Ageing-suitable website recommendation method and device, electronic equipment, storage medium and product

Abstract

The invention discloses an aged website recommendation method and device, electronic equipment, a storage medium and a product. The method comprises the following steps: in response to triggering of an aged website recommendation event, obtaining a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each website in a preset area range in a preset historical time period; determining the data entropy of each window based on the window-service queue data matrix, and determining the data entropy of each window based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix; constructing a KNN network diagram based on the window-service queue matrix, and determining a target minimum spanning tree corresponding to the KNN network diagram based on the inter-data entropy and the intra-data entropy; and determining at least one aging-suitable dot based on the target minimum spanning tree. According to the scheme, the lattice points suitable for aging in the preset area range can be accurately determined.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to an elderly-friendly branch recommendation method and device, electronic equipment, storage medium and product. BACKGROUND

[0002] In recent years, more and more elderly users use online channels to make appointments or directly take numbers, which leads to the problem of crowded queues for business at branches with a large number of elderly users nearby, and the elderly users have to spend a long time in actual queues, which easily causes user complaints. When elderly users take numbers through online channels, how to effectively recommend nearby and non-crowded branches for user diversion becomes a key problem. Under this background, a method for recommending elderly-friendly branches is needed. SUMMARY

[0003] The present application provides an elderly-friendly branch recommendation method and device, electronic equipment, storage medium and product, which can accurately determine elderly-friendly branches within a preset area range.

[0004] According to an aspect of the present application, an elderly-friendly branch recommendation method is provided, comprising:

[0005] In response to an elderly-friendly branch recommendation event being triggered, a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each branch within a preset area range within a preset historical time period are obtained; wherein the window-service queue matrix is used to reflect the total number of numbers taken by each service queue of each window of each branch within the preset area range within the preset historical time period; the advantage matrix is used to reflect whether each service queue of each window of each branch within the preset area range meets the elderly-friendly characteristics; the priority matrix is used to reflect the priority of each service queue of each window of each branch within the preset area range; the historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each branch within the preset area range within the preset historical time period; and the opened service matrix is used to reflect whether each service queue of each window of each branch within the preset area range has been opened;

[0006] Based on the window-service queue data matrix, the inter-data entropy of each window is determined, and based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix, the intra-data entropy of each window is determined;

[0007] A KNN network graph is constructed based on the window-service queue matrix, and a target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy;

[0008] determine at least one elderly-friendly site based on the target minimum spanning tree.

[0009] According to another aspect of the present application, there is provided an elderly-friendly site recommendation device, comprising:

[0010] a matrix obtaining module, configured to, in response to an elderly-friendly site recommendation event being triggered, obtain a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each site within a preset regional range within a preset historical time period; wherein the window-service queue matrix is used to reflect a total number of taking numbers of each service queue of each window of each site within the preset regional range within the preset historical time period; the advantage matrix is used to reflect whether each service queue of each window of each site within the preset regional range meets an elderly-friendly feature; the priority matrix is used to reflect a priority of each service queue of each window of each site within the preset regional range; the historical queuing waiting time matrix is used to reflect an average queuing waiting time of each service queue of each window of each site within the preset regional range within the preset historical time period; and the opened service matrix is used to reflect whether each service queue of each window of each site within the preset regional range has been opened;

[0011] a data entropy determining module, configured to determine an inter-data entropy of each window based on the window-service queue data matrix, and determine an intra-data entropy of each window based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix;

[0012] a target minimum spanning tree determining module, configured to construct a KNN network graph based on the window-service queue matrix, and determine a target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy;

[0013] an elderly-friendly site determining module, configured to determine at least one elderly-friendly site based on the target minimum spanning tree.

[0014] According to another aspect of the present application, there is provided an electronic device, comprising:

[0015] at least one processor; and

[0016] a memory connected with the at least one processor in communication; wherein,

[0017] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the elderly-friendly site recommendation method according to any one of the embodiments of the present application.

[0018] According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for causing a processor to implement the method for recommending a senior citizen-friendly site according to any of the embodiments of the present application when executed by the processor.

[0019] According to another aspect of the present application, there is provided a computer program product comprising a computer program for implementing the method for recommending a senior citizen-friendly site according to any of the embodiments of the present application when executed by a processor.

