obtain a text vector containing the commodity category name and category sales information, then determines the co-occurrence correlation and sales volume correlation between different commodity categories based on the text vector, and then determines the total correlation between commodity categories based on the co-occurrence correlation and sales volume correlation.
[0075] Specifically, the sales volume of each product category and the number of times each product category appears in the sales information can be determined based on the text vector. The co-occurrence correlation between different product categories can be determined based on the number of occurrences. The sales volume correlation between different product categories can be determined based on the sales volume. By comprehensively considering the co-occurrence correlation and sales volume correlation, the total correlation of different product categories can be obtained.
[0076] Furthermore, the total correlation of different product categories can be determined based on the ratio between the co-occurrence correlation and the sales volume correlation.
[0077] Commodities can be divided into different commodity categories at the macro-category level, or at the stock keeping unit (SKU) level, which is the smallest classification level of commodities.
[0078] Through comprehensive evaluation of co-occurrence correlation and sales volume correlation, when importing different sales information, the correlation of products can be analyzed from different dimensions, thereby realizing multi-dimensional product correlation mining.
[0079] Figure 2 A commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. As shown in the figure, the method includes:
[0080] S202, importing sales information to obtain a text vector;
[0081] S204, calculating the correlation between the two product category names according to the text vector;
[0082] S206, obtaining the commodity selling opportunity correlation and sales volume correlation between the two commodity categories, and calculating the total correlation.
[0083] In this embodiment, the sales information is first imported, and the imported sales information is vectorized to obtain a text vector containing the category name information and category sales information corresponding to each commodity category. Then, based on the text vector, the correlation between the names of two commodity categories is calculated. Then, the commodity sales opportunity correlation (i.e., the co-occurrence weight) and sales volume correlation between the two commodity categories are calculated. Finally, the total correlation between the two commodities is determined based on the sales opportunity correlation and the sales volume correlation. Specifically, the co-occurrence correlation is determined based on the sales opportunity correlation, and then the total correlation is determined based on the co-occurrence correlation and the sales volume correlation.
[0084] In some embodiments, determining sales information specifically includes: obtaining order data related to N product categories, wherein the order data includes the ordering account, ordering time and sales volume corresponding to the order; determining first order data, second order data and third order data in the order data, and determining sales information according to at least one of the first order data, the second order data or the third order data; wherein the first order data is order data corresponding to each ordering account, the second order data is order data whose corresponding ordering time is within a preset time period, and the third order data is order data whose corresponding sales volume is within a preset sales volume range.
[0085] Figure 3 A flowchart of a commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. The method comprises:
[0086] S302, obtaining order data related to N product categories, wherein the order data includes the ordering account, ordering time, and sales volume corresponding to the order;
[0087] S304, determining first order data, second order data, and third order data in the order data, and determining sales information according to at least one of the first order data, the second order data, or the third order data.
[0088] In this embodiment, different sales information can be determined based on the order data, and then commodity correlation mining in different dimensions can be achieved based on the different sales information. Specifically, the order data may include the ordering account, ordering time, and sales volume corresponding to the order. The first order data corresponding to each ordering account can be determined based on the ordering account, and the commodity category correlation can be mined from the dimension of the relationship between commodities and people through the first order data. The second order data within a preset time period can be determined based on the ordering time, and the commodity category correlation can be mined from the time dimension through the second order data. The third order data within the preset sales range can be determined based on the sales volume, and the commodity category correlation can be mined from the sales volume dimension through the third order data.
[0089] Furthermore, other data information can be determined based on order data, thereby realizing the mining of product category correlation in other dimensions. By determining sales information, the analysis of product correlation in multiple different dimensions is realized, making the mining calculation more flexible and the mining results more targeted.
[0090] Figure 4 A commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. As shown in the figure, the method includes:
[0091] S402, importing the order purchase category of each customer, and taking the first order data of each purchaser as a text vector;
[0092] S404, obtaining the commodity category relevance based on the purchaser dimension.
[0093] In this embodiment, the order data includes the ordering account (i.e., the customer) corresponding to the order, and the first order data corresponding to each ordering account can be determined according to the ordering account, and then the required sales information can be determined according to the first order data. The sales information is imported, wherein the sales information includes the first order data with each ordering account, wherein the first order data includes the purchased commodity category in the order of each ordering account, and the first order data of each order account is processed into a text vector, and then the commodity category relevance based on the dimension of the order account, i.e., the buyer, is determined according to each text vector, that is, the commodity category relevance is mined from the dimension of the relationship between commodities and people.
