A method for analyzing securities customers, a storage medium, and a computer device

Abstract

The present application discloses a method for analyzing securities customers, a storage medium, and a computer device. The method for analyzing securities customers based on the RFM model includes: obtaining the asset data and transaction data of securities customers; constructing a preference score matrix between users and products based on the asset data and transaction data; constructing a preference propagation network based on the preference score matrix between users and products; calculating a target probability based on the preference value output by the preference propagation network and the extended RFM model; the extended RFM model introduces three factors, namely transaction heat, asset score, and return score, into the asset matching parameters. Through the above method, the present application can accurately analyze the interaction behavior between securities users and products.

Description

Technical Field

[0001] This application relates to the field of data analysis, and particularly to a method for analyzing securities customers, a storage medium, and a computer device. Background Art

[0002] In the securities industry, customer segmentation and value assessment are key steps in achieving precise marketing and service optimization. Although the traditional RFM (Recency, Frequency, Monetary) model can initially classify customers, its limitation is that it fails to fully consider customers' dynamic trading behaviors and asset allocation preferences. Therefore, a more comprehensive method is needed to improve the accuracy and practicality of customer segmentation and value assessment. Summary of the Invention

[0003] This application mainly provides a method for analyzing securities customers, a storage medium, and a computer device to solve the problem that the RFM model is not accurate enough in analyzing securities customers.

[0004] To solve the above technical problems, a technical solution adopted by this application is: to provide a method for analyzing securities customers based on the RFM model, including: obtaining asset data and trading data of securities customers; constructing a preference score matrix between users and products based on the asset data and the trading data; constructing a preference propagation network based on the preference score matrix between users and products; calculating a target probability based on the preference value output by the preference propagation network and the extended RFM model; and the extended RFM model introduces three factors, namely trading heat, asset score, and return score, into the asset matching parameters.

[0005] In some embodiments, constructing the preference score matrix between users and products based on the asset data and trading data includes: constructing an association matrix between users and products based on the asset data and the trading data; calculating the preference similarity between users based on the association matrix between users and products; and calculating the preference degree of users for products according to the preference similarity between users and the actual trading behaviors of users, and generating a preference score matrix.

[0006] In some embodiments, the formula for calculating the preference degree is:

[0007]

[0008] where the preference degree (u i , p k ) represents the preference degree of user u i for product p k , and the preference similarity (u i , u j ) represents the preference similarity between user u iThe preference similarity with user u j The transaction amount between (u j , p k ) represents the transaction amount of user u i for product p k .

[0009] In some embodiments, combining the preference similarity and the actual transaction behavior of the user to calculate the preference degree of the user for the product further includes: obtaining a preset preference weight; the preference weight is set according to the analysis results of business requirements, the asset data, and the transaction data; adjusting the preference degree of the user for the product according to the preference weight to generate a weighted preference score matrix.

[0010] In some embodiments, constructing a preference propagation network based on the preference score matrix between the user and the product includes: initializing the network structure according to the association matrix between the user and the product and the preference similarity between the users; setting an initial preference value for each user node in the network structure; defining rules for the propagation of preferences in the network structure based on the initial preference value to form a preference propagation network.

[0011] In some embodiments, defining rules for the propagation of preferences in the network structure based on the initial preference value to form a preference propagation network includes: updating the preference value of the user for the product through iterative calculation; stopping the iterative calculation in response to the preference value satisfying a preset termination condition.

[0012] In some embodiments, the formula for the iterative calculation is:

[0013]

[0014] where the new preference value (u i ) represents the new preference value used to update the initial preference value of user u i , α is to control the relative importance of the initial preference value and the propagated preference, N(u i ) represents the set of all user nodes adjacent to user u i , N(u j ) represents the set of all user nodes adjacent to user u j , ω(u i , u j ) represents the edge weight between user u i and user u j , ω(u j , u k ) represents the edge weight between user u j and user u k ; where α is greater than 0 and less than 1.

