Fast moving consumer goods correlation analysis method and device based on consumer purchase data

Abstract

The invention relates to the technical field of data processing, and discloses a fast moving consumer goods correlation analysis method and device based on consumer purchase data, and the method comprises the steps: obtaining the product data of fast moving consumer goods and the consumption code scanning data of a plurality of member consumers for the fast moving consumer goods; based on the consumption code scanning data and the product data, clustering fast moving consumer goods varieties of all fast moving consumer goods purchased by the plurality of member consumers to obtain a plurality of fast moving consumer goods clustering results; clustering the plurality of member consumers based on the consumption code scanning data to obtain a plurality of consumer clustering results, the plurality of consumer clustering results including multivariate consumers; a plurality of association rules are determined based on fast moving consumer goods varieties purchased by multivariate consumers, and the support degree and the confidence degree are calculated. By determining the relevance of the fast moving consumer goods of the same brand and the relevance between the fast moving consumer goods and the purchase behavior of the consumer, enterprises can more accurately formulate marketing strategies and optimize product combinations.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for analyzing the association of fast-moving consumer goods based on consumer purchase data. Background Art

[0002] In the context of increasingly fierce competition in the FMCG market, accurately grasping consumer demand has become the key to improving the competitiveness and expanding the market share of FMCG brands. With the rapid development of big data technology, companies are actively using consumer purchasing behavior to deeply analyze and explore the potential correlation of FMCG, in order to develop more accurate and efficient marketing strategies.

[0003] At present, in the data collection stage to determine the relevance of FMCG, traditional methods such as brand membership registration, offline visits and surveys, and business department submissions are mainly relied on. In the analysis stage, the relevance of FMCG is manually judged by combining historical data reports and personal experience. Finally, it is applied to offline promotion scenarios to infer whether the above-obtained FMCG relevance is accurate. However, the above methods are limited by the scope and timeliness of collection, and the data is often incomplete. In addition, due to the complex data sources, it is impossible to accurately divide consumer groups. In addition, manual judgment of the relevance of FMCG is inefficient and inaccurate. Summary of the invention

[0004] In view of this, the present invention provides a method and device for analyzing the association of fast-moving consumer goods based on consumer purchase data to solve the problem of efficiently and accurately analyzing the association of fast-moving consumer goods.

[0005] In a first aspect, the present invention provides a method for analyzing the association of fast-moving consumer goods based on consumer purchase data, the method comprising:

[0006] For any fast-moving consumer goods, obtain the product data of the fast-moving consumer goods and the consumption scanning data of multiple member consumers on the fast-moving consumer goods. The consumption scanning data includes the categories of fast-moving consumer goods.

[0007] Based on the consumption code scanning data and product data, the FMCG categories of all FMCGs purchased by multiple member consumers are clustered to obtain multiple FMCG clustering results. The multiple FMCG clustering results are used to represent the product associations between FMCGs belonging to the same FMCG brand.

[0008] Based on the consumption code scanning data, multiple member consumers are clustered to obtain multiple consumer clustering results, which include multiple consumers;

[0009] Based on the categories of fast-moving consumer goods purchased by multiple consumers, multiple association rules are determined, and the support and confidence corresponding to each association rule are calculated. The support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of multiple consumers.

[0010] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention covers the purchase behaviors of different consumers by collecting the product data of any fast-moving consumer goods and the consumption scanning data of the fast-moving consumer goods by member consumers, which is helpful to discover more extensive and diverse associations and provides a rich data source for subsequent association analysis. By clustering the fast-moving consumer goods belonging to the same fast-moving consumer goods brand, the product associations between fast-moving consumer goods under the same brand can be discovered. By clustering consumers, consumer groups can be accurately divided and consumer types can be subdivided, so as to deeply understand the purchase behaviors and preferences of different types of consumers, and multiple association rules are determined according to the multivariate consumers in the clustering results, and the support and confidence of each association rule are calculated, which can reflect the association between fast-moving consumer goods and consumer purchase behaviors. Through the two association analyses, it is helpful for enterprises to formulate marketing strategies and optimize product portfolios more accurately.

[0011] In an optional implementation, based on the consumption code scanning data and product data, the fast-moving consumer goods categories of all fast-moving consumer goods purchased by multiple member consumers are clustered to obtain multiple fast-moving consumer goods clustering results, including:

[0012] Determine multiple FMCG product features based on the product data of all FMCG categories purchased by multiple member consumers;

[0013] Based on multiple fast-moving consumer goods product features, the fast-moving consumer goods categories of all fast-moving consumer goods purchased by multiple member consumers are clustered to obtain multiple fast-moving consumer goods clustering results.

[0014] The method for analyzing the association of fast-moving consumer goods based on consumer purchase data provided in an embodiment of the present invention determines multiple fast-moving consumer goods product features according to product data, and clusters the fast-moving consumer goods categories based on the fast-moving consumer goods product features, thereby discovering the product associations between the fast-moving consumer goods corresponding to different fast-moving consumer goods categories.

[0015] In an optional implementation, multiple member consumers are clustered based on the consumption code scanning data to obtain multiple consumer clustering results, including:

[0016] Based on the consumption code scanning data, multiple consumption behavior characteristics are determined, and multiple consumption behavior characteristic values ​​corresponding to each member consumer are determined;

[0017] Using the elbow diagram analysis method, the target quantity is determined based on multiple consumer behavior characteristics and multiple consumer behavior characteristic values ​​corresponding to each member consumer;

[0018] Based on the consumption code scanning data and multiple consumption behavior characteristic values ​​corresponding to each member consumer, multiple member consumers are clustered to obtain the target number of consumer clustering results.

[0019] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention can more carefully portray the consumer's portrait by determining multiple consumer behavior characteristics based on consumer scanning data, and use the elbow diagram analysis method to determine the optimal number of clusters based on the consumer behavior characteristics to ensure the rationality and effectiveness of the clustering results. Consumers are then clustered to divide them into different sub-groups, each cluster representing a class of consumers with similar consumer behaviors, so that companies can formulate more accurate marketing strategies based on different consumer clustering results.