[0094] Figure 5 A commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. As shown in the figure, the method includes:
[0095] S502, importing the categories of goods purchased at a specific time, and taking the second order data of each period as a text vector;
[0096] S504, obtaining the commodity category relevance based on the time dimension.
[0097] In this embodiment, the order data includes the order placement time corresponding to the order, and the second order data within the preset time period can be determined according to the order placement time, and then the required sales information can be determined according to the second order data. The sales information is imported, wherein the sales information includes the second order data with the order placement time, wherein the second order data includes the purchased commodity categories in the order within the preset time period, and the second order data within each preset time period is processed into a text vector, and then the commodity category relevance based on the time dimension is determined according to each text vector, that is, the commodity category relevance is mined from the time dimension.
[0098] Figure 6 A commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. As shown in the figure, the method includes:
[0099] S602, importing product categories in a specific sales range;
[0100] S604, obtaining commodity category relevance based on sales volume.
[0101] In this embodiment, the order data includes the sales volume corresponding to the order, and the third order data within the preset sales volume range can be determined according to the sales volume, and then the required sales information can be determined according to the third order data. The sales information is imported, wherein the sales information includes the third order data with sales volume, wherein the third order data includes the purchased commodity categories in the order within the preset sales volume range, and each third order data within the preset sales volume range is processed into a text vector, and then the commodity category relevance based on the sales volume dimension is determined according to each text vector, that is, the commodity category relevance is mined from the sales volume dimension.
[0102] In some embodiments, the co-occurrence correlation between any two commodity categories is determined based on the number of occurrences, specifically including: determining a first weight of any commodity category based on the number of occurrences of any commodity category and the number of text vectors; determining a second weight of any two commodity categories based on the sum of the first weights corresponding to each commodity category in any two commodity categories; determining the number of co-occurrences of any two commodity categories appearing at the same time; determining a ratio of the number of co-occurrences to the number of text vectors, and determining a weight of the number of co-occurrences between any two commodity categories based on the ratio; and determining the co-occurrence correlation based on the co-occurrence weight and the second weight.
[0103] Figure 7 A flowchart of a commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. The method comprises:
[0104] S702, determining a first weight of any product category according to the number of occurrences of any product category and the number of text vectors;
[0105] S704, determining the second weights of any two commodity categories according to the sum of the first weights corresponding to each commodity category in any two commodity categories;
[0106] S706, determining the number of co-occurrences of any two product categories appearing at the same time;
[0107] S708, determining a ratio of the number of co-occurrences to the number of text vectors, and determining a weight of the number of co-occurrences between any two product categories according to the ratio;
[0108] S710, determining the co-occurrence correlation according to the co-occurrence number weight and the second weight.
[0109] In this embodiment, the number of co-occurrences of any two commodity categories can be determined based on the number of occurrences, and then the weight of the number of co-occurrences of any two commodity categories, that is, the commodity sales opportunity correlation, can be determined, and then the co-occurrence correlation can be determined based on the number of occurrences weight. Specifically, based on the number of occurrences of any commodity and the number of text vectors, the weight of the commodity category appearing in all text vectors, that is, the first weight, can be determined. The first weights of the two commodities are summed to obtain the second weights of the two commodities. The number of times the two commodities appear at the same time, that is, the number of co-occurrences, is determined, and the ratio of the number of co-occurrences to the number of text vectors is used as the weight of the number of co-occurrences of the two commodity categories, that is, the commodity sales opportunity correlation. Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight.
[0110] For example, there are 89 text vectors, of which product category A appears in the 1st and 10th vectors respectively; product category B appears in the 1st, 10th, 17th and 81st vectors respectively. Product category C appears in the 1st, 10th, 67th, 83rd and 88th vectors respectively. At this time, the ratio of the number of occurrences of product category A to the number of text vectors, 2 / 89, is used as the first weight of product category A; similarly, the first weight of product category B can be determined to be 4 / 89; and the first weight of product category C is 5 / 89. The first weight of product category A and the first weight of product category B are summed, and the second weights of product category A and product category B are obtained as 2 / 89+4 / 89=6 / 89. Similarly, the second weights of product category A and product category C can be obtained as 2 / 89+5 / 89=7 / 89.