[0015] In some embodiments, calculating the target probability based on the preference value output by the preference propagation network and the extended RFM model includes: constructing a probability model for each user to select each product according to the preference value of each user node in the preference propagation network, so as to generate a selection probability matrix for each user to select each product; integrating the selection probability matrix and the parameters of the extended RFM model to perform hierarchical clustering to assign cluster labels to each user; analyzing the characteristics of each cluster in the hierarchical clustering to identify typical user behavior patterns..

[0016] To solve the above technical problems, another technical solution adopted by this application is: to provide a storage medium, on which program data is stored, and when the program data is executed by a processor, the steps of the above-mentioned securities customer analysis method are implemented.

[0017] This application also provides a computer device, including a processor and a memory connected to each other, the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned securities customer analysis method are implemented.

[0018] The beneficial effects of this application are: different from the prior art, this application discloses a securities customer analysis method, a storage medium and a computer device. By obtaining the asset data and transaction data of securities customers; constructing a preference score matrix between users and products based on the asset data and transaction data to reflect the transaction situation of users for products; constructing a preference propagation network based on the preference score matrix between users and products to simulate the preference propagation process between users, so as to more accurately predict the potential interest of users in un-purchased products; calculating the target probability based on the preference value output by the preference propagation network and the extended RFM model; the extended RFM model introduces three factors, namely transaction heat, asset score and return score, into the asset matching parameters to convert the preferences of users into specific decisions, so as to dynamically adjust the clustering method according to the visualization results and business requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings, where:

[0020] Figure 1 is a schematic flowchart of an embodiment of the securities customer analysis method provided by this application;

[0021] Figure 2 is as Figure 1Flow diagram of an embodiment of method step 200 shown;

[0022] Figure 3 is as shown in Figure 2 Flow diagram of an embodiment of method step 230 shown;

[0023] Figure 4 is as shown in Figure 1 Flow diagram of an embodiment of method step 300 shown;

[0024] Figure 5 is as shown in Figure 4 Flow diagram of an embodiment of method step 330 shown;

[0025] Figure 6 is as shown in Figure 1 Flow diagram of an embodiment of method step 100 shown;

[0026] Figure 7 Structural diagram of an embodiment of the storage medium provided by this application;

[0027] Figure 8 Structural diagram of an embodiment of the computer device provided by this application. Detailed implementation manners

[0028] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0029] The terms "first", "second", and "third" in the embodiments of this application are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", and "third" may explicitly or implicitly include at least one of such features. In the description of this application, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically and clearly defined. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but optionally further includes unlisted steps or units, or optionally further includes other steps or units inherent to these processes, methods, products, or devices.

[0030] References herein to "embodiments" mean that particular features, structures, or characteristics described in connection with the embodiments can be included in at least one embodiment of the present application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0031] Refer to Figure 1 , Figure 1 which is a schematic flowchart of an embodiment of the securities customer analysis method provided by the present application. The securities customer analysis method based on the RFM model includes:

[0032] 100: Obtain the asset data and transaction data of securities customers.

[0033] Collecting and organizing the asset data of securities customers includes historical asset allocation data, such as the holding ratio and trading frequency of various types of assets. Obtaining the transaction data of securities users includes transaction behavior records, such as the trading behavior of customers, such as buy, sell, and holding time records.

[0034] Optionally, perform data cleaning and preprocessing on the obtained asset data and transaction data to ensure data integrity and consistency.

[0035] 200: Construct a preference score matrix between users and products based on the asset data and transaction data.

[0036] Construct a preference score matrix between users and products based on the asset data and transaction data. The preference score matrix records the preference scores of users for products and can reflect the trading tendency of users for products.

[0037] Further, refer to Figure 2 , 200 includes:

[0038] 210: Construct an association matrix between users and products based on the asset data and transaction data.

[0039] Aggregate the collected transaction data, summarize the transaction amounts by customer and product as labels, and generate an association matrix between users and products, for example, the following table:

[0040]

[0041] In the association matrix between users and products, each row represents a user, such as u1, u2, u3; each column represents a product, such as p1, p2, p3, p4; the values in the matrix represent the total transaction amounts of users for the products, such as user 1 consumes 10,000 for product 1, and user 2 consumes 15,000 for product 3, etc.