[0020] The method for recommending a senior citizen-friendly site according to the embodiments of the present application, in response to a senior citizen-friendly site recommendation event being triggered, obtains a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each site within a preset area range within a preset historical time period. The window-service queue matrix is used to reflect the total number of taking numbers of each service queue of each window of each site within the preset area range within the preset historical time period. The advantage matrix is used to reflect whether each service queue of each window of each site within the preset area range meets the senior citizen-friendly characteristics. The priority matrix is used to reflect the priority of each service queue of each window of each site within the preset area range. The historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each site within the preset area range within the preset historical time period. The opened service matrix is used to reflect whether each service queue of each window of each site within the preset area range has been opened. The inter-data entropy of each window is determined based on the window-service queue data matrix, and the intra-data entropy of each window is determined based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix. The KNN network graph is constructed based on the window-service queue matrix, and the target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy. At least one senior citizen-friendly site is determined based on the target minimum spanning tree. Through the technical solution provided by the embodiments of the present application, the senior citizen-friendly sites within the preset area range can be accurately determined, and the calculation amount is small, the robustness is strong, and the efficiency is high.

[0021] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to make the technical solution in the embodiments of the present application clearer, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0023] Figure 1 A flow chart of the old-age-adapted net point recommendation method provided by the embodiment of the present application is shown in FIG. 1.

[0024] Figure 2 A flow chart of the old-age-adapted net point recommendation method provided by the embodiment of the present application is shown in FIG. 1.

[0025] Figure 3 A structural schematic diagram of the old-age-adapted net point recommendation device provided by the embodiment of the present application is shown in FIG. 2.

[0026] Figure 4 A structural schematic diagram of the electronic device for implementing the old-age-adapted net point recommendation method of the embodiment of the present application is shown in FIG. 3. DETAILED DESCRIPTION

[0027] In order to make the technical solution in the embodiments of the present application clearer, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0028] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. 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 including a series of steps or units does not necessarily have to include only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0029] Figure 1A flowchart of a method for recommending an age-friendly site is provided in the embodiments of the present application. The embodiments can be applied to the case of recommending an age-friendly site. The method can be executed by an age-friendly site recommendation device, which can be realized in the form of hardware and / or software and can be configured in an electronic device. As shown in FIG. 13, the method comprises the following steps. Figure 1

[0030] S110, in response to the age-friendly site recommendation event being triggered, obtaining a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each site in a preset area range within a preset historical time period.

[0031] The window-service queue matrix is used to reflect the total number of taking numbers of each service queue of each window of each site in the preset area range within the preset historical time period. The advantage matrix is used to reflect whether each service queue of each window of each site in the preset area range meets the age-friendly characteristics. The priority matrix is used to reflect the priority of each service queue of each window of each site in the preset area range. The historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each site in the preset area range within the preset historical time period. The opened service matrix is used to reflect whether each service queue of each window of each site in the preset area range has been opened.

[0032] In the embodiments of the present application, when an age-friendly site recommendation request is received, it is determined that the age-friendly site recommendation event is triggered. In response to the age-friendly site recommendation event being triggered, window-service queue data of each site in a preset area range within a preset historical time period is obtained. The window-service queue data is the total number of taking numbers of each service queue of each window of the site within the preset historical time period. The preset historical time period can be a time period of tracing back a preset time length from the time when the age-friendly site recommendation event is triggered. The preset area range can be the entire urban area or part of the urban area. Based on the window-service queue data, a window-service queue data matrix A = { {a 11 ,...,a 1,n},...,{a m1 ,...,a mn}} is constructed, where the number of rows of the window-service queue data matrix is the total number of windows of each site in the preset area range, and the number of columns of the window-service queue data matrix is the total number of service queues involved by the windows of all sites in the preset area range. It can be understood that the element a ij ​The total number of taking numbers of the jth service queue of the ith window in a preset historical time period.

[0033] In the embodiment of the present application, the old people generally meet the following two rules when they handle business in the network point: one is that the old people take VIP numbers when they handle personal cash and personal non-cash business; and two is that the old people can directly take old people numbers (the old people numbers can be set to high priority). Based on the above rule one, the dominance matrix D = {{d 11 ,...,d 1n},...,{d m1 ,...,d mn}} can be determined, wherein the element d ij in the dominance matrix D indicates whether the jth service queue of the ith window meets the aging characteristics, if it meets, the element d ij can be set to 1, if it does not meet, the element d ij can be set to 0.5. According to the above rule two, the priority of each service queue of each window can be determined, and based on the priority of each service queue of each window, the priority matrix B = {{b 11 ,...,b 1n},...,{b m1 ,...,b mn}} can be constructed, wherein the element b ij in the priority matrix B indicates the priority of the jth service queue of the ith window. The average queuing waiting time of each service queue of each window of each network point in a preset historical time period in a preset area range is obtained, and based on the average queuing waiting time of each service queue of each window, the historical queuing waiting time matrix T = {{t 11 ,...,t 1n},...,{t m1 ,...,t mn}} is constructed, wherein the element t ij in the historical queuing waiting time matrix T indicates the average queuing waiting time of the jth service queue of the ith window in a preset historical time period in a preset area range. Based on whether each service queue of each window of each network point in a preset historical time period in a preset area range is opened, the opened business matrix O = {{o 11 ,...,o 1n},...,{o m1 ,...,o mn}} is constructed, wherein the element o ij in the opened business matrix O indicates whether the jth service queue of the ith window is opened, if it is opened, the element o ij can be set to 1, if it is not opened, the element o ij can be set to 0.5.