[0111] Since product category A and product category B appear in both the 1st vector and the 10th vector, the co-occurrence times of product category A and product category B are determined to be 2. The ratio of the co-occurrence times 2 to the number of text vectors 89 is used as the co-occurrence weight of the two product categories, that is, the co-occurrence weight (product sales opportunity correlation) is 2 / 89. Similarly, the co-occurrence times of product category A and product category C are determined to be 2, and the co-occurrence weight (product sales opportunity correlation) is 2 / 89.
[0112] Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight. The ratio between the co-occurrence weight 2 / 89 and the second weight 6 / 89 between product category A and product category B is taken as the co-occurrence correlation, that is, (2 / 89) ÷ (6 / 89) = 0.33. The co-occurrence correlation between product category A and product category B is 0.33.
[0113] Figure 8 A flowchart of a commodity correlation mining method according to another embodiment of the first aspect of the present invention is shown. The method comprises:
[0114] S802, determining a first sales volume of a first product category and a second sales volume of a second product category in any two product categories, and determining a total sales volume of all product categories in the sales information;
[0115] S804, determining a first sales volume weight of the first product category and a second sales volume weight of the second product category respectively according to the first sales volume, the second sales volume and the total sales volume;
[0116] S806: Determine the sales volume correlation between any two commodity categories according to the first sales volume weight and the second sales volume weight.
[0117] In this embodiment, the sales correlation between any two commodity categories can be determined based on the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories. Specifically, first determine the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories in the sales information. Then determine the first sales volume weight based on the first sales volume and the total sales volume, and determine the second sales volume weight based on the second sales volume and the total sales volume. Finally, determine the sales correlation between the two commodity categories based on the first sales volume weight and the second sales volume weight.
[0118] For example, there are 89 text vectors in total. Based on these 89 text information, it can be determined that the total sales of all product categories is 1180. In addition, based on the text vectors, it can be determined that the first sales of product category A is 16, the second sales of product category B is 12, and the second sales of product category C is 110. Then the ratio between the first sales of product category A and the total sales of all product categories is used as the first sales weight of product category A, that is, 12÷1180=0.010. Similarly, it can be determined that the second sales weight of product category B is 16÷1180=0.013.
[0119] The correlation between commodity category A and commodity category B is determined based on the first sales weight of commodity category A and the second sales weight of commodity category B. Specifically, the ratio between the first sales weight of commodity category A and the second sales weight of commodity category B is used as the sales correlation between commodity category A and commodity category B, that is, 0.010÷0.013=0.769×100%=76.9%.
[0120] Similarly, the weight of the sales volume of product category C is 110÷1180=0.093, and the sales volume correlation between product category A and product category C is determined to be 0.010÷0.093=0.1075×100%=10.75%.
[0121] An embodiment of the second aspect of the present invention proposes a commodity correlation mining device 100, including: an acquisition unit 110, used to determine sales information, wherein the sales information includes category name information and category sales information corresponding to each commodity category in N commodity categories, wherein N is an integer greater than 1; the acquisition unit 110 is also used to vectorize the sales information to obtain a text vector containing the category name information and the category sales information; an operation unit 120 is used to determine the sales volume of each commodity category in the N commodity categories and the number of occurrences of each commodity category in the sales information based on the text vector; the operation unit 120 is also used to determine the co-occurrence correlation between any two commodity categories based on the number of occurrences; the operation unit 120 is also used to determine the sales volume correlation between any two commodity categories based on the sales volume; the operation unit 120 is also used to determine the total correlation between any two commodity categories based on the co-occurrence correlation and the sales volume correlation.
[0122] The commodity correlation mining device 100 proposed in this embodiment includes an acquisition unit 110 and a computing unit 120. First, the acquisition unit 110 determines the required sales information and processes the sales information to obtain a text vector containing the category name and category sales information of the commodity. Then, the computing unit 120 determines the co-occurrence correlation and sales correlation between different commodity categories based on the text vector, and then determines the total correlation between the commodity categories based on the co-occurrence correlation and sales correlation.
[0123] Specifically, the operation unit 120 can determine the sales volume of each commodity category and the number of times each commodity category appears in the sales information based on the text vector, determine the co-occurrence correlation between different commodity categories based on the number of occurrences, and determine the sales volume correlation between different commodity categories based on the sales volume. By comprehensively considering the co-occurrence correlation and the sales volume correlation, the total correlation of different commodity categories can be obtained.
[0124] Furthermore, the total correlation of different product categories can be determined based on the ratio between the co-occurrence correlation and the sales volume correlation.