[0020] In an optional implementation, based on the consumption code scanning data and multiple consumption behavior feature values ​​corresponding to each member consumer, multiple member consumers are clustered to obtain a target number of consumer clustering results, including:

[0021] For any member consumer, based on multiple consumption behavior characteristics and the member consumer's consumption code scanning data, determine the consumption behavior characteristic values ​​corresponding to the multiple consumption behavior characteristics of the member consumer;

[0022] Get the membership matrix obtained in the previous iteration;

[0023] For any member consumer, based on the membership matrix obtained in the previous iteration, determine the membership of the member consumer to the target number of clusters in the previous iteration;

[0024] For any cluster, based on the membership of multiple member consumers to the cluster in the previous iteration and the consumption behavior characteristic values ​​of multiple member consumers, determine the cluster center of the current iteration of the cluster;

[0025] Based on the distance between each member consumer and the cluster center of each cluster in this iteration, the membership matrix obtained in the previous iteration is updated to obtain the membership matrix of this iteration;

[0026] Based on the membership matrix of this iteration and the consumption behavior characteristic values ​​of multiple member consumers, determine the objective function of this iteration. When the objective function of this iteration is greater than the preset threshold, repeat the above process of updating the membership matrix until the objective function of this iteration is less than the preset threshold, and obtain the target number of clusters and the member consumers in each cluster;

[0027] For any cluster, the number of all member consumers and the mean of the consumption behavior characteristic values ​​in the cluster are taken as the consumer clustering result corresponding to the cluster.

[0028] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention can more accurately quantify the purchase behavior of each consumer by determining the consumption behavior characteristic value of each member consumer, and ensure that each iteration can more accurately reflect the degree of belonging of each consumer to each clustering cluster by continuously updating the membership matrix, so that the clustering result is more stable and reliable, and set the objective function and preset threshold to ensure that the iteration is stopped after reaching a certain accuracy to improve efficiency. The center of each clustering cluster is calculated based on the consumption behavior characteristic values ​​of all consumers in the cluster, which can better represent the consumer characteristics in the cluster. The cluster center will be recalculated in each iteration to ensure that the cluster center always reflects the latest clustering results. The target number of clusters finally obtained represents different types of consumer groups, so that the consumers in each clustering cluster have similar consumption behavior characteristics, which is conducive to formulating more accurate marketing strategies.

[0029] In an optional implementation, based on the categories of fast-moving consumer goods purchased by multiple consumers, multiple association rules are determined, including:

[0030] Construct a correlation matrix based on the FMCG categories of all FMCGs purchased by multiplex consumers;

[0031] For any multiple-type consumer, all the types of fast-moving consumer goods purchased by the multiple-type consumer are taken as consumption records, and the total consumption records corresponding to all multiple-type consumers are obtained;

[0032] Based on the association matrix and the total consumption records, multiple association rules are determined.

[0033] The fast-moving consumer goods correlation analysis method based on consumer purchase data provided by the embodiment of the present invention constructs a correlation matrix to structurally represent the correlation information between the types of fast-moving consumer goods purchased by multiple consumers, and then takes all the types of fast-moving consumer goods purchased by each multiple consumer as a consumption record, which are summarized into a total consumption record. This makes it possible to mine frequently occurring fast-moving consumer goods combinations based on the correlation matrix and the total consumption records, thereby generating association rules, which is helpful to optimize marketing strategies and product combinations.

[0034] In an optional implementation, a correlation matrix is ​​constructed based on the categories of all fast-moving consumer goods purchased by multiple consumers, including:

[0035] Determine multiple FMCG related features based on product data corresponding to the FMCG categories purchased by diverse consumers;

[0036] For any FMCG combination among all FMCG categories purchased by the diversified consumer, determining association information of the FMCG combination based on the product data and multiple FMCG association features, where the FMCG combination includes at least two FMCG categories;

[0037] Based on the association information corresponding to all FMCG combinations, a correlation matrix is ​​constructed.

[0038] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention determines multiple fast-moving consumer goods association features by analyzing product data. For any fast-moving consumer goods combination, data containing multiple fast-moving consumer goods association features is determined from the product data, so as to determine the association information of the fast-moving consumer goods combination, thereby generating an association matrix, which is helpful to identify the association between fast-moving consumer goods combinations and provide support for determining association rules.

[0039] In an optional implementation, for any fast-moving consumer product, before obtaining product data of the fast-moving consumer product and consumption code scanning data of multiple member consumers on the fast-moving consumer product, the method further includes:

[0040] For any fast-moving consumer goods, obtain the initial product data of the fast-moving consumer goods;

[0041] Obtain initial consumption code scanning data of multiple member consumers on fast-moving consumer goods from multiple sales channels;

[0042] The initial product data and the initial consumer code scanning data are preprocessed respectively to obtain product data and consumer code scanning data.

[0043] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention ensures the integrity and accuracy of the data by obtaining the initial product data and initial consumption scanning data of any fast-moving consumer goods, and pre-processes the initial data to improve the data quality, thereby improving the accuracy of subsequent fast-moving consumer goods association analysis.

[0044] In a second aspect, the present invention provides a device for analyzing the association of fast-moving consumer goods based on consumer purchase data, the device comprising:

[0045] The first acquisition module is used to acquire, for any fast-moving consumer product, product data of the fast-moving consumer product and consumption code scanning data of the fast-moving consumer product by multiple member consumers, where the consumption code scanning data includes the category of the fast-moving consumer product;

[0046] A first clustering module is used to cluster the FMCG categories of all FMCGs purchased by multiple member consumers based on the consumption code scanning data and product data, and obtain multiple FMCG clustering results, wherein the multiple FMCG clustering results are used to represent the product associations between FMCGs belonging to the same FMCG brand;

[0047] The second clustering module is used to cluster multiple member consumers based on the consumption code scanning data to obtain multiple consumer clustering results, and the multiple consumer clustering results include multiple consumers;

[0048] The association analysis module is used to determine multiple association rules based on the categories of fast-moving consumer goods purchased by multiple consumers, and calculate the support and confidence corresponding to each association rule. The support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of multiple consumers.

