Unified payment cashier desk processing system for Internet medical treatment

Abstract

The invention belongs to the technical field of medical payment, and discloses a unified payment cashier desk processing system for Internet medical treatment. The method comprises the following steps: establishing channel configuration of a payment channel; docking the payment channel configuration according to a preset standard interface; partitioning the payment channel to obtain n payment regions; for each payment area, constructing an effective configuration combination of payment channels; the combination of payment scene identifiers is acquired in real time, and a payment channel is selected from the effective configuration combination of the corresponding payment area for payment processing; and the payment strategy can be continuously optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical payment, and more specifically, to a unified payment cashier processing system for Internet-based medical services. Background Art

[0002] The patent application with the publication number CN118863877A discloses a full-scenario intelligent payment system based on an Internet medical platform, including: a business processing layer: an interface between the system and various hospital business systems, responsible for processing payment requests received from various hospital business systems; a payment gateway layer: responsible for managing and coordinating different payment channels; a cashier core layer: responsible for coordinating the execution of payment operations and managing transaction security; a docking layer: responsible for the interaction between the system and the hospital HIS system and external payment platforms; a settlement and reconciliation layer: a layer responsible for financial verification and accounting processing; a user layer: an interface for the system to interact with end-users. In the present invention, through standardized API interfaces and modular design, the system can seamlessly access various hospital business systems and multiple payment channels, achieving high integration. This integration reduces the workload of the hospital for docking and maintenance between different systems and improves the overall efficiency of the system.

[0003] However, in complex and ever-changing medical payment scenarios, traditional systems are difficult to effectively manage and optimize multiple payment channels, resulting in low payment efficiency, poor user experience, inability to select the optimal payment channel based on real-time data, and low capital turnover efficiency; in addition, the differences in interfaces of different payment channels increase the complexity of system integration and maintenance, making it difficult for medical institutions to expand payment methods; when dealing with large-scale payment transactions, existing systems lack scientific optimization methods and are difficult to handle high-concurrency payment requirements, especially during special periods such as holidays or the epidemic; at the same time, due to the lack of a comprehensive performance evaluation mechanism, medical institutions are difficult to accurately identify and solve bottleneck problems in the payment system, affecting the overall operation efficiency; these problems not only increase the operating costs of medical institutions but also have a negative impact on the quality of medical services.

[0004] In view of this, the present invention proposes a unified payment cashier processing system for Internet-based medical services to solve the above problems. Summary of the Invention

[0005] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: A unified payment cashier processing system for Internet-based medical services, including: a channel configuration module, used to establish the channel configuration of payment channels; and dock the payment channel configuration according to a preset standard interface;

[0006] a channel partition module, used to partition payment channels to obtain n payment regions;

[0007] A configuration module, which is used to build an effective configuration combination of payment channels for each payment area;

[0008] A payment aggregation module, which is used to obtain the combination of payment scenario identifiers in real time, and select a payment channel from the effective configuration combination of the corresponding payment area for payment processing; each module is connected by wired and / or wireless means.

[0009] Furthermore, the method for establishing the channel configuration of the payment channel includes:

[0010] Collect and maintain the configuration information of each payment channel, including the payment channel name, payment channel ID, payment channel merchant ID, and payment channel key;

[0011] Define a data structure for storing the configuration information of the payment channel. The payment channel ID of each payment channel serves as the identifier of the corresponding payment channel; collect and maintain several payment scenario identifiers, each payment scenario identifier corresponding to a payment scenario, and these payment scenario identifiers will be used as key values to correspond to the configuration information of different payment channels; establish a mapping relationship, store the combination of payment scenario identifiers as the key value and the corresponding payment channel ID as the value in the data structure;

[0012] Define a rule for dynamically adjusting the mapping relationship, which is used to dynamically adjust the mapping relationship in the data structure, that is, the establishment of the payment channel configuration is completed.

[0013] Furthermore, the definition method of the rule for dynamically adjusting the mapping relationship includes:

[0014] Map the payment scenario identifiers to a multi-dimensional space, denoted as the payment scenario space. Each point in the payment scenario space represents a specific combination of payment scenarios; collect payment requests and payment data within a fixed historical time; record the payment scenario identifier and the used payment channel ID of each payment request, and map the payment data to the corresponding space point according to the payment scenario identifier; construct a scan line in the payment scenario space;

[0015] Traverse each point in the payment scenario space along the direction of the scan line, and calculate the evaluation index of the payment data at each point for all available payment channels; the calculation formula of the evaluation index is:

[0016]

[0017] where CS is the evaluation index, PS is the payment success rate, PD is the payment delay, PE is the payment exception rate, PF is the payment fee, PR is the payment refund rate, w1, w2, w3, w4, and w5 are the weight coefficients of each index, and the values are between 0 and 1;

[0018] Set a preset evaluation threshold, filter out payment channels whose evaluation metrics are greater than the evaluation threshold as backup payment channels. For each point, sort the backup payment channels in descending order according to the evaluation metrics, and select the top one or several payment channels as the recommended payment channels for this point. If there are no backup payment channels for a certain point, select the payment channel with the highest evaluation metric, or borrow the recommended payment channel from the adjacent point; update the payment channel IDs corresponding to the recommended payment channels of each point into the mapping relationship, and dynamically adjust the scan line during this process.