[0042] 220: Calculate the preference similarity between users based on the association matrix between users and products.

[0043] Specifically, the preference similarity between users is calculated by the following formula:

[0044]

[0045] Among them, the preference similarity (u i , u j ) represents the preference similarity between user ui and user uj; Ai and Aj respectively represent the preference vectors of user ui and user uj, that is, the vectors composed of the transaction amounts of user ui and user uj on all products.

[0046] Among them, Ai·Aj can be calculated by the following formula:

[0047]

[0048] Among them, A i，p and A j，p respectively represent the transaction amounts of user u i and user u j on product p; A i ·A j means that for each product p, multiply the transaction amounts of user u i and user u j on this product, and add up the results of all products.

[0049] Among them, ||Ai|| and ||Aj|| respectively represent the Euclidean norms of vector Ai and vector Aj, and the calculation method is as follows:

[0050] and .

[0051] The vectors Ai and Aj represent the preference vectors for each user. Square the transaction amounts on each product, sum them up, and then take the square root to obtain the Euclidean norm of the vector.

[0052] The similarity score between user ui and user uj is obtained through the above calculation. The range of the similarity score is [-1, 1], where 1 means exactly the same and 0 means no correlation.

[0053] For example, there are two users u1 and u3, and three products p1, p2, p3, p4. The transaction amounts of users u1 and u3 on these three products are as follows:

[0054] User u1: [A 1，1 , A 1，2 , A 1，3 , A1，4 =[10000, 5000, 0, 0],

[0055] User u3: [A 3，1 , A 3，2 , A 3，3 , A 3，4 =[20000, 0, 0, 0].

[0056] Calculate the preference similarity between user u1 and u3 as follows:

[0057] First, calculate the dot product:

[0058] ;

[0059] Calculate the Euclidean norm of each vector based on the dot product:

[0060] ,

[0061] ;

[0062] Calculate the preference similarity between u1 and u3:

[0063] .

[0064] 230: Calculate the preference degree of users for products based on the preference similarity between users and their actual transaction behaviors, and generate a preference score matrix.

[0065] Specifically, the formula for calculating the preference degree is:

[0066]

[0067] Among them, the preference degree (u i , p k ) represents the preference degree of user u i for product p k , the preference similarity (u i , u j ) represents the preference similarity between user u i and user u j , and the transaction amount (u j , p k ) represents the transaction amount of user u i for product p k .

[0068] Calculate the preference degree of users for products through matrix multiplication to generate a preference score matrix. Each row in the preference score matrix represents a user, each column represents a product, and the value in the preference score matrix represents the preference degree of the user for the product.

[0069] Optionally, refer to Figure 3 , 230 also includes:

[0070] 231: Obtain preset preference weights. The preference weights are set according to the analysis results of business requirements, asset data, and transaction data.

[0071] Assign weights to the user's preference degree for the product to ensure that the importance of different factors is reasonably reflected.

[0072] According to business requirements and data analysis results, assign appropriate weights to each factor.

[0073] For example, set the weights as follows: user - to - user preference similarity weight: 0.7, user actual transaction amount weight: 0.3.

[0074] 232: Adjust the user's preference degree for the product according to the preference weights to generate a weighted preference score matrix.

[0075] For example, when the user - to - user preference similarity weight is 0.7 and the user actual transaction amount weight is 0.3, calculate the user's weighted preference degree for the product through the following formula:

[0076] .

[0077] According to the set weights, adjust the user's preference degree for the product to generate a weighted preference score matrix. Each row of the weighted preference score matrix represents a user, each column represents a product, and the value in the matrix represents the user's weighted preference degree for the product.

[0078] 300: Construct a preference propagation network based on the preference score matrix between users and products.

[0079] The preference propagation network is used to simulate the preference propagation process between users to more accurately predict the user's potential interest in un - purchased products.

[0080] The construction of the preference propagation network depends on the similarity between users and the known interaction information between users and products.