[0049] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to execute the fast-moving consumer goods correlation analysis method based on consumer purchase data of the above-mentioned first aspect or any corresponding embodiment thereof.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for analyzing the association of fast-moving consumer goods based on consumer purchase data of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 is a flow chart of a method for analyzing the association of fast-moving consumer goods based on consumer purchase data according to an embodiment of the present invention;

[0053] Figure 2 is a schematic diagram of an elbow graph according to an embodiment of the present invention;

[0054] Figure 3 is a structural block diagram of a device for analyzing association of fast-moving consumer goods based on consumer purchase data according to an embodiment of the present invention;

[0055] Figure 4 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0057] The existing fast-moving consumer goods association analysis is limited by the scope and timeliness of data collection, and the data is often incomplete. In addition, due to the complex data sources, it is impossible to accurately divide consumer groups. In addition, when conducting specific analysis, manual judgment is performed, which is inefficient and inaccurate. The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention covers the purchase behavior of different consumers by collecting the product data of any fast-moving consumer goods and the consumption scanning data of the fast-moving consumer goods by member consumers, which helps to discover more extensive and diverse associations and provides a rich data source for subsequent association analysis. By clustering the categories of fast-moving consumer goods based on product data, the intrinsic associations between fast-moving consumer goods belonging to the same brand can be found. At the same time, multiple association rules are determined according to multiple consumers, and the support and confidence of each association rule are calculated, which can reflect the association between fast-moving consumer goods and consumer purchase behavior. Through two association analyses, it helps enterprises to formulate marketing strategies and optimize product portfolios more accurately.

[0058] According to an embodiment of the present invention, an embodiment of a method for analyzing the association of fast-moving consumer goods based on consumer purchase data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0059] In this embodiment, a method for analyzing the association of fast-moving consumer goods based on consumer purchase data is provided, which can be used in terminals such as computers, Figure 1 FIG. 1 is a flow chart of a method for analyzing the association of fast-moving consumer goods based on consumer purchase data according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0060] Step S101, for any fast-moving consumer goods, obtain product data of the fast-moving consumer goods and consumption code scanning data of multiple member consumers on the fast-moving consumer goods, and the consumption code scanning data includes the category of fast-moving consumer goods. Specifically, the embodiment of the present invention takes cigarettes as an example for explanation. For enterprises selling cigarettes, product data of all cigarettes sold by the enterprise itself can be obtained according to the type of cigarettes. The product data of any type includes data such as type, product specifications, and production process. In addition, the consumption code scanning data of consumers belonging to members is collected from all sales channels sold by the enterprise. The consumption code scanning data refers to the data generated by scanning a QR code or a barcode when a member consumer purchases cigarettes, including data such as sales channels, categories of fast-moving consumer goods, quantity of fast-moving consumer goods, brand of fast-moving consumer goods, and consumption time. Optionally, consumption code scanning data for ten days, thirty days, or three months can be obtained, and the embodiment of the present invention does not limit this. By obtaining the above data, a data basis is provided for subsequent correlation analysis, and when collecting consumption code scanning data, the consumption code scanning data of member consumers is automatically obtained from various consumption channels, providing a rich source of data, and is not limited by the scope and timeliness of collection.

[0061] Step S102, based on the consumption code scanning data and product data, cluster the FMCG categories of all FMCGs purchased by multiple member consumers to obtain multiple FMCG clustering results, and the multiple FMCG clustering results are used to represent the product associations between FMCGs belonging to the same FMCG brand. Specifically, since an enterprise generally only sells one brand of cigarettes, and a brand can include multiple products. Therefore, all cigarette types can be clustered according to the product data corresponding to all cigarette types purchased by all member consumers, and cigarettes with similar product data under the same cigarette brand can be clustered into one category, so that the product associations between FMCGs under the same brand can be discovered to locate similar FMCGs.

[0062] Step S103, based on the consumption code scanning data, multiple member consumers are clustered to obtain multiple consumer clustering results, and the multiple consumer clustering results include multiple consumers. Specifically, since different member consumers have different selection tendencies and preference characteristics for cigarettes, multiple consumer clustering results can be obtained by clustering all member consumers, and each clustering result represents a group of member consumers with similar purchasing behaviors. The member consumers in the multiple consumers have a large number of types and frequencies of fast-moving consumer goods purchased, indicating that they have a high willingness to explore fast-moving consumer goods and diversified consumption needs, but their loyalty to specific types of fast-moving consumer goods is relatively low. Therefore, the member consumers in the multiple consumers can be further analyzed to find the correlation between purchasing behavior and the types of fast-moving consumer goods.

[0063] Step S104, based on the FMCG categories of the FMCG purchased by the multiple consumers, multiple association rules are determined, and the support and confidence corresponding to each association rule are calculated. The support and confidence are used to represent the correlation between FMCGs through the purchasing behavior of multiple consumers. Specifically, since member consumers among multiple consumers usually purchase multiple types of cigarettes when purchasing cigarettes, the association rules can be determined by analyzing the consumption behavior of member consumers. The association rules represent the combination of FMCG categories purchased by member consumers together. By calculating the support of each association rule, the frequency of a combination of FMCG categories appearing in the consumption behavior of member consumers is represented. The higher the support, the more common the combination. At the same time, the confidence of the association rule is calculated to represent the probability of purchasing another type of FMCG at the same time when purchasing a certain type of FMCG. The higher the confidence, the stronger the correlation between the two types of FMCG.

[0064] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention covers the purchase behaviors of different consumers by collecting the product data of any fast-moving consumer goods and the consumption scanning data of the fast-moving consumer goods by member consumers, which is helpful to discover more extensive and diverse associations and provides a rich data source for subsequent association analysis. By clustering the fast-moving consumer goods belonging to the same fast-moving consumer goods brand, the product associations between fast-moving consumer goods under the same brand can be discovered. By clustering consumers, consumer groups can be accurately divided and consumer types can be subdivided, so as to deeply understand the purchase behaviors and preferences of different types of consumers, and multiple association rules are determined according to the multivariate consumers in the clustering results, and the support and confidence of each association rule are calculated, which can reflect the association between fast-moving consumer goods and consumer purchase behaviors. Through the two association analyses, it is helpful for enterprises to formulate marketing strategies and optimize product portfolios more accurately.