[0125] Commodities can be divided into different commodity categories at the macro-category level, or at the stock keeping unit (SKU) level, which is the smallest classification level of commodities.
[0126] Through comprehensive evaluation of co-occurrence correlation and sales volume correlation, when importing different sales information, the correlation of products can be analyzed from different dimensions, thereby realizing multi-dimensional product correlation mining.
[0127] In some embodiments, the acquisition unit 110 is specifically used to: acquire order data related to N commodity categories, wherein the order data includes the order account, order time and sales volume corresponding to the order; determine first order data, second order data and third order data in the order data, and determine sales information according to at least one of the first order data, the second order data or the third order data; wherein the first order data is the order data corresponding to each order account, the second order data is the order data whose corresponding order time is within a preset time period, and the third order data is the order data whose corresponding sales volume is within a preset sales volume range.
[0128] In this embodiment, the acquisition unit 110 can determine different sales information based on the order data, and then implement commodity correlation mining in different dimensions based on the different sales information. Specifically, the order data may include the ordering account, ordering time and sales volume corresponding to the order. The first order data corresponding to each ordering account can be determined according to the ordering account, and the commodity category correlation can be mined from the dimension of the relationship between the commodity and the person through the first order data. The second order data within the preset time period can be determined according to the ordering time, and the commodity category correlation can be mined from the time dimension through the second order data. The third order data within the preset sales range can be determined according to the sales volume, and the commodity category correlation can be mined from the sales volume dimension through the third order data.
[0129] Furthermore, the acquisition unit 110 can also determine other data information based on the order data, thereby realizing the mining of commodity category relevance in other dimensions. By determining the sales information, the analysis of commodity relevance in multiple different dimensions is realized, making the mining calculation more flexible and the mining results more targeted.
[0130] In some embodiments, the operation unit 120 includes a first operation subunit 122, which is used to determine the co-occurrence correlation between any two commodity categories based on the number of occurrences, specifically including: determining the first weight of any commodity category based on the number of occurrences of any commodity category and the number of text vectors; determining the second weight of any two commodity categories based on the sum of the first weights corresponding to each commodity category in any two commodity categories; determining the number of co-occurrences of any two commodity categories appearing at the same time; determining the ratio of the number of co-occurrences to the number of text vectors, and determining the co-occurrence weight between any two commodity categories based on the ratio; determining the co-occurrence correlation based on the co-occurrence weight and the second weight.
[0131] In this embodiment, the calculation unit includes a first calculation subunit, which can determine the number of co-occurrences of any two commodity categories based on the number of occurrences, and then determine the weight of the number of co-occurrences of any two commodity categories, that is, the commodity sales opportunity correlation, and then determine the co-occurrence correlation based on the number of occurrences weight. Specifically, according to the number of occurrences of any commodity and the number of text vectors, the weight of the commodity category appearing in all text vectors can be determined, that is, the first weight. The first weights of the two commodities are summed to obtain the second weights of the two commodities. The number of times the two commodities appear at the same time, that is, the number of co-occurrences, is determined, and the ratio of the number of co-occurrences to the number of text vectors is used as the weight of the number of co-occurrences of the two commodity categories, that is, the commodity sales opportunity correlation. Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight.
[0132] For example, there are 89 text vectors, of which product category A appears in the 1st and 10th vectors respectively; product category B appears in the 1st, 10th, 17th and 81st vectors respectively. Product category C appears in the 1st, 10th, 67th, 83rd and 88th vectors respectively. At this time, the ratio of the number of occurrences of product category A to the number of text vectors, 2 / 89, is used as the first weight of product category A; similarly, the first weight of product category B can be determined to be 4 / 89; and the first weight of product category C is 5 / 89. The first weight of product category A and the first weight of product category B are summed, and the second weights of product category A and product category B are obtained as 2 / 89+4 / 89=6 / 89. Similarly, the second weights of product category A and product category C can be obtained as 2 / 89+5 / 89=7 / 89.
[0133] Since product category A and product category B appear in both the 1st vector and the 10th vector, the co-occurrence times of product category A and product category B are determined to be 2. The ratio of the co-occurrence times 2 to the number of text vectors 89 is used as the co-occurrence weight of the two product categories, that is, the co-occurrence weight (product sales opportunity correlation) is 2 / 89. Similarly, the co-occurrence times of product category A and product category C are determined to be 2, and the co-occurrence weight (product sales opportunity correlation) is 2 / 89.