[0081] Further, refer to Figure 4 , 300 includes:

[0082] 310: Initialize the network structure according to the association matrix between users and products and the preference similarity between users.

[0083] Use the networkx library to create a network, where nodes represent users or products, and edges represent the strength of the relationship between users and products, as well as between users and users.

[0084] Initialize the network structure based on the association matrix between users and products and the user similarity matrix. Each node in the network represents a user or a product, and the weight of the edge represents the transaction amount between the user and the product, or the similarity score between users.

[0085] 320: Set an initial preference value for each user node in the network structure.

[0086] Operate on the node attributes of the networkx graph object, traverse the association matrix between users and products, calculate the total transaction amount of each user as its initial preference value, and store it in the network node attributes. Each user node now has an initial preference value, which reflects the degree of product preference currently known to the user.

[0087] Specifically, the initial preference value is calculated by the following formula:

[0088] .

[0089] 330: Define the rules for the propagation of preferences in the network structure to form a preference propagation network.

[0090] Define the rules for the propagation of preferences on the network, which determine how preferences are passed from one node to another.

[0091] Furthermore, referring to Figure 5 , 330 also includes:

[0092] 331: Update the preference value of the user for the product through iterative calculation.

[0093] Specifically, the formula for iterative calculation is:

[0094]

[0095] where the new preference value (u i ) represents the new preference value used to update the initial preference value of user u i , α is to control the relative importance of the initial preference value and the propagated preference, N(u i ) represents the set of all user nodes adjacent to user u i , N(u j ) represents the set of all user nodes adjacent to user u j , ω(u i , u j ) represents the edge weight between user u i and user u j , ω(u j , u k ) represents the edge weight between user u j and user u kThe edge weights between...

[0096] Where α is greater than 0 and less than 1.

[0097] Update the preference values of each user node through the above formula, taking into account both the initial preferences and the preference propagation from neighbor nodes. After the preference values are updated, they can better reflect the users' interests in products they have not encountered.

[0098] 332: In response to the preference values satisfying a preset termination condition, stop the iterative calculation.

[0099] Update the preference parameters of users for products through iterative calculation, which is used to simulate the propagation of users' preferences in the network until the preference values converge to a stable state. Set the iterative termination condition to ensure that the algorithm stops when the preference values converge or reach the maximum number of iterations.

[0100] Set an initial preference value for each user node and make a copy as the old preference value for comparison. Iteratively update the preference values of each user node according to the preference propagation rule until the termination condition is met.

[0101] Furthermore, update the preference values of each user node according to the preference propagation formula and check for convergence. If the preference values do not converge, prepare the data for the next iteration; otherwise, terminate the iteration.

[0102] Furthermore, perform the preference propagation update within the maximum number of iterations until the preference values converge. Finally, obtain the stable user preference values.

[0103] Specifically, calculate the difference between the new preference value and the old preference value, and stop the iterative calculation after this difference satisfies a preset stop condition.

[0104] The formula for judging iterative convergence is:

[0105]

[0106] Where ε is a small positive number representing the maximum allowable error range; if the preference changes of all users are less than ε, it is considered that the preference values have converged.

[0107] Traverse each node in the network and judge whether the preference value change of each node in the network conforms to a preset range; in response to the preference value change of each node in the network conforming to the preset range, return a boolean value indicating that the preference values have converged.

[0108] Or, in response to the preference value change of a node in the network not conforming to the preset range, return a boolean value indicating that the preference values have not converged.

[0109] 400: Calculate the target probability based on the preference value output by the preference propagation network and the extended RFM model. The extended RFM model introduces three factors, namely trading heat, asset score, and return score, into the asset matching parameters.

[0110] Based on the preference parameters finally obtained in the preference propagation network, calculate the probability that a user selects a specific product. This step is a key link in transforming the user's preferences into specific decisions.

[0111] Specifically, the calculation methods of the extended parameters include:

[0112] Asset score = Ʃ (the amount of this type of asset ÷ the total amount of assets × the weight of this type of asset).