[0065] In this embodiment, a method for analyzing the association of fast-moving consumer goods based on consumer purchase data is provided, which can be used in the above-mentioned terminal, such as a computer, etc. The method specifically includes the following steps:

[0066] Step S201, for any fast-moving consumer product, initial product data of the fast-moving consumer product is obtained. Specifically, referring to the above step S101, initial product data of each type of fast-moving consumer product is obtained according to the type of the fast-moving consumer product.

[0067] Step S202, obtaining initial consumption code scanning data of multiple member consumers on fast-moving consumer goods from multiple sales channels. Specifically, refer to the above step S101 to obtain the initial consumption code scanning data.

[0068] Step S203, preprocess the initial product data and initial consumer scan code data respectively to obtain product data and consumer scan code data. Specifically, a variety of preprocessing operations can be used. The embodiment of the present invention takes the following preprocessing operations as an example for illustration: remove duplicate, invalid and erroneous data from the initial product data and initial consumer scan code data to ensure data accuracy; unify the data format to facilitate subsequent analysis; use mean filling, median filling or model prediction filling to ensure data integrity; identify and process values ​​that are significantly deviated from the normal range, optimize the data set by removal or correction, and improve data quality and reliability. By obtaining the initial product data and initial consumer scan code data of any fast-moving consumer product, the integrity and accuracy of the data are ensured, and the initial data is preprocessed to improve the data quality, so as to improve the accuracy of subsequent fast-moving consumer product association analysis.

[0069] Step S204: For any fast-moving consumer goods, obtain the product data of the fast-moving consumer goods and the consumption code scanning data of multiple member consumers on the fast-moving consumer goods, and the consumption code scanning data includes the category of the fast-moving consumer goods. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0070] Step S205, based on the consumption code scanning data and product data, cluster the FMCG categories of all FMCGs purchased by multiple member consumers to obtain multiple FMCG clustering results, and the multiple FMCG clustering results are used to represent the product associations between FMCGs belonging to the same FMCG brand.

[0071] Specifically, the above step S205 includes:

[0072] Step S2051, based on the product data corresponding to the FMCG categories of all FMCGs purchased by multiple member consumers, multiple FMCG product characteristics are determined. Specifically, based on the product data corresponding to different types of cigarettes, important parameters of the cigarettes can be determined as FMCG product characteristics, such as cigarette price, filter length, packaging specifications, tar content, smoke nicotine content, carbon monoxide content, and cigarette length.

[0073] Step S2052, based on multiple fast-moving consumer goods product features, cluster the fast-moving consumer goods categories of all fast-moving consumer goods purchased by multiple member consumers to obtain multiple fast-moving consumer goods clustering results. Specifically, clustering algorithms such as C-means (Fuzzy C-means Clustering), K-means (K-means Clustering), hierarchical clustering, etc. are used to cluster the fast-moving consumer goods purchased by all member consumers, and cluster cigarette types with similar product parameters under the same brand into one category to locate similar fast-moving consumer goods. Assume that after removing duplicates, there are 20 types of cigarettes purchased by all member consumers. By clustering the 20 types of cigarettes based on the above-mentioned 7 fast-moving consumer goods product features, 3 categories can be obtained. For the first category, the price is relatively low, the tar, carbon monoxide, and nicotine content of the smoke are relatively high, and it can be positioned as a cigarette to expand the market; for the second category, the price is high, the filter tip is long, the cigarette length is short, the tar, carbon monoxide, and nicotine content of the smoke are relatively low, and it can be positioned as a high-end brand cigarette; for the third category, the price is moderate, the tar, carbon monoxide, and nicotine content of the smoke are high, and it can be positioned as a popular main cigarette. The following table 1 is the clustering result of fast-moving consumer goods obtained by the above assumptions, which is only an example and is not limited.

[0074] Table 1

[0075]

[0076]

[0077] Step S206, clustering multiple member consumers based on the consumption code scanning data to obtain multiple consumer clustering results, and the multiple consumer clustering results include multiple consumers.

[0078] Specifically, the above step S206 includes:

[0079] Step S2061, based on the consumption code scanning data, multiple consumption behavior characteristics are determined, and multiple consumption behavior characteristic values ​​corresponding to each member consumer are determined. Specifically, in order to fully reflect the consumption characteristics of member consumers when purchasing fast-moving consumer goods, the following three consumption behavior characteristics can be selected: Total number of purchases: the total frequency of member consumers purchasing fast-moving consumer goods within a preset time period, which is used to measure the purchasing activity of member consumers; The proportion of fast-moving consumer goods purchased with the most times: by calculating the proportion of the fast-moving consumer goods purchased with the most times by member consumers to their total purchase times, it is used to evaluate the preference and loyalty of member consumers to specific fast-moving consumer goods categories; The number of fast-moving consumer goods purchased: the number of different fast-moving consumer goods purchased by member consumers within a preset time period is counted, which is used to reflect the purchase diversity and market exploration behavior of member consumers. After determining the above three consumption behavior characteristics, the consumption behavior characteristic values ​​of each member consumer are determined respectively. For example, the consumption behavior characteristic values ​​corresponding to member consumer A are: (total number of purchases: 100, the proportion of fast-moving consumer goods purchased with the most times: 0.6, the number of fast-moving consumer goods purchased: 10).