[0019] Further, the method for dynamically adjusting the scan line includes:

[0020] Within a predetermined time period, collect new payment requests and payment data to form a payment request data set D_new; for each point p in the payment scenario space S, calculate the shortest distance d(p, L_old) from p to the original scan line L_old, define a coverage radius r. If d(p, L_old) is less than or equal to r, then the point p is covered by the scan line L_old, and count the number of covered points Nc and the number of uncovered points Nu.

[0021] Map D_new to the payment scenario space S to obtain a new point set P_new. For each point p in S, calculate the nearest neighbor distance d_min(p, P_new) from p to P_new, define a change threshold TL. If d_min(p, P_new) is greater than TL, then mark the area where the point p is located as a change area Rc.

[0022] For each point p' in the change area Rc, calculate the shortest distance d(p, L_old) from p' to the scan line L_old, define an adjustment radius ra. If d(p, L_old) is greater than ra, then use a B-spline curve to fit the points in the change area Rc to obtain a new scan line segment L_new, and insert the new scan line segment L_new into the scan line L_old to obtain an adjusted scan line Lad.

[0023] Define an expected length. If the length of the adjusted scan line Lad is greater than the expected length, then shorten the length of the adjusted scan line Lad to the expected length; if the length of the adjusted scan line Lad is less than the expected length, then extend the length of the adjusted scan line Lad to the expected length.

[0024] Use the kernel density estimation method to construct the distribution density function f of the scan line L_old in the payment scenario space S. For each point in the point set P_new, calculate its log-likelihood value lu with respect to the current distribution density function f and sum them to obtain the total log-likelihood value LU. Use the gradient ascent optimization algorithm to adjust the parameters of the distribution density function f to maximize the total log-likelihood value LU. Obtain the adjusted distribution density function f'. Resample the adjusted scan line Lad according to the adjusted distribution density function f' to obtain the optimized scan line Lop. Subsequently, use the optimized scan line Lop to traverse each point p in S and select a recommended payment channel for p.

[0025] Further, the method for partitioning payment channels includes:

[0026] Extract the key features of the channel configuration of the payment channels; map the key features of each payment channel to a multi-dimensional feature space with the dimension of m, that is, m key features; each dimension of the multi-dimensional feature space corresponds to a key feature, and in the multi-dimensional feature space, each payment channel corresponds to a feature point;

[0027] For each dimension i, set a threshold range [low_i, high_i]; where low_i is the lower limit of dimension i and high_i is the upper limit of dimension i; define a clipping window, which is an m-dimensional hyper-rectangle in the multi-dimensional feature space. For each dimension i, the boundary of the clipping window in this dimension is determined by [low_i, high_i]; that is, the lower left vertex of the clipping window is low_i and the upper right vertex is high_i; for each dimension i, define two code elements 0_i and 1_i. If the coordinate value of the feature point s is less than the lower limit of the clipping window in this dimension, the corresponding code element is 1_i; if the coordinate value of the feature point s is greater than the upper limit of the clipping window in this dimension, the corresponding code element is 0_i; otherwise, the corresponding code element is 0;

[0028] Connect the code elements in all dimensions to obtain an m-bit code. If the code is all 0, the feature point s is completely within the clipping window. If each bit of the code is not 0, the feature point s is completely outside the clipping window; otherwise, it means that the feature point s intersects the boundary of the clipping window;

[0029] For the feature points that intersect with the boundary of the clipping window, use parametric equations to calculate the intersection points of the corresponding feature points and the boundary of the clipping window. Based on the intersection points, divide the corresponding feature points into two new points, which are located inside and outside the clipping window respectively; classify the feature points that are completely inside the clipping window into the same region. For adjacent regions, construct the key features inside them into vectors, denoted as feature vectors. For any two regions, calculate the similarity between their feature vectors, set a similarity threshold. If the similarity is greater than or equal to the similarity threshold, merge the corresponding regions into a new region; repeat the merging until no further merging can be done; adjust the position and size of the clipping window, and repeat the iteration. Each iteration will obtain a new set of regions until all feature points are assigned to a certain region; output the regions obtained in the last iteration as the final result, that is, obtain n payment regions.

[0030] Furthermore, the construction method of the effective configuration combination includes:

[0031] Regard the payment region as a two-dimensional lattice model. The two-dimensional lattice model consists of a series of grids and nodes, and each node represents a payment channel within the payment region; define the model parameters of the two-dimensional lattice model, and the model parameters include the hopping parameter T, the interaction parameter Y, and the chemical potential μ;

[0032] Define an energy function H for calculating the total energy of the two-dimensional lattice model;

[0033] Randomly assign the payment channels within the payment region to the nodes to form an initial state; use the Metropolis algorithm to sample the initial state; at each step, randomly select a node and change its state; calculate the difference in the total energy of the two-dimensional lattice model before and after the state change, denoted as the energy difference ΔE, and decide whether to accept the new state with a probability of min(1, exp(-β×ΔE)); where β is the inverse temperature; record the change of the total energy with the number of steps, gradually increase the value of the inverse temperature to make the two-dimensional lattice model tend to be stable, and repeat until the total energy is the lowest. At this time, the two-dimensional lattice model is the ground state configuration; map the ground state configuration to the two-dimensional plane, count the number of payment channels on each node to obtain the overall distribution of the number of payment channels, and based on the distribution of the number of payment channels, perform cluster analysis on the ground state configuration to identify the aggregation regions of the payment channels; within the aggregation regions, identify the fixed payment channel combination patterns, which are the effective configuration combinations.