[0034] S120, determining data inter-entropy of each window based on the window-service queue data matrix, and determining data intra-entropy of each window based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix.

[0035] In the embodiment of the present application, the data intra-entropy of each window is determined based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix. Optionally, the data intra-entropy of each window is determined based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix, comprising: determining a window-service queue interaction weight matrix based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix; wherein the window-service queue interaction weight matrix is used to reflect the interaction importance degree of each window of each network point in the preset area range connected with each service queue; taking the interaction weight of each window in the window-service queue interaction weight matrix as a whole, the data intra-entropy of each window is calculated based on the window-service queue interaction weight matrix.

[0036] For example, according to the element d ij in the advantage matrix D, the element b ij in the priority matrix B, the element t ij in the historical queuing waiting time matrix T and the element o ij in the opened service matrix O, the interaction weight w ij of the i-th window connected with the j-th service queue is calculated, wherein the interaction weight w ij is used to reflect the interaction importance degree of the i-th window connected with the j-th service queue. For example, the interaction weight w ij may be calculated according to the following formula:

[0037]

[0038] The window-service queue interaction weight matrix W is constructed based on the interaction weight of each window connected with each service queue. Since each element in each row of the window-service queue interaction weight matrix W respectively represents the interaction weight (i.e. the interaction importance degree) of each window connected with each service queue, taking the interaction weight of each window in the window-service queue interaction weight matrix W as a whole, the data intra-entropy of each window is calculated based on the window-service queue interaction weight matrix. For example, the data intra-entropy of the i-th window can be calculated according to the following formula:

[0039]

[0040] wherein, πi denotes an invariant measure, satisfying π i W = π i and the normalization constraint π t 1 = 1. ITE i denotes the intra-data entropy of the ith window, nITE i denotes the normalized intra-data entropy of the ith window, max(ITE) denotes the maximum value among the intra-data entropies of all windows.

[0041] In the embodiments of the present application, the inter-data entropy of each window is determined based on the window-service queue data matrix A. Optionally, the inter-data entropy of each window is determined based on the window-service queue data matrix, comprising: determining the sample variance and the neighborhood variance of each element node in the window-service queue data matrix; performing denoising processing on the window-service queue data matrix based on the sample variance and the neighborhood variance to generate an optimized data matrix; taking the service quantity element of each window in the optimized data matrix as a whole, and calculating the inter-data entropy of each window based on the optimized data matrix.

[0042] Since there may be a large amount of noise in high-dimensional data, the existence of the noise may affect the accuracy of subsequent data processing, therefore, it is necessary to perform denoising processing on the window-service queue data matrix A. Optionally, before performing denoising processing on the window-service queue data matrix A, the window-service queue data matrix A can be normalized, and the normalized window-service queue data matrix is taken as a normalized matrix X. Illustratively, the preset area range contains k network points, then the distances ST = {st1,..., st k} between each network point and the user in the preset area range are obtained, wherein the distance between each network point and the user is the average distance between all users who have conducted business at the network point in a preset historical time period and the network point. Illustratively, the elements a ij in the window-service queue data matrix A are normalized according to the following formula:

[0043]

[0044] wherein x ij denotes the element after normalization of the element a ij in the window-service queue data matrix A, s i denotes the location distance between the network point and the customer, max(A i ) denotes the maximum value among the total number of numbers taken by all service queues of the ith window (i.e., the maximum total number of numbers taken by the ith window), and max(ST) denotes the maximum location distance between the network point and the user (i.e., ST = {st1,..., st kmax(ST) represents the maximum distance between the grid points and the users (i.e. ST = {st1,...,stn}, max(ST) = max{st1,...,stn} ), and min(ST) represents the minimum distance between the grid points and the users (i.e. ST = {st1,...,stn}, min(ST) = min{st1,...,stn} ). k i

[0045] In the embodiment of the present application, a sequence composed of elements in each row of the normalized window-service queue data matrix (i.e. normalized matrix X) is taken as a sample sequence corresponding to each node, and the sample variance of each node is calculated according to the sample sequence corresponding to each node. wherein, represents the sample variance of the ith node corresponding to the sample sequence composed of elements in the ith row of the normalized matrix X. For example, the sample variance of the ith node can be calculated according to the following formula:

[0046]

[0047] wherein, x i represents the sample sequence of the ith node (containing x i1 ,..., x i2 ,..., x in ,..., x ij n sample elements), represents the sample mean of the sample sequence of the ith node, and n represents the number of samples contained in the sample sequence of the ith node, i.e. the number of elements contained in the ith row of the normalized matrix X, x i represents the jth element in the sample sequence of the ith node, i.e. the element in the jth column of the ith row of the normalized matrix X, is the sample variance of the sample sequence x i . It can be understood that the sample variance of the ith node is taken as the sample variance of each element in the ith row of the window-service queue data matrix A, i.e. the sample variance of each element in the same row of the window-service queue data matrix A is the same.

[0048] For example, the neighborhood variance of the jth column element node x j in the normalized matrix X can be calculated according to the following formula:

[0049]

[0050] wherein, k c represents the ith element x ijThe number of adjacent neighbor nodes can be manually adjusted, where t represents the t-th neighbor node and m represents the number of elements in the j-th column of the element sequence. This can be understood as placing the j-th column element node x... j neighborhood variance The neighborhood variance of each element in the j-th column of the normalized matrix X is the same as the neighborhood variance of each element in the same column of the normalized matrix X.

[0051] The normalized window-business queue data matrix is ​​denoised based on sample variance and neighborhood variance to generate an optimized data matrix. For example, for each element x in the normalized matrix X... ij The sample determines the neighborhood variance corresponding to the element. Is it smaller than the sample size? If the difference is significant, then the element after denoising is x. ij Otherwise, the denoised elements will be 0. For example, the normalized matrix X is denoised according to the following formula:

[0052]

[0053] Among them, y ij This represents the elements x in the normalized matrix X. ij In this embodiment of the invention, the matrix generated after denoising the normalized matrix X is called the optimized data matrix Y. It should be noted that after denoising the normalized matrix X, there may be columns where all elements are 0. Therefore, these columns are deleted, resulting in an optimized data matrix of size m×h, where h is less than or equal to n. It should also be noted that if a column containing all 0 elements is deleted after denoising the normalized matrix X, then before determining the data entropy of each window based on the advantage matrix, priority matrix, historical queuing time matrix, and activated service matrix, all elements in the corresponding columns of these matrices are deleted.

[0054] In this embodiment of the invention, the business quantity elements of each window in the optimized data matrix are treated as a whole, and the data entropy between each window is calculated based on the optimized data matrix. It can be understood that the number of data entropy elements is the same as the number of rows in the optimized data matrix. For example, the optimized data matrix Y is decomposed using a random nonnegative matrix factorization algorithm, and the calculation method is as follows:

[0055]

[0056] wherein, W(h*v) is a basic matrix, H(v*m) is a coefficient matrix, is a random constraint, is a regularization parameter determined by BIC criterion, I is an m*m unit matrix value at 0-1, R is an m*m unit matrix value at 0-1, and v is determined by Gap statistics.

[0057] The service quantity element of each window in the optimized data matrix Y is taken as a whole. Since the part of the whole belongs to the changing intermediate state, the inter-data entropy can be defined to capture the certainty of the continuous change relationship between each whole. After obtaining H(v*m) according to the above formula, the probability matrix P is converted, and the inter-data entropy IRE is calculated, which is calculated as follows:

[0058]

[0059] wherein, p ji represents the probability that window j belongs to cluster i, H ij is a coefficient matrix, IRE i represents the inter-data entropy of the i th window, nIRE i represents the normalized inter-data entropy of the i th window, and max(IRE) represents the maximum value in the inter-data entropy of all windows.

[0060] In S130, a KNN network graph is constructed based on the window-service queue matrix, and a target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy.

[0061] In the embodiment of the present application, the window-service queue matrix is analyzed to construct a KNN network graph. Optionally, the KNN network graph is constructed based on the window-service queue matrix, which includes: denoising the window-service queue data matrix to generate an optimized data matrix, and reducing the dimension of the optimized data matrix based on the UMAP dimension reduction algorithm to generate a target data matrix; and constructing a KNN network graph based on the target data matrix by a KNN algorithm. Illustratively, the window-service queue data matrix A is denoised according to the above denoising method to generate an optimized data matrix Y, and the optimized data matrix Y is reduced in dimension based on the UMAP dimension reduction algorithm to generate a target data matrix Z. It can be understood that, due to the problems of high-dimensional complex data such as complex calculation, difficult analysis result, and difficult abstract understanding, a dimension reduction technique is used to obtain a low-dimensional data matrix. The dimension reduction algorithm refers to a series of techniques that map data in the original high-dimensional space to a low-dimensional space by using a certain mathematical mapping method, which has the effects of reducing data dimension, removing redundant features in data, reducing calculation cost, improving data analysis accuracy, and facilitating data visualization. In the embodiment of the present application, the UMAP dimension reduction algorithm is used to reduce the dimension of the optimized data matrix Y. Illustratively, the target data matrix Z generated by dimension reduction can be represented as:

[0062] In the embodiments of the present application, based on the target data matrix Z, the following steps can be used to construct the KNN network graph: ① Based on the matrix Z, a two-dimensional graph containing m nodes is initially drawn, each node in the graph is taken as an initial node, and the distance between two nodes is taken as the similarity between the two nodes. ② The KNN algorithm is used to construct the KNN network graph in the two-dimensional coordinates. When the node z i is the nearest K neighbor node of the node z j , or the node z j is the nearest K neighbor node of the node z i , the node z i is connected to the node z j , and Z knn is formed. The value of K in the KNN algorithm can be manually adjusted. Each node in the KNN network graph can be connected by an edge, and the weight on the edge is determined by the Euclidean distance between the two nodes.

[0063] In the embodiments of the present application, after the KNN network graph is constructed based on the window-service queue matrix, the target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy. Determining the target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy includes: randomly extracting a preset number of target nodes from the KNN network graph, and dividing the preset number of target nodes into at least two node clusters based on a preset clustering algorithm; for each node cluster, determining a cluster average inter-data entropy corresponding to the node cluster according to the inter-data entropy corresponding to each node included in the node cluster, and determining a cluster average intra-data entropy corresponding to the node cluster according to the intra-data entropy corresponding to each node included in the node cluster; and constructing the target minimum spanning tree corresponding to the KNN network graph based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to the at least two node clusters.

[0064] In the embodiments of the present application, in order to improve the robustness of the old-age network point recommendation result and further reduce the influence of noise on the recommendation result, a preset number of target nodes are randomly extracted from the KNN network graph, for example, V% target nodes are extracted, and the value of V can be 90, and the preset number of target nodes are divided into at least two node clusters based on a preset clustering algorithm. Optionally, the preset clustering algorithm can be the Louvain algorithm. The preset number of target nodes are divided based on the Louvain algorithm, wherein nodes with high similarity are divided into the same node cluster. For example, a node cluster can be represented as C={c1,...,c i ,...,c N, where N represents the number of nodes contained in the node cluster. For each node cluster, the cluster average inter-data entropy corresponding to the node cluster is determined according to the inter-data entropy corresponding to each node contained in the node cluster, and the cluster average intra-data entropy corresponding to the node cluster is determined according to the intra-data entropy corresponding to each node contained in the node cluster. For example, for the ith node cluster C i The cluster average inter-data entropy and the cluster average intra-data entropy of the node cluster can be determined according to the following formula:

[0065]

[0066] where C i represents the node cluster i, represents the cluster average intra-data entropy of the node cluster C i , and N(C i ) represents the number of nodes contained in the cluster C i . represents the cluster average inter-data entropy of the node cluster C i .

[0067] It can be understood that the cluster average intra-data entropy and the cluster average inter-data entropy corresponding to each node cluster can be determined in the above manner. In the embodiment of the present application, the target minimum spanning tree corresponding to the KNN network graph is constructed based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to at least two node clusters. Optionally, the target minimum spanning tree corresponding to the KNN network graph is constructed based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to the at least two node clusters, and the method comprises: constructing a network connection graph based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to the at least two node clusters; determining the cluster minimum spanning tree corresponding to the network connection graph based on a preset minimum spanning tree construction strategy; determining the smooth node trajectory corresponding to the cluster minimum spanning tree through a principal curve algorithm, and returning to execute the step of randomly extracting a preset number of target nodes from the KNN network graph until K smooth node trajectories are determined; wherein each smooth node trajectory contains N fitting nodes; and determining the target minimum spanning tree corresponding to the KNN network graph constructed based on the K*N fitting nodes on the K smooth node trajectories.