[0134] Finally, the co-occurrence correlation is determined based on the co-occurrence weight and the second weight. The ratio between the co-occurrence weight 2 / 89 and the second weight 6 / 89 between product category A and product category B is taken as the co-occurrence correlation, that is, (2 / 89) ÷ (6 / 89) = 0.33. The co-occurrence correlation between product category A and product category B is 0.33.
[0135] In some embodiments, the operation unit 120 also includes a second operation subunit 124, which is used to determine the sales correlation between any two commodity categories based on the sales volume, specifically including: determining the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, and determining the total sales volume of all commodity categories in the sales information; determining the first sales weight of the first commodity category and the second sales weight of the second commodity category according to the first sales volume, the second sales volume and the total sales volume; determining the sales correlation between any two commodity categories according to the first sales weight and the second sales weight.
[0136] In this embodiment, the calculation unit also includes a second calculation subunit, which can determine the sales correlation between any two commodity categories based on the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories. Specifically, first determine the first sales volume of the first commodity category and the second sales volume of the second commodity category in any two commodity categories, as well as the total sales volume of all commodity categories in the sales information. Then determine the first sales volume weight based on the first sales volume and the total sales volume, and determine the second sales volume weight based on the second sales volume and the total sales volume. Finally, determine the sales correlation between the two commodity categories based on the first sales volume weight and the second sales volume weight.
[0137] For example, there are 89 text vectors in total. Based on these 89 text information, it can be determined that the total sales of all product categories is 1180. In addition, based on the text vectors, it can be determined that the first sales of product category A is 16, the second sales of product category B is 12, and the second sales of product category C is 110. Then the ratio between the first sales of product category A and the total sales of all product categories is used as the first sales weight of product category A, that is, 12÷1180=0.010. Similarly, it can be determined that the second sales weight of product category B is 16÷1180=0.013.
[0138] The correlation between commodity category A and commodity category B is determined based on the first sales weight of commodity category A and the second sales weight of commodity category B. Specifically, the ratio between the first sales weight of commodity category A and the second sales weight of commodity category B is used as the sales correlation between commodity category A and commodity category B, that is, 0.010÷0.013=0.769×100%=76.9%.
[0139] Similarly, the weight of the sales volume of product category C is 110÷1180=0.093, and the sales volume correlation between product category A and product category C is determined to be 0.010÷0.093=0.1075×100%=10.75%.
[0140] Fig. 9Another embodiment of the second aspect of the present invention proposes a commodity correlation mining device 100, including an acquisition unit 110 and a calculation unit, wherein the calculation unit includes a first calculation subunit and a second calculation subunit. Specifically, the acquisition unit 110 can determine sales information, and process the sales information to obtain a text vector. The calculation unit can determine the co-occurrence correlation and sales correlation of different commodity categories based on the text vector, and then determine the total correlation based on the co-occurrence correlation and sales correlation. The calculation unit includes a first calculation subunit and a second calculation subunit, wherein the first calculation subunit can determine the co-occurrence correlation based on the text vector, and the second calculation subunit can determine the sales correlation based on the text vector.
[0141] like Fig.10 As shown, an embodiment of the third aspect of the present invention proposes an electronic device 200, including a processor 210 and a memory 220, wherein the processor 210 is used to implement the steps of the commodity correlation mining method as designed in any one of the first aspects of the present invention when executing the program or instructions stored in the memory 220.
[0142] The electronic device 200 provided in this embodiment includes a memory 220 and a processor 210. The memory 220 can store programs, and the processor 210 can execute the programs stored in the memory 220 to implement the steps of the commodity correlation mining method of any design of the first aspect above, and thus has all the beneficial effects of the commodity correlation mining method, which will not be repeated here.
[0143] An embodiment of the fourth aspect of the present invention provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the product correlation mining method as in any embodiment of the first aspect are implemented, and thus the product correlation mining method has all the beneficial effects, which will not be repeated here.
[0144] The readable storage medium may include any medium capable of storing or transmitting information. Examples of readable storage media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, and the like. The code segments may be downloaded via a network such as the Internet, an intranet, and the like.
[0145] It should be noted that, in this article, the term "comprises", "includes" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or computer device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or computer device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or computer device including the element. In addition, it should be pointed out that the scope of the method and computer device in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.
[0146] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.