[0034] Furthermore, the formula of the energy function is:

[0035] H=-T·∑(c + _u·c_v+c + _v·c_u)-Y·∑(n_u·n_v)-μ·∑n_u-z·GH;

[0036] Among them, c + _u is the creation factor of node u, which acts on node u to create a payment channel on node u. c_v is the elimination factor of node v, which acts on node v to remove the payment channel on node v; c + _v is the creation factor of node v, which acts on node v to create a payment channel on node v. c_u is the elimination factor of node u, which acts on node u to remove the payment channel on node u;

[0037] n_u is the number operator of the payment channels on node u, which acts on node u. The eigenvalue of n_u is 0 or 1, corresponding to whether there is a payment channel on node u respectively; z is the distance decay parameter, with a value between 0 and 1; GH is the distance decay function;

[0038] The distance decay function GH = ∑ (u,v) (|r_u - r_v|·n_u·n_v); where, r_u is the position vector of node u on the plane of the two-dimensional lattice model, and r_v is the position vector of node v on the plane of the two-dimensional lattice model, both of which are two-dimensional vectors.

[0039] Furthermore, the method for identifying the fixed payment channel combination pattern includes:

[0040] For each aggregation area, construct a binary matrix, denoted as the combination pattern matrix; the rows and columns of the combination pattern matrix correspond to the nodes in the aggregation area respectively; for any two aggregation areas A and B, calculate the pattern similarity between their corresponding combination pattern matrices M_A and M_B; according to the pattern similarity, perform hierarchical clustering analysis on all aggregation areas to obtain several categories, and for each category, select the aggregation area with the highest pattern similarity as the representative area, and the combination pattern of the payment channels in the representative area is the fixed payment channel combination pattern.

[0041] Furthermore, the method for selecting payment channels from the effective configuration combinations corresponding to the payment area for payment processing includes:

[0042] Establish a payment scenario mapping table to map the combination of each payment scenario identifier to the corresponding payment area number, load all the effective configuration combinations corresponding to the payment area according to the payment area number, and for each effective configuration combination, calculate its comprehensive evaluation score in the current payment scenario; according to the comprehensive evaluation score, sort all the effective configuration combinations in descending order, and select the one with the highest comprehensive evaluation score from the sorted effective configuration combinations as the optimal configuration combination, and extract all the payment channel IDs and related configuration parameters included in the optimal configuration combination. These payment channels will be used as candidate channels to process the current payment request.

[0043] Technical effects and advantages of the unified payment cashier desk processing system for Internet medical care according to the present invention:

[0044] The present invention improves the overall performance and reliability of the payment system, significantly enhances the adaptability of the system to complex and changing payment environments; through intelligent channel management and optimized configuration, it greatly improves the payment processing efficiency and success rate, while reducing the operating costs; the adaptive mechanism of the system ensures optimal performance in a dynamic market environment, enhancing the user experience and satisfaction; the design of standardized interfaces significantly reduces the complexity of system integration and maintenance, improving the scalability and flexibility of the system; the comprehensive performance evaluation mechanism provides a scientific basis for decision-making, contributing to the continuous optimization of payment strategies; in addition, the innovative method of the present invention provides a new idea for solving the resource allocation problem in large-scale payment scenarios. Description of the Drawings

[0045] Figure 1 It is a schematic diagram of the unified payment cashier desk processing system for Internet medical care according to the present invention;

[0046] Figure 2 It is a schematic diagram of the unified payment cashier desk processing method for Internet medical care according to the present invention. Detailed Embodiments

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] Embodiment 1

[0049] Please refer to Figure 1 As shown, the unified payment cashier desk processing system for Internet medical care described in this embodiment includes:

[0050] A channel configuration module, used to establish the channel configuration of payment channels; connect the payment channel configuration according to a preset standard interface;

[0051] A channel partition module, used to partition the payment channels to obtain n payment regions; n is an integer greater than 1;

[0052] A configuration module, used to construct an effective configuration combination of payment channels for each payment region;

[0053] The payment aggregation module is used to obtain the combination of payment scenario identifiers in real time, and select a payment channel from the valid configuration combinations in the corresponding payment area for payment processing; each module is connected by wired and / or wireless means to achieve data transmission between modules.