[0068] In the embodiment of the present application, the cluster average intra-data entropy decreases as the feature expression trend of the node cluster strengthens, and the cluster average inter-data entropy increases when the node cluster is in an intermediate state. For each pair of node clusters {C i ,C j}, if and If the Euclidean distance between the two node clusters is less than the preset threshold, it indicates that the relevance between the two node clusters is strong, and thus the two node clusters can be regarded as having a connection relationship, and the Euclidean distance between the two node clusters is taken as the connection edge weight of the two node clusters. Based on the above strategy, a network connection graph can be constructed. Based on the preset minimum spanning tree construction strategy, a cluster minimum spanning tree corresponding to the network connection graph is determined. For example, for a network connection graph including N nodes, the preset minimum spanning tree construction strategy can include the following four points: a. Any two nodes on the tree are reachable; b. There is no loop in the tree; c. Only N-1 edges are used; and d. The sum of the weights of the edges is minimum. Assuming that the determined cluster minimum spanning tree includes N nodes, the selected data points are projected onto the cluster minimum spanning tree by using the principal curve algorithm, and a smooth node trajectory is obtained after iterative calculation, wherein the smooth node trajectory includes N fitting nodes. The above step is repeatedly executed K times, and K smooth node trajectories are obtained. Then, a target KNN network graph is constructed based on the K*N fitting nodes in the K smooth node trajectories, and a target minimum spanning tree corresponding to the target KNN network graph is determined based on the preset minimum spanning tree construction strategy.

[0069] In S140, at least one aging adaptation network point is determined based on the target minimum spanning tree.

[0070] For example, the principal curve algorithm can be used to determine a target smooth node trajectory corresponding to the target minimum spanning tree, and each node in the target smooth node trajectory is arranged in descending order according to the feature transformation trend. The more forward the node is, the stronger the aging adaptation feature of the window corresponding to the node is, and the more suitable the node is for the elderly to conduct business. Optionally, the network point where the window corresponding to the r most forward nodes is located can be taken as an aging adaptation network point recommended to the elderly, and the r aging adaptation network points are recommended to the elderly. The elderly can independently select a network point close to the r aging adaptation network points and having configured aging adaptation parameters from the r aging adaptation network points.

[0071] Figure 2 A flowchart of an aging adaptation network point recommendation method provided by an embodiment of the present application can be understood in combination with the description of the above embodiment, and will not be described herein again. Figure 2

[0072] ​The old-age adaptation point recommendation method provided by the embodiment of the application is triggered in response to an old-age adaptation point recommendation event, and a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix and an opened service matrix of each point in a preset area range within a preset historical time period are acquired; the window-service queue matrix is used to reflect the total number of taking numbers of each service queue of each window of each point in the preset area range within the preset historical time period; the advantage matrix is used to reflect whether each service queue of each window of each point in the preset area range meets the old-age adaptation characteristics; the priority matrix is used to reflect the priority of each service queue of each window of each point in the preset area range; the historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each point in the preset area range within the preset historical time period; and the opened service matrix is used to reflect whether each service queue of each window of each point in the preset area range has been opened; the inter-data entropy of each window is determined based on the window-service queue data matrix, and the intra-data entropy of each window is determined based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix and the opened service matrix; a KNN network graph is constructed based on the window-service queue matrix, and the target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy; and at least one old-age adaptation point is determined based on the target minimum spanning tree. Through the technical scheme provided by the embodiment of the application, the old-age adaptation point in the preset area range can be accurately determined, and the calculation amount is small, the robustness is strong, and the efficiency is high.

[0073] Figure 3 A structural schematic diagram of an old-age adaptation point recommendation device provided by the embodiment of the application is shown in FIG. 1. Figure 3 As shown in the figure, the device comprises:

[0074] The matrix obtaining module 310 is configured to, in response to the elderly-friendly site recommendation event being triggered, obtain a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix, and an opened service matrix of each site within a preset regional range in a preset historical time period. The window-service queue matrix is used to reflect a total number of taking numbers of each service queue of each window of each site within the preset regional range in the preset historical time period. The advantage matrix is used to reflect whether each service queue of each window of each site within the preset regional range meets the elderly-friendly characteristics. The priority matrix is used to reflect a priority of each service queue of each window of each site within the preset regional range. The historical queuing waiting time matrix is used to reflect an average queuing waiting time of each service queue of each window of each site within the preset regional range in the preset historical time period. The opened service matrix is used to reflect whether each service queue of each window of each site within the preset regional range has been opened.

[0075] The data entropy determining module 320 is configured to determine an inter-data entropy of each window based on the window-service queue data matrix, and determine an intra-data entropy of each window based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix.

[0076] The target minimum spanning tree determining module 330 is configured to construct a KNN network graph based on the window-service queue matrix, and determine a target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy.

[0077] The elderly-friendly site determining module 340 is configured to determine at least one elderly-friendly site based on the target minimum spanning tree.

[0078] Optionally, the data entropy determining module is configured to:

[0079] determine a window-service queue interaction weight matrix based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix. The window-service queue interaction weight matrix is used to reflect an interaction importance degree of each window and each service queue of each site within the preset regional range.

[0080] take the interaction weight of each window in the window-service queue interaction weight matrix as a whole, and calculate the intra-data entropy of each window based on the window-service queue interaction weight matrix.