[0080] Step S2062, using the elbow diagram analysis method, based on multiple consumer behavior characteristics and multiple consumer behavior characteristic values ​​corresponding to each member consumer, determine the target number. Specifically, using clustering algorithms such as K-means, try different cluster numbers k, such as k=1, 2, 3, 4, etc. For each k, calculate the sum of the squares of the distances from all data points to the cluster center of the category to which they belong. Among them, the data point includes three consumer behavior characteristic values ​​corresponding to each member consumer and each member consumption. The cluster center can be obtained by randomly selecting k data points, and can also be set by itself or calculated in other ways. The embodiment of the present invention does not limit this. Figure 2 is a schematic diagram of an elbow diagram according to an embodiment of the present invention, such as Figure 2 As shown in Figure 1, the sum of squared distances corresponding to different k values ​​is plotted into a graph to form an elbow graph. The elbow graph usually shows an "elbow" (i.e., the inflection point of the curve). The k value at this inflection point is considered to be the optimal number of clusters, i.e., the target number, which is given by Figure 2 It can be seen that the target number is 3. Through the elbow diagram analysis method, the optimal number of clusters can be scientifically determined to better classify and analyze the consumption behavior of member consumers.

[0081] Step S2063: Cluster multiple member consumers based on the consumption code scanning data and multiple consumption behavior feature values ​​corresponding to each member consumer to obtain a target number of consumer clustering results.

[0082] In some optional implementations, the above step S2063 includes:

[0083] Step a1, obtain the membership matrix obtained in the last iteration. Specifically, the C-means clustering algorithm is used, which optimizes the membership of each data point to its cluster through an iterative optimization process, thereby achieving effective clustering of consumers.

[0084] Step a2: for any member consumer, based on the membership matrix obtained in the previous iteration, determine the membership of the member consumer to the target number of clusters in the previous iteration. Specifically, each row in the membership matrix represents a cluster, each column represents a data point, and the elements in the matrix represent the membership. Therefore, assuming that the target number is 3, for any member consumer, the membership matrix obtained in the previous iteration can be used to obtain the membership of the member consumer to the three clusters obtained in the previous iteration.

[0085] Step a3: for any cluster, based on the membership of multiple member consumers to the cluster in the previous iteration and the consumption behavior characteristic values ​​of multiple member consumers, determine the cluster center of the current iteration of the cluster. Specifically, for each cluster, based on the three consumption behavior characteristic values ​​of each member consumer and the membership of the cluster obtained in the previous iteration, the cluster center is updated by the following formula (1) to obtain the cluster center of the current iteration.

[0086]

[0087] Among them, μ i represents the cluster center of the i-th cluster; n represents the number of data points, that is, the number of member consumers; uij represents the membership of the j-th data point to the i-th cluster; m represents the fuzzy index; x j Represents the value of the j-th data point, that is, the three consumption behavior characteristic values ​​mentioned above.

[0088] Step a4, based on the distance between each member consumer and the cluster center of each cluster in this iteration, the membership matrix obtained in the previous iteration is updated to obtain the membership matrix of this iteration. Specifically, the distance between each member consumer and the three cluster centers obtained in this iteration is calculated, and the distance can be the Euclidean distance shown in the following formula (2), so as to update the membership matrix obtained in the previous iteration based on the distance through the following formula (3) to obtain the membership matrix of this iteration.

[0089]

[0090] in, represents the Euclidean distance between the jth data point and the cluster center of the i-th cluster at the k-th iteration; m represents the fuzzy index; x jl represents the lth consumption behavior characteristic value of the jth data point; Represents the lth consumption behavior characteristic value of the cluster center of the i-th cluster at the k-th iteration.

[0091]

[0092] in, represents the membership of the jth data point and the ith cluster obtained in this iteration, i.e. the k+1th iteration; h represents the iterative cluster index variable; c represents the number of clusters, i.e. the target number; m represents the fuzzy index; represents the Euclidean distance between the j-th data point and the cluster center of the i-th cluster at the k-th iteration; represents the Euclidean distance between the j-th data point and the cluster center of the h-th cluster at the k-th iteration.

[0093] Step a5, based on the membership matrix of this iteration and the consumption behavior characteristic values ​​of multiple member consumers, determine the objective function of this iteration. When the objective function of this iteration is greater than the preset threshold, repeat the above process of updating the membership matrix until the objective function of this iteration is less than the preset threshold, and obtain the target number of clusters and the member consumers in each cluster. Specifically, the objective function can be expressed by the following formula (4), the purpose of which is to minimize the sum of the weighted distances from all data points to the cluster center of the cluster to which they belong.

[0094]

[0095] Where J represents the objective function; c represents the number of clusters, that is, the target number. Figure 2 The target number shown is 3 for example; n represents the number of data points, that is, the number of member consumers; u ij represents the membership of the jth data point to the ith cluster; m represents the fuzzy index, with a value range of [1, +∞), determined by cross-validation; x j represents the value of the jth data point, i.e., the three consumption behavior characteristic values ​​mentioned above; μ i Represents the cluster center of the i-th cluster.

[0096] Based on the membership matrix of this iteration, the membership of each member consumer to each cluster in this iteration can be obtained, which is substituted into the above formula (4) to calculate the objective function of this iteration and compare it with the preset threshold. If it is greater than the preset threshold, return to step a1 to re-obtain the membership matrix of the previous iteration, and repeat the process of updating the membership matrix until the objective function obtained by a certain iteration is less than the preset threshold. At this time, clustering is completed, and 3 clusters and member consumers belonging to each cluster are obtained.

[0097] Step a6: For any cluster, the number of all member consumers and the mean of the consumption behavior characteristic values ​​in the cluster are used as the consumer clustering result corresponding to the cluster. Specifically, for each cluster, the number of member consumers belonging to the cluster is determined, and the mean of the consumption behavior characteristic values ​​of these member consumers, that is, the consumption behavior characteristic value of the cluster center of the cluster, is used as the consumption behavior characteristic value of the cluster, thereby obtaining the consumer clustering result. For the first category, the member consumers in this category purchase fewer times and types, but the value of the proportion of the most fast-moving consumer goods purchased in the consumption behavior characteristics is extremely high, indicating that most of them are individual customers who occasionally purchase, and their contribution and value to fast-moving consumer goods are relatively low. They can be positioned as low-value individual customers. For the second category, the member consumers in this category not only purchase more times, but also have a higher proportion of purchases of a specific type, showing strong loyalty and product preference. They are important supporters of fast-moving consumer goods and can be positioned as loyal consumers. For the third category, the member consumers in this category purchase more types and times, indicating a high willingness to explore and diversified consumption needs, but relatively low loyalty to specific types, and can be positioned as diversified consumers. Table 2 below shows the consumer clustering results obtained by the above assumptions, which is only an example and is not intended to be limiting.