[0054] The ways to establish the channel configuration of the payment channel include:

[0055] Collect and maintain the configuration information of each payment channel, including the payment channel name, payment channel ID, payment channel merchant ID, and payment channel key; define a data structure (such as a database table or key-value pair mapping table) to store the configuration information of the payment channel, and the payment channel ID of each payment channel serves as the identifier of the corresponding payment channel; collect and maintain several payment scenario identifiers, each payment scenario identifier corresponds to a payment scenario, such as institution ID, terminal Key, business type Key, and pharmaceutical company Key, etc. These payment scenario identifiers will be used as key values to correspond to the configuration information of different payment channels; establish a mapping relationship, use the combination of payment scenario identifiers as the key value, and the corresponding payment channel ID as the value, and store it in the data structure.

[0056] Define the rules for dynamically adjusting the mapping relationship, which are used to dynamically adjust the mapping relationship in the data structure, allowing modification, addition, or deletion of the mapping between the institution ID, terminal Key, business type Key, and pharmaceutical company Key and the payment channel ID; that is, the establishment of the payment channel configuration is completed.

[0057] The definition methods of the rules for dynamically adjusting the mapping relationship include:

[0058] Map the payment scenario identifiers (institution ID, terminal Key, business type Key, and pharmaceutical company Key) to a multi-dimensional space, denoted as the payment scenario space, and each point in the payment scenario space represents a specific combination of payment scenarios; collect payment requests and payment data within a fixed historical time; record the payment scenario identifiers and the used payment channel IDs of each payment request, and map the payment data to the corresponding space points according to the payment scenario identifiers.

[0059] It should be noted that payment data usually includes payment order information, payment channel information, payment process information, payment result information, and other auxiliary information; payment order information such as order number, payment amount, payment time, payment scenario, etc.; payment channel information such as the used payment channel ID, payment channel type (such as bank card, third-party payment, etc.), payment channel merchant ID, payment channel configuration parameters, etc.; payment process information such as the time of initiating the payment request, payment channel response time, payment result (success / failure), failure reason code, transaction serial number returned by the payment channel, etc.

[0060] Payment result information such as the actual payment amount, payment completion time, payment channel handling fee, refund flag (whether a refund has occurred), etc.; other auxiliary information such as error logs, call stack information, and network environment information (such as latency, packet loss rate, etc.).

[0061] These payment data are obtained from multiple sources, such as the log files of the payment system, database records, response information returned by payment channels, etc.; by cleaning, transforming, and aggregating these raw payment data, we can extract evaluation metrics such as payment success rate, payment latency, payment error rate, etc., for the comprehensive evaluation and optimization of the configuration combination.

[0062] Construct a scan line in the payment scenario space. The scan line can be a straight line, curve, or other shape, used to traverse the points in the payment scenario space; along the direction of the scan line, traverse each point in the payment scenario space and calculate the evaluation metrics of the payment data at each point for all available payment channels; the calculation formula for the evaluation metrics is:

[0063] Among them, CS is the evaluation metric, PS is the payment success rate, PD is the payment latency, PE is the payment exception rate, PF is the payment fee, PR is the payment refund rate, w1, w2, w3, w4, and w5 are the weight coefficients of each metric, and the values range from 0 to 1, which are set according to actual needs; the payment success rate is the number of successful payment transactions divided by the total number of payment requests, the payment latency is the average value of the difference between all payment completion times and payment request times, the payment exception rate is the number of abnormal payment transactions divided by the total number of payment requests, and the payment refund rate is the number of refund transactions divided by the number of successful payment transactions; actually, set an evaluation period (such as daily, weekly, or monthly), collect payment data within this period, and then calculate the values of each metric according to the formula.

[0064] Preset an evaluation threshold, filter out the payment channels whose evaluation metrics are greater than the evaluation threshold as alternative payment channels. For each point, sort the alternative payment channels in descending order according to the evaluation metrics, and select the highest-ranked one or several payment channels as the recommended payment channels for this point. If there are no alternative payment channels for a certain point, then select the payment channel with the highest evaluation metric, or borrow the recommended payment channel from adjacent points.

[0065] Update the payment channel IDs corresponding to the recommended payment channels for each point into the mapping relationship. The new mapping relationship will be applied to subsequent payment request processing. Regularly or when new payment data arrives, repeat the above steps, and dynamically adjust the recommended payment channels for each point according to the new evaluation metrics. Parameters such as the evaluation threshold and metric weights can be optimized to improve the accuracy of recommendations; and dynamically adjust the scan line during this process.

[0066] Within a predetermined time period (such as daily, weekly, or monthly), new payment requests and payment data are collected to form a payment request dataset D_new; for each point p in the payment scenario space S, calculate the shortest distance d(p, L_old) from p to the original scan line L_old, define a coverage radius r, if d(p, L_old) is less than or equal to r, then the point p is covered by the scan line L_old, and count the number of covered points Nc and the number of uncovered points Nu.

[0067] Map D_new to the payment scenario space S to obtain a new point set P_new, for each point p in S, calculate the nearest neighbor distance d_min(p, P_new) from p to P_new, define a change threshold TL, if d_min(p, P_new) is greater than TL, then the area where the point p is located has changed, and mark the area where the point p is located as the changed area Rc.