[0081] Optionally, the data entropy determining module is configured to:

[0082] determining a sample variance and a neighborhood variance of each element node in the window-service queue data matrix;

[0083] performing denoising processing on the window-service queue data matrix based on the sample variance and the neighborhood variance to generate an optimized data matrix;

[0084] calculating, based on the optimized data matrix, a data inter-entropy of each window by taking the service quantity element of each window in the optimized data matrix as a whole.

[0085] Optionally, the target minimum spanning tree determination module is configured to:

[0086] performing denoising processing on the window-service queue data matrix to generate an optimized data matrix, and performing dimension reduction on the optimized data matrix based on a UMAP dimension reduction algorithm to generate a target data matrix;

[0087] constructing a KNN network graph based on the target data matrix by using a KNN algorithm.

[0088] Optionally, the target minimum spanning tree determination module comprises:

[0089] a node cluster determination unit configured to randomly extract a preset number of target nodes from the KNN network graph, and divide the preset number of target nodes into at least two node clusters based on a preset clustering algorithm;

[0090] a cluster average data entropy determination unit configured to determine, for each node cluster, a cluster average data inter-entropy corresponding to the node cluster according to the data inter-entropy corresponding to each node included in the node cluster, and determine a cluster average data intra-entropy corresponding to the node cluster according to the data intra-entropy corresponding to each node included in the node cluster;

[0091] a target minimum spanning tree construction unit configured to construct a target minimum spanning tree corresponding to the KNN network graph based on the cluster average data inter-entropy and the cluster average data intra-entropy corresponding to the at least two node clusters.

[0092] Optionally, the target minimum spanning tree construction unit is configured to:

[0093] construct a network connection graph based on the cluster average data inter-entropy and the cluster average data intra-entropy corresponding to the at least two node clusters;

[0094] determine a cluster minimum spanning tree corresponding to the network connection graph based on a preset minimum spanning tree construction strategy;

[0095] Determine the smooth node trajectory corresponding to the cluster minimum spanning tree through the principal curve algorithm, and return to execute random extraction of a preset number of target nodes from the KNN network graph until K smooth node trajectories are determined; wherein each smooth node trajectory contains N fitting nodes.

[0096] Determine the target minimum spanning tree corresponding to the KNN network graph constructed based on the K*N fitting nodes on the K smooth node trajectories.

[0097] The suitable aging network point recommendation device provided in the embodiments of the application can execute the suitable aging network point recommendation method provided in any embodiment of the application, has the function modules and beneficial effects corresponding to the execution method.

[0098] Figure 4 A structural schematic diagram of an electronic device 10 that can be used to implement embodiments of the application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the application described and / or claimed in this document.

[0099] As shown in Figure 4 The electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11, wherein the memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0100] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0101] The processor 11 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the age-appropriate web point recommendation method.

[0102] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program according to embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication unit 19, or installed from the storage unit 18, or installed from the ROM 12. When the computer program is executed by the processor 11, the above-described functions defined in the methods of embodiments of the present application are performed.

[0103] In some embodiments, the age-appropriate web point recommendation method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded onto the RAM 13 and executed by the processor 11, one or more steps of the age-appropriate web point recommendation method described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the age-appropriate web point recommendation method by any other appropriate means, such as by means of firmware.

[0104] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a special-purpose standard product (ASSP), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0105] Computer programs used to implement the methods of the application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the computer program, when executed by the processor of the machine, implements the functions / acts specified in the flowcharts and / or block diagrams. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package, and partially on a machine or entirely on a remote machine or server.

[0106] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store computer programs for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0107] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0108] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0109] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0110] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be executed in parallel, executed in sequence, or executed in a different order, as long as the desired results of the present disclosure are achieved, and the present disclosure is not limited herein.

[0111] The specific embodiments described hereinabove are illustrative only and not restrictive. One skilled in the art will appreciate that variations and modifications can be made to the specifics described herein without departing from the spirit and principles of the application. Accordingly, the scope of protection is not limited to the specific details described herein but is given by the appended claims.

Claims

1. A method for recommending age-friendly service outlets, characterized in that, The method comprises the following steps: In response to an elderly adaptation network point recommendation event being triggered, a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix, and an opened service matrix of each network point in a preset regional range within a preset historical time period are obtained; wherein the window-service queue matrix is used to reflect the total number of taking numbers of each service queue of each window of each network point in the preset regional range within the preset historical time period; the advantage matrix is used to reflect whether each service queue of each window of each network point in the preset regional range meets the characteristics of elderly adaptation; the priority matrix is used to reflect the priority of each service queue of each window of each network point in the preset regional range; the historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each network point in the preset regional range within the preset historical time period; and the opened service matrix is used to reflect whether each service queue of each window of each network point in the preset regional range has been opened; Based on the window-service queue data matrix, the inter-data entropy of each window is determined, and based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix, the intra-data entropy of each window is determined; Based on the window-service queue matrix, a KNN network graph is constructed, and the target minimum spanning tree corresponding to the KNN network graph is determined based on the inter-data entropy and the intra-data entropy; Based on the target minimum spanning tree, at least one elderly adaptation network point is determined.