[0098] Table 2

[0099]

[0100]

[0101] Step S207, based on the categories of fast-moving consumer goods purchased by multiple consumers, multiple association rules are determined, and the support and confidence corresponding to each association rule are calculated, where the support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of multiple consumers.

[0102] Specifically, step S207 determines a plurality of association rules based on the categories of fast-moving consumer goods purchased by the multiple consumers, including:

[0103] Step S2071, constructing a correlation matrix based on the categories of all fast-moving consumer goods purchased by the multiple consumers.

[0104] In some optional implementations, the above step S2071 includes:

[0105] Step b1, based on the product data corresponding to the types of fast-moving consumer goods purchased by multiple consumers, multiple fast-moving consumer goods related features are determined. Specifically, when determining the fast-moving consumer goods related features, the factors that affect the selection of different types of fast-moving consumer goods by multiple consumers are considered. For example, the three dimensions of product specifications, sales conditions, and production processes of cigarettes are used as three fast-moving consumer goods related features, so that the three aspects of information can be integrated and the selection tendency and preference characteristics of multiple consumers can be presented through the cross-correlation of each dimension.

[0106] Step b2, for any FMCG combination among all FMCG categories purchased by multiple consumers, determine the association information of all FMCG combinations based on product data and multiple FMCG association characteristics, where the FMCG combination includes at least two FMCG categories. Specifically, assume that all multiple consumers purchased a total of 4 types of cigarettes: cigarette type 1, cigarette type 5, cigarette type 7, and cigarette type 20, and randomly combine these 4 types to form multiple FMCG combinations. Assume that cigarette type 1 and cigarette type 5 constitute a FMCG combination, obtain data on the association characteristics of the above three FMCGs of cigarette type 1 and cigarette type 5 from the product data, such as product specifications, prices, production processes, and product target consumer groups, and the relevant staff of the cigarette company manually determine the association information between cigarette type 1 and cigarette type 5 based on the above data combined with their own business experience, such as what is the association between cigarette type 1 and cigarette type 5 in terms of production process, who influences whose price positioning, who promotes whose innovation and development, or who interacts with whom, etc.

[0107] Step b3: construct a correlation matrix based on the correlation information corresponding to all FMCG combinations. Specifically, the following Table 3 is a correlation matrix constructed based on the correlation information of the above-mentioned four types of cigarettes, which is only an example and not a limitation.

[0108] Table 3

[0109]

[0110]

[0111] Step S2072: For any multiplex consumer, all the types of fast-moving consumer goods purchased by the multiplex consumer are taken as consumption records, and the total consumption records corresponding to all multiplex consumers are obtained. Specifically, for the assumed multiplex consumer B, all the types of fast-moving consumer goods purchased by him are: cigarette type 1 and cigarette type 5; for multiplex consumer C, all the types of fast-moving consumer goods purchased by him are: cigarette type 1 and cigarette type 20; for multiplex consumer D, all the types of fast-moving consumer goods purchased by him are: cigarette type 1, cigarette type 5, cigarette type 7, and cigarette type 20. All the types of fast-moving consumer goods purchased by the above three multiplex consumers are taken as a consumption record respectively, so as to obtain the total consumption records of all multiplex consumers.

[0112] Step S2073, based on the association matrix and the total consumption records, multiple association rules are determined. Specifically, since multiple consumers are a group that purchases a certain fast-moving consumer product with a high overall frequency but has no obvious preference for specific types of fast-moving consumer products, the purpose is to discover the combination of fast-moving consumer products that multiple consumers often purchase together through association rule mining. Apply the association rule mining algorithm to determine the association rules from the fast-moving consumer product combination and the total consumption records displayed in the association matrix. Optionally, the process of applying the algorithm above is a prior art and will not be repeated here.

[0113] In some optional implementations, assuming that any association rule X is (A, B)→C, it contains two item sets A and B. Support and confidence are two commonly used indicators in association rule analysis, which are used to measure the degree of association between item sets. Support can be calculated by the following formula (5), which indicates the popularity of any item set in the total consumption records. Confidence can be calculated by the following formula (6), which indicates the possibility of another item set appearing when one item set appears. For example, the confidence of (A, B)→C is 0.5, which means that among the consumption records containing "A", 50% of the consumption records also contain "B".

[0114]

[0115] Among them, X represents the association rule; Support(X) represents the support; q ABC It represents the consumption records including item sets A, B and C at the same time; Q represents the total consumption records.

[0116]

[0117] Among them, X represents the association rule; Confidence(X) represents the confidence; q ABC represents the consumption records including item sets A, B and C at the same time; q AB Indicates consumption records that include both item set A and item set B.

[0118] Optionally, when the support and confidence of each association rule are calculated using the above formula, it is assumed that the support threshold is 2% and the confidence threshold is 60%. For all association rules, the association rules that are greater than the support threshold and the confidence threshold are regarded as valid association rules, which serve as the basis for enterprises to formulate marketing strategies and optimize product portfolios. Through association rule analysis, it helps to understand consumers' purchasing behavior, discover the correlation between the purchase of fast-moving consumer goods, and mine hidden patterns and laws from large-scale data to provide support and guidance for decision-making, thereby improving efficiency, reducing costs, and optimizing resource allocation. The following Table 4 shows some valid association rules for the four types of cigarettes based on the above assumptions, as well as their support and confidence. It is only an example and is not limited to this.

[0119] Table 4

[0120]

[0121]

[0122] The fast-moving consumer goods association analysis method based on consumer purchase data provided by the embodiment of the present invention covers the purchase behaviors of different consumers by collecting the product data of any fast-moving consumer goods and the consumption scanning data of the fast-moving consumer goods by member consumers, which is helpful to discover more extensive and diverse associations and provides a rich data source for subsequent association analysis. By clustering the fast-moving consumer goods belonging to the same fast-moving consumer goods brand, the product associations between fast-moving consumer goods under the same brand can be discovered. By clustering consumers, consumer groups can be accurately divided and consumer types can be subdivided, so as to deeply understand the purchase behaviors and preferences of different types of consumers, and multiple association rules are determined according to the multivariate consumers in the clustering results, and the support and confidence of each association rule are calculated, which can reflect the association between fast-moving consumer goods and consumer purchase behaviors. Through the two association analyses, it is helpful for enterprises to formulate marketing strategies and optimize product portfolios more accurately.