[0068] For each point p' in the changed area Rc, calculate the shortest distance d(p, L_old) from p' to the scan line L_old, define an adjustment radius ra, if d(p, L_old) is greater than ra, then use a B-spline curve to fit the points in the changed area Rc to obtain a new scan line segment L_new, and insert the new scan line segment L_new into the scan line L_old to obtain an adjusted scan line Lad.

[0069] Define an expected length, if the length of the adjusted scan line Lad is greater than the expected length, then shorten the length of the adjusted scan line Lad to the expected length; if the length of the adjusted scan line Lad is less than the expected length, then extend the length of the adjusted scan line Lad to the expected length.

[0070] Use the kernel density estimation method to construct a distribution density function f of the scan line L_old in the payment scenario space S, for each point in the point set P_new, calculate its log-likelihood value lu with respect to the current distribution density function f and sum them to obtain a total log-likelihood value LU; use the gradient ascent optimization algorithm to adjust the parameters of the distribution density function f to maximize the total log-likelihood value LU; obtain an adjusted distribution density function f'; resample the adjusted scan line Lad according to the adjusted distribution density function f' to obtain an optimized scan line Lop; subsequently, use the optimized scan line Lop to traverse each point p in S and select a recommended payment channel for p.

[0071] Preset standard interfaces, and for each interface function, formulate unified interface specifications, including interface addresses, request methods, request parameters, response formats, etc.; for example, the specification of the payment order placement interface can be defined as:

[0072] Interface address: https: / / pay.example.com / order (for example only and not runnable);

[0073] Request method: POST;

[0074] Request parameters: including order number, amount, payment method, etc.;

[0075] Response format: JSON format, including payment order number, payment link, etc.;

[0076] For each payment channel to be docked, collect the specific implementation information of its five standard interfaces; this information includes interface address, request method, request parameter format, response format, etc.

[0077] Build an interface adaptation layer for each payment channel to convert the standard interface request into the actual interface request of that channel; in the adaptation layer, according to the standard interface specification and the actual interface information of the payment channel, complete the conversion and formatting of the request parameters; at the same time, it is also necessary to convert the response result of the payment channel into the response format defined by the standard interface specification.

[0078] Integrate the interface adaptation layers of all payment channels into a unified interface gateway; the interface gateway provides five standard interfaces for external systems to call, and internally routes to the corresponding adaptation layer according to the configuration information of the payment channel; configure necessary parameters for each payment channel, such as merchant ID, key, certificate, etc.; these parameters will be used in the adaptation layer to construct the actual interface request of the corresponding channel.

[0079] Through the above process, the interfaces of numerous payment channels are uniformly docked to the preset five standard interfaces, realizing standardized calls for functions such as payment order placement, query, refund, etc.; this standardization is beneficial to the integration of external systems and payment systems, and is also beneficial to subsequent operation and maintenance and optimization work; at the same time, the design of the interface adaptation layer makes the integration of new channels more flexible and extensible.

[0080] The ways to partition payment channels include:

[0081] Extract the key features of the channel configuration of the payment channel, such as supported payment methods, transaction limits, handling fees, etc.; map the key features of each payment channel to a multi-dimensional feature space, where the dimension of the multi-dimensional feature space is m, that is, m key features; each dimension of the multi-dimensional feature space corresponds to a key feature, and in the multi-dimensional feature space, each payment channel corresponds to a feature point.

[0082] For each dimension i (key feature), set a threshold range [low_i, high_i]; where low_i is the lower limit of dimension i and high_i is the upper limit of dimension i. This threshold range indicates that payment channels with similar feature values are expected to be divided into the same region; define a clipping window. In the multi-dimensional feature space, the clipping window is an m-dimensional hyper-rectangle. For each dimension i (key feature), the boundaries of the clipping window in this dimension are determined by [low_i, high_i]; that is, the lower left vertex of the clipping window is low_i and the upper right vertex is high_i.

[0083] For each dimension i, define two code elements 0_i and 1_i. If the coordinate value of the feature point s is less than the lower limit of the clipping window in this dimension, the corresponding code element is 1_i; if the coordinate value of the feature point s is greater than the upper limit of the clipping window in this dimension, the corresponding code element is 0_i; otherwise, the corresponding code element is 0; connect the code elements in all dimensions to obtain an m-bit code. If this code is all 0, the feature point s is completely within the clipping window. If each bit of this code is not 0, the feature point s is completely outside the clipping window; otherwise, it means that the feature point s intersects the boundary of the clipping window.

[0084] For the feature points that intersect the boundary of the clipping window, use the parametric equation to calculate the intersection points of the corresponding feature points and the boundary of the clipping window. The parametric equation can be a straight-line parametric equation; based on the intersection points, divide the corresponding feature points into two new points, which are located inside and outside the clipping window respectively; classify the feature points that are completely within the clipping window into the same region. For adjacent regions, construct the key features within them into vectors, denoted as feature vectors. For any two regions, calculate the similarity (cosine similarity or Euclidean distance) between their feature vectors, set a similarity threshold. If the similarity is greater than or equal to the similarity threshold, merge the corresponding regions into a new region; repeat the merging until no further merging is possible.