2. The method of claim 1, wherein, Based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix, the intra-data entropy of each window is determined, comprising: Based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix, a window-service queue interaction weight matrix is determined; wherein the window-service queue interaction weight matrix is used to reflect the interaction importance degree of the connection between each window and each service queue of each network point in the preset regional range; The interaction weight of each window in the window-service queue interaction weight matrix is taken as a whole, and the intra-data entropy of each window is calculated based on the window-service queue interaction weight matrix.

3. The method of claim 1, wherein, Based on the window-service queue data matrix, the inter-data entropy of each window is determined, comprising: For each element node in the window-service queue data matrix, the sample variance and the neighborhood variance of the element node are determined; Based on the sample variance and the neighborhood variance, the window-service queue data matrix is denoised to generate an optimized data matrix; The number of services of each window in the optimized data matrix is taken as a whole, and the inter-data entropy of each window is calculated based on the optimized data matrix.

4. The method of claim 1, wherein, Based on the window-service queue matrix, a KNN network graph is constructed, comprising: The window-service queue data matrix is denoised to generate an optimized data matrix, and the optimized data matrix is reduced in dimension based on a UMAP dimension reduction algorithm to generate a target data matrix; Construct a KNN network graph based on the target data matrix through a KNN algorithm.

5. The method of claim 1, wherein, Determine a target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy, including: Randomly extract a preset number of target nodes from the KNN network graph, and divide the preset number of target nodes into at least two node clusters based on a preset clustering algorithm; For each node cluster, determine a cluster average inter-data entropy corresponding to the node cluster according to the inter-data entropy corresponding to each node included in the node cluster, and determine a cluster average intra-data entropy corresponding to the node cluster according to the intra-data entropy corresponding to each node included in the node cluster; Construct the target minimum spanning tree corresponding to the KNN network graph based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to the at least two node clusters.

6. The method of claim 5, wherein, Determine a target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy, including: Construct a network connection graph based on the cluster average inter-data entropy and the cluster average intra-data entropy corresponding to the at least two node clusters; Determine a cluster minimum spanning tree corresponding to the network connection graph based on a preset minimum spanning tree construction strategy; Determine a smooth node trajectory corresponding to the cluster minimum spanning tree through a principal curve algorithm, and return to execute the step of randomly extracting a preset number of target nodes from the KNN network graph until K smooth node trajectories are determined; each smooth node trajectory includes N fitting nodes. Determine a target minimum spanning tree corresponding to a KNN network graph constructed based on K*N fitting nodes on the K smooth node trajectories.

7. An aging-appropriate mesh point recommendation device, comprising: Including: A matrix obtaining module is configured to, in response to an event of recommending a suitable aging network point being triggered, obtain a window-service queue matrix, an advantage matrix, a priority matrix, a historical queuing waiting time matrix, and an opened service matrix of each network point in a preset regional range within a preset historical time period. The window-service queue matrix is used to reflect the total number of tickets drawn by each service queue of each window of each network point in the preset regional range within the preset historical time period. The advantage matrix is used to reflect whether each service queue of each window of each network point in the preset regional range meets the characteristics of suitable aging. The priority matrix is used to reflect the priority of each service queue of each window of each network point in the preset regional range. The historical queuing waiting time matrix is used to reflect the average queuing waiting time of each service queue of each window of each network point in the preset regional range within the preset historical time period. The opened service matrix is used to reflect whether each service queue of each window of each network point in the preset regional range has been opened. A data entropy determining module is configured to determine an inter-data entropy of each window based on the window-service queue data matrix, and determine an intra-data entropy of each window based on the advantage matrix, the priority matrix, the historical queuing waiting time matrix, and the opened service matrix. The target minimum spanning tree determination module is configured for constructing a KNN network graph based on the window-service queue matrix, and determining a target minimum spanning tree corresponding to the KNN network graph based on the inter-data entropy and the intra-data entropy; The elderly-friendly net point determination module is configured for determining at least one elderly-friendly net point based on the target minimum spanning tree.

8. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the elderly-friendly net point recommendation method in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for enabling the processor to execute the elderly-friendly net point recommendation method in any one of claims 1-6 when executed.

10. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program, when executed by the processor, implements the elderly-friendly net point recommendation method according to any one of claims 1-6.

Images