[0113] Trading heat score = Ʃ (the number of transactions on the current day ÷ the total number of transactions × the weight of the current day's transactions).

[0114] Return score = Ʃ (profit and loss ÷ total profit and loss × the weight of profit and loss).

[0115] Further, refer to Figure 6 , 440 includes:

[0116] 441: According to the preference values of each user node in the preference propagation network, construct a probability model for each user to select each product, so as to generate a selection probability matrix for each user to select each product.

[0117] To ensure that the preference scores are within a reasonable range, use StandardScaler to standardize the preference scores. This step helps to prevent certain features from dominating the clustering results due to their large dimensions.

[0118] Model the probability of each user selecting each product, and use the softmax function to ensure that the sum of probabilities is 1 and can reflect the relative preferences between different products.

[0119] For example, construct the selection probability matrix of users for products as follows:

[0120]

[0121] Among them, each row corresponds to customers u1, u2, u3 respectively, and each column corresponds to products p1, p2, p3, p4 respectively. Each parameter represents the selection probability of the user for the product. For example, the selection probability of user 1 for product 2 is 0.1112, and the selection probability of user 3 for product 4 is 0.0471.

[0122] 442: Integrate the selection probability matrix and the parameters of the extended RFM model to perform hierarchical clustering to assign cluster labels to each user.

[0123] Perform hierarchical clustering using the scipy.cluster.hierarchy.linkage function, and select the ward method; iterate through different numbers of clusters from the user-defined minimum number of clusters (min_clusters) to the user-defined maximum number of clusters (max_clusters), and calculate the Silhouette Score for each number of clusters. Calculate the Silhouette Score and select the optimal number of clusters. Re-perform hierarchical clustering using the optimal number of clusters and assign the final cluster labels to each user.

[0124] Specifically, the Silhouette Score is a metric for evaluating the clustering effect, ranging from -1 to 1, and the closer the value is to 1, the better the clustering. The Silhouette Score function is used to calculate the scores for each number of clusters.

[0125] Record the highest Silhouette Score and the corresponding number of clusters in each iteration, and finally select the number of clusters with the highest score as the optimal number of clusters.

[0126] Among them, the selection probability matrix is derived from the preference propagation network, indicating the selection probability of each user for different products. The parameters of the extended RFM model include indicators such as the recency of the last purchase (Recency), the frequency of purchase (Frequency), the amount of purchase (Monetary), etc., as well as extended transaction heat, asset score, return score, and customer comprehensive score parameters.

[0127] 443: Analyze the characteristics of each cluster in hierarchical clustering to identify typical user behavior patterns.

[0128] Analyze the characteristics of each cluster, identify typical behavior patterns, and provide support for personalized strategies.

[0129] Optionally, use t-SNE technology to project high-dimensional data into three-dimensional space to visually display the distribution of different value groups.

[0130] Optionally, if there is overlap or entanglement, adjust the feature weight assignment and the number of clustering clusters, and re-perform clustering.

[0131] Optionally, according to the visualization results and business requirements, dynamically adjust the clustering method to optimize the hierarchical clustering parameters.

[0132] See Figure 7 , Figure 7 is a schematic structural diagram of an embodiment of the storage medium provided by this application.

[0133] This storage medium 500 stores program data 510, and when the program data 510 is executed by a processor, it implements the steps of the distributed processing engine integration method as described in Figure 1 the above.

[0134] The program data 510 is stored in a storage medium 500, including a number of instructions for causing a network device (such as a router, a personal computer, a server, etc.) or a processor to execute all or part of the steps of the methods of the various embodiments of the present application.

[0135] Optionally, the storage medium 500 can be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RFM), a magnetic disk, an optical disk, or other various media that can store program data.

[0136] See Figure 7 , Figure 7 is a schematic structural diagram of an embodiment of the computer device provided by the present application.

[0137] The computer device 600 includes a processor 620 and a memory 610 that are interconnected. The memory 610 stores a computer program. When the processor 620 executes the computer program, the steps of the distributed processing engine integration method as described above are implemented.