[0123] In this embodiment, a device for analyzing the association of fast-moving consumer goods based on consumer purchase data is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0124] This embodiment provides a fast-moving consumer goods correlation analysis device based on consumer purchase data, such as Figure 3 As shown, including:

[0125] The first acquisition module 301 is used to acquire, for any fast-moving consumer product, product data of the fast-moving consumer product and consumption scan code data of the fast-moving consumer product by multiple member consumers, where the consumption scan code data includes the category of the fast-moving consumer product.

[0126] The first clustering module 302 is used to cluster the FMCG categories of all FMCGs purchased by multiple member consumers based on the consumption code scanning data and product data, and obtain multiple FMCG clustering results. The multiple FMCG clustering results are used to represent the product associations between FMCGs belonging to the same FMCG brand.

[0127] The second clustering module 303 is used to cluster multiple member consumers based on the consumption code scanning data to obtain multiple consumer clustering results, and the multiple consumer clustering results include multiple consumers.

[0128] The association analysis module 304 is used to determine multiple association rules based on the categories of fast-moving consumer goods purchased by multiple consumers, and calculate the support and confidence corresponding to each association rule. The support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of multiple consumers.

[0129] In some optional implementations, the first clustering module 302 includes:

[0130] The first determining unit is used to determine a plurality of fast-moving consumer goods product features based on product data corresponding to the fast-moving consumer goods categories of all fast-moving consumer goods purchased by a plurality of member consumers.

[0131] The first clustering unit is used to cluster the fast-moving consumer goods categories of all fast-moving consumer goods purchased by multiple member consumers based on multiple fast-moving consumer goods product features to obtain multiple fast-moving consumer goods clustering results.

[0132] In some optional implementations, the second clustering module 303 includes:

[0133] The second determination unit is used to determine multiple consumption behavior characteristics based on the consumption code scanning data, and determine multiple consumption behavior characteristic values ​​corresponding to each member consumer.

[0134] The analysis unit is used to determine the target quantity based on multiple consumption behavior characteristics and multiple consumption behavior characteristic values ​​corresponding to each member consumer by using an elbow diagram analysis method.

[0135] The second clustering unit is used to cluster multiple member consumers based on the consumption code scanning data and multiple consumption behavior feature values ​​corresponding to each member consumer, and obtain a target number of consumer clustering results.

[0136] In some optional embodiments, the second clustering unit includes:

[0137] Get subunit, used to get the membership matrix obtained in the previous iteration.

[0138] The first determination subunit is used to determine, for any member consumer, the membership of the member consumer to the target number of clusters in the previous iteration based on the membership matrix obtained in the previous iteration.

[0139] The second determination subunit is used to determine, for any cluster, the cluster center of the current iteration of the cluster based on the membership of multiple member consumers to the cluster in the previous iteration and the consumption behavior characteristic values ​​of multiple member consumers.

[0140] The updating subunit is used to update the membership matrix obtained in the previous iteration based on the distance between each member consumer and the cluster center of each cluster cluster in this iteration, so as to obtain the membership matrix of this iteration.

[0141] The third determination subunit is used to determine the objective function of this iteration based on the membership matrix of this iteration and the consumption behavior characteristic values ​​of multiple member consumers. When the objective function of this iteration is greater than a preset threshold, the above process of updating the membership matrix is ​​repeated until the objective function of this iteration is less than the preset threshold, thereby obtaining a target number of clusters and the member consumers in each cluster.

[0142] The fourth determination subunit is used to, for any cluster, take the number of all member consumers and the average of the consumption behavior characteristic values ​​in the cluster as the consumer clustering result corresponding to the cluster.

[0143] In some optional implementations, the association analysis module 304 includes:

[0144] A building block is used to construct a correlation matrix based on the FMCG categories of all FMCGs purchased by multiple consumers.

[0145] The third determining unit is used to, for any multiple-type consumer, take all types of fast-moving consumer goods purchased by the multiple-type consumer as consumption records, and obtain the total consumption records corresponding to all the multiple-type consumers.

[0146] The fourth determining unit is used to determine a plurality of association rules based on the association matrix and the total consumption records.

[0147] In some optional embodiments, the building block includes:

[0148] The fifth determination subunit is used to determine a plurality of fast-moving consumer goods related features based on product data corresponding to the types of fast-moving consumer goods purchased by the diverse consumers.

[0149] The sixth determination subunit is used to determine, from the product data, association information containing multiple FMCG association features for any FMCG combination among all FMCG categories purchased by the diversified consumer, where the FMCG combination includes at least two FMCG categories.

[0150] The sub-unit is constructed to construct the association matrix based on the association information corresponding to all FMCG combinations.

[0151] In some optional implementations, before the first acquisition module 301, the device further includes:

[0152] The second acquisition module is used to acquire initial product data of any fast-moving consumer product.

[0153] The third acquisition module is used to obtain the initial consumption scanning code data of multiple member consumers on fast-moving consumer goods from multiple sales channels.

[0154] The preprocessing module is used to preprocess the initial product data and the initial consumer scan code data respectively to obtain product data and consumer scan code data.

[0155] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0156] The fast-moving consumer goods correlation analysis device based on consumer purchase data in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0157] The embodiment of the present invention also provides a computer device having the above Figure 3 The device shown is a fast-moving consumer goods correlation analysis device based on consumer purchase data.

[0158] See also Figure 4 , Figure 4 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 4As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 4 A processor 10 is taken as an example.