[0085] Adjust the position and size of the clipping window and repeat the iteration. Each iteration will obtain a new set of regions until all feature points are assigned to a certain region; output the regions obtained in the last iteration as the final result, that is, obtain n payment regions; each region contains a set of payment channels with similar configurations.

[0086] The construction methods of valid configuration combinations include:

[0087] The payment area is regarded as a two-dimensional lattice model, which consists of a series of grids and grid points (nodes). The shape of the grids can be square, rectangular or other regular shapes; each grid point represents a payment channel within the payment area; there is an interaction between adjacent grid points; the interaction is an attractive or repulsive force, indicating the degree of compatibility or conflict between different payment channels.

[0088] Define the model parameters of the two-dimensional lattice model. The model parameters include the hopping parameter T, the interaction parameter Y, and the chemical potential μ. The hopping parameter controls the hopping intensity between payment channels, indicating the ability of a payment channel to jump from one node to an adjacent node in the lattice system. The larger the value, the stronger the migration ability of the payment channel in the grid. The interaction parameter controls the interaction intensity between payment channels, indicating the magnitude of the attractive or repulsive force between different payment channel nodes. A positive value indicates attraction, and a negative value indicates repulsion. The chemical potential controls the total number of payment channels in the two-dimensional lattice model. The larger the value, the more payment channels can be accommodated in the lattice.

[0089] Define the energy function H, which is used to calculate the total energy of the two-dimensional lattice model.

[0090] H = -T·∑(c + _u·c_v + c + _v·c_u) - Y·∑(n_u·n_v) - μ·∑n_u - z·GH;

[0091] Among them, c + _u is the creation factor of node u, which acts on node u to create a payment channel on node u. c_v is the annihilation factor of node v, which acts on node v to remove the payment channel on node v. c + _v is the creation factor of node v, which acts on node v to create a payment channel on node v. c_u is the annihilation factor of node u, which acts on node u to remove the payment channel on node u. It should be noted that the meaning of c + _u·c_v is to first remove a payment channel on node v and then create this payment channel on node u.

[0092] n_u is the number operator of the payment channels on node u, which acts on node u. The eigenvalue of n_u is 0 or 1, corresponding to whether there is a payment channel on node u or not. 0 indicates that there is no payment channel on the node, and 1 indicates that there is one payment channel on the node. z is the distance attenuation parameter, with a value between 0 and 1, which controls the distance attenuation intensity. GH is the distance attenuation function.

[0093] The distance attenuation function GH = ∑ (u,v)(|r_u - r_v|·n_u·n_v); where r_u is the position vector of node u on the plane of the two-dimensional lattice model, and r_v is the position vector of node v on the plane of the two-dimensional lattice model, both being two-dimensional vectors.

[0094] Randomly assign the payment channels within the payment area to the nodes to form an initial state, ensuring that the initial state satisfies the constraint conditions during this process, such as the total number of payment channels, adjacent compatibility, etc.; use the Metropolis algorithm to sample the initial state and simulate the time evolution of the two-dimensional lattice model; at each step, randomly select a node and change its state (add or remove a payment channel); calculate the difference in the total energy of the two-dimensional lattice model before and after the state change, denoted as the energy difference ΔE, and decide whether to accept the new state (the changed state) with a probability of min(1, exp(-β×ΔE)); where β is the inverse temperature, controlling the degree of thermal fluctuations of the two-dimensional lattice model; at high temperature (low β), it is easier to accept high-energy states.

[0095] Record the change of the total energy with the number of steps (time), gradually increase the value of the inverse temperature to make the two-dimensional lattice model tend to be stable, repeat until the total energy is the lowest, and the two-dimensional lattice model at this time is the ground state configuration; map the ground state configuration onto the two-dimensional plane to show the distribution of payment channels in the two-dimensional lattice model, count the number of payment channels on each node to obtain the overall distribution of the number of payment channels, and based on the distribution of the number of payment channels, perform clustering analysis (such as DBSCAN) on the ground state configuration to identify the aggregation areas of payment channels; within the aggregation areas, identify the fixed payment channel combination patterns, which are the effective configuration combinations.

[0096] Specifically, for each aggregation area, construct a binary matrix, denoted as the combination pattern matrix; the rows and columns of the combination pattern matrix correspond to the nodes within the aggregation area respectively, and this matrix describes the combination pattern of payment channels within the aggregation area; for any two aggregation areas A and B, calculate the pattern similarity between their corresponding combination pattern matrices M_A and M_B, and matrix similarity metrics such as cosine similarity, Jaccard similarity, or edit distance can be used; according to the pattern similarity, perform hierarchical clustering analysis on all aggregation areas, which can be achieved using a hierarchical clustering algorithm, to obtain several categories. The aggregation areas within the same category have highly similar payment channel combination patterns. For each category, select the aggregation area with the highest pattern similarity as the representative area, and the combination pattern of payment channels within the representative area is the fixed payment channel combination pattern.

[0097] Construct a set of optimized and effective payment channel configuration combinations for each payment area; these combinations consider the interactions between payment channels and the overall system performance, providing a good initial solution for subsequent optimization algorithms.