[0138] Different from the prior art, the present application makes the RFM model more applicable to the application scenarios of customer segmentation and value evaluation by expanding the RFM model and introducing new parameters. By constructing an association matrix between users and products, the interaction situation of each user with each product is reflected. By simulating the preference propagation process between users through a preference propagation network, the potential interest of users in un-purchased products can be predicted more accurately. And the extended RFM model is combined with a selection probability matrix to segment users through a clustering algorithm, and the selection profile of each user for each product is obtained, so as to more accurately complete the analysis of the interaction mode of securities customers.

[0139] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for the storage medium embodiment and the computer device embodiment, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.

[0140] The present application can be used in many general or special computing system environments or configurations. For example: personal computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, network PCs, small computers, distributed computing environments including any of the above systems or devices, and so on.

[0141] In several embodiments provided by the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0142] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] In addition, each functional unit in various embodiments of the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0144] The above are only the embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for analyzing securities customers based on the RFM model, characterized in that, Including: Obtain the asset data and transaction data of securities customers; Construct a preference score matrix between users and products based on the asset data and the transaction data; Initialize the network structure according to the association matrix between users and products and the preference similarity between users; Set an initial preference value for each user node in the network structure; Define the rules for the propagation of preferences in the network structure based on the initial preference value to form a preference propagation network; Construct a probability model for each user to select each product according to the preference value of each user node in the preference propagation network to generate a selection probability matrix for each user to select each product; Integrate the selection probability matrix and the extended RFM model parameters to perform hierarchical clustering to assign cluster labels to each user; Analyze the characteristics of each cluster in the hierarchical clustering to identify typical user behavior patterns; The extended RFM model introduces three factors, namely transaction heat, asset score, and return score, into the asset matching parameters.

2. The method for analyzing securities customers according to claim 1, characterized in that, The constructing a preference score matrix between users and products based on the asset data and the transaction data includes: Construct an association matrix between users and products based on the asset data and the transaction data; Calculate the preference similarity between users based on the association matrix between users and products; Calculate the preference degree of users for products according to the preference similarity between users and the actual transaction behavior of users to generate the preference score matrix.

3. The securities customer analysis method according to claim 2, wherein The formula for calculating the preference degree is: Among them, the preference degree (u i , p k ) represents the preference degree of user u i for product p k . The preference similarity (u i , u j ) represents the preference similarity between user u i and user u j . The transaction amount (u j , p k ) represents the transaction amount of user u j for product p k .

4. The method for analyzing securities customers according to claim 2, characterized in that, The calculating the preference degree of users for products according to the preference similarity between users and the actual transaction behavior of users further includes: Obtain a preset preference weight; the preference weight is set according to business requirements and the analysis results of the asset data and the transaction data; Adjust the preference degree of users for products according to the preference weight to generate a weighted preference score matrix.

5. The method for analyzing securities customers according to claim 1, characterized in that, The defining the rules for the propagation of preferences in the network structure based on the initial preference value to form a preference propagation network includes: Update the preference value of users for products through iterative calculation; In response to the preference value satisfying a preset termination condition, stop the iterative calculation.

6. The method for analyzing securities customers according to claim 5, wherein The formula for the iterative calculation is: Among them, the new preference value (u i ) represents the new preference value used to update the initial preference value of user u i . α is the relative importance that controls the initial preference value and the propagated preference value. N(u i ) represents the set of all user nodes adjacent to user u i . N(u j ) represents the set of all user nodes adjacent to user u j . ω(u i , u j ) represents the edge weight between user u i and user u j . ω(u j , u k ) represents the edge weight between user u j and user u k . The preference value (u j ) represents the preference value of user u j ; where α is greater than 0 and less than 1.

7. A storage medium having program data stored thereon, characterized in that, When the program data is executed by a processor, the steps of the securities customer analysis method according to any one of claims 1-6 are implemented.

8. A computer device, characterized in that, Including a processor and a memory connected to each other, the memory stores a computer program, and when the processor executes the computer program, the steps of the securities customer analysis method according to any one of claims 1-6 are implemented.

Images