[0147] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A commodity correlation mining method, characterized in that: include: Determine sales information, wherein the sales information includes category name information and category sales information corresponding to each of N product categories, where N is an integer greater than 1; Performing vectorization processing on the sales information to obtain a text vector including the category name information and the category sales information; Determine, according to the text vector, the sales volume of each of the N product categories and the number of occurrences of each product category in the sales information; Determine the co-occurrence correlation between any two product categories based on the occurrence counts; Determine the sales volume correlation between any two commodity categories according to the sales volume; Determine the total correlation between any two commodity categories according to the co-occurrence correlation and the sales volume correlation; The determining of sales information specifically includes: obtaining order data related to the N commodity categories, wherein the order data includes an ordering account, ordering time, and sales volume corresponding to the order; Determining first order data, second order data, and third order data in the order data, and determining the sales information according to at least one of the first order data, the second order data, or the third order data; The first order data is the order data corresponding to each ordering account, the second order data is the order data corresponding to the ordering time within a preset time period, and the third order data is the order data corresponding to the sales volume within a preset sales volume range; Determining the co-occurrence correlation between any two commodity categories according to the number of occurrences specifically includes: Determine a first weight of any product category according to the number of occurrences of any product category and the number of the text vectors; Determine the second weights of any two commodity categories according to the sum of the first weights corresponding to each commodity category in any two commodity categories; Determine the number of co-occurrences of any two product categories appearing at the same time; Determine a ratio of the number of co-occurrences to the number of text vectors, and determine a weight of the number of co-occurrences between any two commodity categories according to the ratio; Determining the co-occurrence correlation according to the co-occurrence number weight and the second weight; Determining the sales volume correlation between any two commodity categories according to the sales volume specifically includes: Determine a first sales volume of a first commodity category and a second sales volume of a second commodity category in any two commodity categories, and determine a total sales volume of all commodity categories in the sales information; Determine, according to the first sales volume, the second sales volume, and the total sales volume, a first sales volume weight of the first commodity category and a second sales volume weight of the second commodity category; The sales volume correlation between any two commodity categories is determined according to the first sales volume weight and the second sales volume weight.
2. A commodity correlation mining device, characterized in that: include: An acquisition unit, configured to determine sales information, wherein the sales information includes category name information and category sales information corresponding to each of N commodity categories, wherein N is an integer greater than 1; The acquisition unit is further used to perform vectorization processing on the sales information to obtain a text vector containing the category name information and the category sales information; a computing unit, configured to determine, based on the text vector, the sales volume of each of the N commodity categories and the number of occurrences of each commodity category in the sales information; The computing unit is further used to determine the co-occurrence correlation between any two commodity categories according to the number of occurrences; The computing unit is further configured to determine the sales volume correlation between any two commodity categories according to the sales volume; The computing unit is further configured to determine the total correlation between any two commodity categories according to the co-occurrence correlation and the sales volume correlation; The acquisition unit is specifically used for: Acquire order data related to the N product categories, wherein the order data includes the ordering account, ordering time, and sales volume corresponding to the order; Determining first order data, second order data, and third order data in the order data, and determining the sales information according to at least one of the first order data, the second order data, or the third order data; The first order data is the order data corresponding to each ordering account, the second order data is the order data corresponding to the ordering time within a preset time period, and the third order data is the order data corresponding to the sales volume within a preset sales volume range; The operation unit comprises a first operation subunit, which is used for: Determine a first weight of any product category according to the number of occurrences of any product category and the number of the text vectors; Determine the second weights of any two commodity categories according to the sum of the first weights corresponding to each commodity category in any two commodity categories; Determine the number of co-occurrences of any two product categories appearing at the same time; Determine a ratio of the number of co-occurrences to the number of text vectors, and determine a weight of the number of co-occurrences between any two commodity categories according to the ratio; Determining the co-occurrence correlation according to the co-occurrence number weight and the second weight; The operation unit further includes a second operation subunit, which is used for: Determine a first sales volume of a first commodity category and a second sales volume of a second commodity category in any two commodity categories, and determine a total sales volume of all commodity categories in the sales information; Determine, according to the first sales volume, the second sales volume, and the total sales volume, a first sales volume weight of the first commodity category and a second sales volume weight of the second commodity category; The sales volume correlation between any two commodity categories is determined according to the first sales volume weight and the second sales volume weight.
3. An electronic device, characterized in that: The electronic device includes a processor and a memory, and the processor is used to implement the steps of the commodity correlation mining method according to claim 1 when executing the program or instructions stored in the memory.
4. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps of the commodity correlation mining method according to claim 1 are implemented.