[0098] Establish a payment scenario mapping table to map each combination of payment scenario identifiers to the corresponding payment area number. Load all valid configuration combinations corresponding to the payment area according to the payment area number. For each valid configuration combination, calculate its comprehensive evaluation score in the current payment scenario; the comprehensive evaluation score considers multiple performance indicators such as payment rejection rate, payment retry rate, payment channel utilization rate, and payment amount distribution, and assigns different weights. The payment rejection rate is the number of payment requests rejected by the payment channel divided by the total number of payment requests initiated, which reflects the degree of rejection of payment requests by the payment channel; the payment retry rate is the number of payment requests that need to be retried divided by the total number of payment requests initiated, which reflects the proportion of payment requests that need to be re-initiated; for each payment channel, its utilization rate is the number of orders successfully paid through this channel divided by the total number of successfully paid orders, which reflects the usage frequency and importance of different payment channels; count the quantity distribution of successfully paid orders within different amount ranges, which is the payment amount distribution, and can understand the performance of the system when processing payment requests of different amounts.

[0099] According to the comprehensive evaluation score, sort all valid configuration combinations in descending order. From the sorted valid configuration combinations, select the one with the highest comprehensive evaluation score as the optimal configuration combination. Extract all payment channel IDs and related configuration parameters included in the optimal configuration combination. These payment channels will be used as candidate channels to process the current payment request. According to the extracted payment channel information, construct a payment request message, and call the payment order placement interface of each payment channel in sequence according to the preset standard interface to attempt payment, and record the detailed information of this payment request, including payment channels used, payment results, payment latency and other data; these payment data will be used for subsequent data analysis and model optimization to improve the overall performance of the payment system.

[0100] Find the optimal valid configuration combination from the corresponding payment area, extract the recommended payment channels for payment processing, so as to improve the success rate and efficiency of payment; at the same time, the feedback payment data will also be used to continuously optimize the configuration combination to form a closed-loop adaptive system.

[0101] This embodiment improves the overall performance and reliability of the payment system, significantly enhances the system's adaptability to complex and changing payment environments; through intelligent channel management and optimized configuration, it greatly improves the payment processing efficiency and success rate, while reducing operating costs; the system's adaptive mechanism ensures optimal performance in a dynamic market environment, enhancing user experience and satisfaction; the design of standardized interfaces significantly reduces the complexity of system integration and maintenance, improving the system's scalability and flexibility; a comprehensive performance evaluation mechanism provides a scientific basis for decision-making, contributing to the continuous optimization of payment strategies; in addition, the innovative method of the present invention provides a new idea for solving the resource allocation problem in large-scale payment scenarios.

[0102] Embodiment 2

[0103] Please refer to Figure 2 As shown, for the parts not described in detail in this embodiment, refer to the description in Embodiment 1. A unified payment cashier desk processing method for Internet medical care is provided, including:

[0104] Step 1: Establish the channel configuration of the payment channels; dock the payment channel configuration according to the preset standard interface;

[0105] Step 2: Partition the payment channels to obtain n payment regions;

[0106] Step 3: For each payment region, construct an effective configuration combination of the payment channels;

[0107] Step 4: Obtain the combination of payment scene identifiers in real time, and select a payment channel from the effective configuration combination of the corresponding payment region for payment processing.

[0108] Embodiment 3

[0109] This embodiment discloses and provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it realizes the operation mode of the above-mentioned unified payment cashier desk processing method for Internet medical care.

[0110] Since the electronic device introduced in this embodiment is the electronic device adopted for implementing the unified payment cashier desk processing method for Internet medical care in the embodiments of the present application, based on the unified payment cashier desk processing method for Internet medical care introduced in the embodiments of the present application, those skilled in the art can understand the specific implementation manners and various variations of the electronic device in this embodiment. Therefore, the specific implementation of how this electronic device realizes the method in the embodiments of the present application will not be described in detail here. As long as those skilled in the art implement the electronic device adopted for the unified payment cashier desk processing method for Internet medical care in the embodiments of the present application, it falls within the scope of protection of the present application.

[0111] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0112] The above are only the preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the idea of the present invention belong to the protection scope of the present invention. It should be noted that for ordinary users in the technical field, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of the present invention.

Claims

1. A unified payment cashier processing system for Internet medical care, characterized by: include: The channel configuration module is used to establish the channel configuration of the payment channel; the payment channel configuration is connected according to the preset standard interface; The channel partitioning module is used to partition the payment channel to obtain n payment areas; Configuration module, used to build effective configuration combinations of payment channels for each payment area; The payment aggregation module is used to obtain the combination of payment scenario identifiers in real time and select a payment channel from the valid configuration combination of the corresponding payment area for payment processing; each module is connected by wired and / or wireless means.