[0159] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0160] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0161] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0162] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0163] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0164] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0165] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0166] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0167] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for analyzing the association of fast-moving consumer goods based on consumer purchase data, characterized in that: The method comprises: For any fast-moving consumer product, obtain product data of the fast-moving consumer product and consumption code scanning data of the fast-moving consumer product by multiple member consumers, wherein the consumption code scanning data includes the category of the fast-moving consumer product; Based on the consumption code scanning data and the product data, clustering the FMCG categories of all FMCGs purchased by the multiple member consumers to obtain multiple FMCG clustering results, wherein the multiple FMCG clustering results are used to represent product associations between FMCGs belonging to the same FMCG brand; Based on the consumption code scanning data, clustering the multiple member consumers to obtain multiple consumer clustering results, wherein the multiple consumer clustering results include multiple consumers; Based on the categories of fast-moving consumer goods purchased by the diverse consumers, multiple association rules are determined, and the support and confidence corresponding to each association rule are calculated. The support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of the diverse consumers.

2. The method according to claim 1, characterized in that The method of clustering the fast-moving consumer goods categories of all fast-moving consumer goods purchased by the multiple member consumers based on the consumption code scanning data and the product data to obtain multiple fast-moving consumer goods clustering results includes: Determine multiple fast-moving consumer goods product features based on the product data corresponding to the fast-moving consumer goods categories of all fast-moving consumer goods purchased by the multiple member consumers; Based on the multiple fast-moving consumer goods product features, the fast-moving consumer goods categories of all the fast-moving consumer goods purchased by the multiple member consumers are clustered to obtain the multiple fast-moving consumer goods clustering results.

3. The method according to claim 1, characterized in that The clustering of the plurality of member consumers based on the consumption code scanning data to obtain a plurality of consumer clustering results includes: Based on the consumption code scanning data, multiple consumption behavior characteristics are determined, and multiple consumption behavior characteristic values ​​corresponding to each member consumer are determined; Using an elbow diagram analysis method, based on the multiple consumer behavior characteristics and multiple consumer behavior characteristic values ​​corresponding to each member consumer, determine the target quantity; Based on the consumption code scanning data and multiple consumption behavior feature values ​​corresponding to each member consumer, the multiple member consumers are clustered to obtain a target number of consumer clustering results.

4. The method according to claim 3, characterized in that The clustering of the plurality of member consumers based on the consumption code scanning data to obtain a target number of consumer clustering results includes: Get the membership matrix obtained in the previous iteration; For any member consumer, based on the membership matrix obtained in the previous iteration, determine the membership of the member consumer to the target number of clusters in the previous iteration; For any cluster, based on the membership of the multiple member consumers to the cluster in the previous iteration and the consumption behavior characteristic values ​​of the multiple member consumers, determine the cluster center of the current iteration of the cluster; Based on the distance between each member consumer and the cluster center of each cluster in this iteration, the membership matrix obtained in the previous iteration is updated to obtain the membership matrix of this iteration; Based on the membership matrix of this iteration and the consumption behavior characteristic values ​​of the plurality of member consumers, the objective function of this iteration is determined. When the objective function of this iteration is greater than a preset threshold, the above-mentioned process of updating the membership matrix is ​​repeated until the objective function of this iteration is less than the preset threshold, and the target number of clusters and the member consumers in each cluster are obtained; For any cluster, the number of all member consumers and the average of the consumption behavior characteristic values ​​in the cluster are taken as the consumer clustering result corresponding to the cluster.

5. The method according to claim 1, characterized in that The determining of multiple association rules based on the categories of fast-moving consumer goods purchased by the multiple consumers includes: Constructing a correlation matrix based on the categories of all fast-moving consumer goods purchased by the multiple consumers; For any multiple-type consumer, all the types of fast-moving consumer goods purchased by the multiple-type consumer are taken as consumption records to obtain the total consumption records corresponding to all the multiple-type consumers; The plurality of association rules are determined based on the association matrix and the total consumption records.

6. The method according to claim 5, characterized in that The construction of a correlation matrix based on the categories of all fast-moving consumer goods purchased by the multiple consumers includes: Determine multiple fast-moving consumer goods related features based on product data corresponding to the categories of fast-moving consumer goods purchased by the multiple consumers; For any FMCG combination among all FMCG categories purchased by the multiple consumer, determining association information of the FMCG combination based on the product data and the multiple FMCG association features, wherein the FMCG combination includes at least two FMCG categories; The association matrix is ​​constructed based on the association information corresponding to all fast-moving consumer goods combinations.

7. The method according to claim 1, characterized in that Before obtaining the product data of any fast-moving consumer product and the consumption code scanning data of the fast-moving consumer product by multiple member consumers, the method further includes: For any fast-moving consumer product, obtaining initial product data of the fast-moving consumer product; Acquire initial consumption code scanning data of the fast-moving consumer goods by the multiple member consumers from multiple sales channels; The initial product data and the initial consumption code scanning data are preprocessed respectively to obtain the product data and the consumption code scanning data.

8. A device for analyzing the association of fast-moving consumer goods based on consumer purchase data, characterized in that: The device comprises: A first acquisition module is used to acquire, for any fast-moving consumer product, product data of the fast-moving consumer product and consumption code scanning data of the fast-moving consumer product by multiple member consumers, wherein the consumption code scanning data includes the category of the fast-moving consumer product; A first clustering module is used to cluster the fast-moving consumer goods categories of all fast-moving consumer goods purchased by the multiple member consumers based on the consumption code scanning data and the product data to obtain multiple fast-moving consumer goods clustering results, wherein the multiple fast-moving consumer goods clustering results are used to represent product associations between fast-moving consumer goods belonging to the same fast-moving consumer goods brand; A second clustering module is used to cluster the multiple member consumers based on the consumption code scanning data to obtain multiple consumer clustering results, wherein the multiple consumer clustering results include multiple consumers; The association analysis module is used to determine multiple association rules based on the categories of fast-moving consumer goods purchased by the multiple consumers, and calculate the support and confidence corresponding to each association rule, wherein the support and confidence are used to represent the association between fast-moving consumer goods through the purchasing behavior of the multiple consumers.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the fast-moving consumer goods association analysis method based on consumer purchase data according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the fast-moving consumer goods association analysis method based on consumer purchase data according to any one of claims 1 to 7.

Images