2. The unified payment cashier processing system for Internet medical care according to claim 1 is characterized in that: The channel configuration method for establishing the payment channel includes: Collect and maintain configuration information for each payment channel, including payment channel name, payment channel ID, payment channel merchant ID, and payment channel key; Define a data structure for storing the configuration information of payment channels, with the payment channel ID of each payment channel as the identifier of the corresponding payment channel; collect and maintain several payment scenario identifiers, each of which corresponds to a payment scenario, and these payment scenario identifiers will be used as key values ​​to correspond to the configuration information of different payment channels; establish a mapping relationship, with the combination of payment scenario identifiers as the key value and the corresponding payment channel ID as the value, and store them in the data structure; Define the rules for dynamically adjusting the mapping relationship, which is used to dynamically adjust the mapping relationship in the data structure, that is, the payment channel configuration is established.

3. The unified payment cashier processing system for Internet medical care according to claim 2 is characterized in that: Methods for defining dynamically adjusting mapping relationship rules include: Map the payment scenario identifier to a multidimensional space, recorded as the payment scenario space, where each point in the payment scenario space represents a specific combination of payment scenarios; collect payment requests and payment data within a fixed period of time in history; record the payment scenario identifier and the payment channel ID used for each payment request, and map the payment data to the corresponding space point according to the payment scenario identifier; construct a scan line in the payment scenario space; Along the direction of the scan line, traverse each point in the payment scenario space and calculate the evaluation index of the payment data at each point in all available payment channels; the calculation formula of the evaluation index is: Among them, CS is the evaluation index, PS is the payment success rate, PD is the payment delay, PE is the payment abnormality rate, PF is the payment fee, PR is the payment refund rate, w1, w2, w3, w4 and w5 are the weight coefficients of each index, and the value is between 0 and 1; An evaluation threshold is preset, and payment channels with evaluation indicators greater than the evaluation threshold are screened out as backup payment channels. For each point, the backup payment channels are sorted in descending order according to the evaluation indicators, and the highest-ranked one or several payment channels are selected as the recommended payment channels for the point. If a point does not have any backup payment channels, the payment channel with the highest evaluation indicator is selected, or the recommended payment channel is borrowed from the adjacent point; the payment channel ID corresponding to the recommended payment channel of each point is updated to the mapping relationship, and the scan line is dynamically adjusted in the process.

4. The unified payment cashier processing system for Internet medical care according to claim 3 is characterized in that: The method of dynamically adjusting the scan line includes: Within a predetermined time period, new payment requests and payment data are collected to form a payment request data set D_new; for each point p in the payment scene space S, the shortest distance d(p, L_old) from p to the original scan line L_old is calculated, and the coverage radius r is defined. If d(p, L_old) is less than or equal to r, the point p is covered by the scan line L_old, and the number of covered points Nc and the number of uncovered points Nu are counted; Map D_new to the payment scenario space S to obtain a new point set P_new. For each point p in S, calculate the nearest neighbor distance d_min(p, P_new) from p to P_new, and define the change threshold TL. If d_min(p, P_new) is greater than TL, the area where the marked point p is located is the change area Rc. For each point p' in the change region Rc, calculate the shortest distance d(p, L_old) from p' to the scan line L_old, define the adjustment radius ra, if d(p, L_old) is greater than ra, use the Bezier curve to fit the points in the change region Rc, get the new scan line segment L_new, insert the new scan line segment L_new into the scan line L_old, and get the adjusted scan line Lad; Define an expected length, if the length of the adjusted scanning line Lad is greater than the expected length, shorten the length of the adjusted scanning line Lad to the expected length; if the length of the adjusted scanning line Lad is less than the expected length, extend the length of the adjusted scanning line Lad to the expected length; Use the kernel density estimation method to construct the distribution density function f of the scan line L_old in the payment scene space S. For each point in the point set P_new, calculate its logarithmic likelihood value lu to the current distribution density function f and sum them to obtain the total logarithmic likelihood value LU; use the gradient ascent optimization algorithm to adjust the parameters of the distribution density function f to maximize the total logarithmic likelihood value LU; obtain the adjusted distribution density function f'; resample the adjusted scan line Lad according to the adjusted distribution density function f' to obtain the optimized scan line Lop; subsequently apply the optimized scan line Lop to traverse each point p in S and select the recommended payment channel for p.

5. The unified payment cashier processing system for Internet medical care according to claim 4 is characterized in that: The methods for partitioning payment channels include: Extract key features of the channel configuration of the payment channel; map the key features of each payment channel into a multidimensional feature space, where the dimension of the multidimensional feature space is m, i.e., m key features; each dimension of the multidimensional feature space corresponds to a key feature, and in the multidimensional feature space, each payment channel corresponds to a feature point; For each dimension i, set a threshold range [low_i, high_i]; where low_i is the lower limit of dimension i, and high_i is the upper limit of dimension i; define a cropping window, which is an m-dimensional hyperrectangle in the multidimensional feature space. For each dimension i, the boundary of the cropping window in this dimension is determined by [low_i, high_i]; that is, the lower left vertex of the cropping window is low_i, and the upper right vertex is high_i; for each dimension i, define two code elements 0_i and 1_i. If the coordinate value of the feature point s is less than the lower limit of the cropping window in this dimension, the corresponding code element is 1_i; if the coordinate value of the feature point s is greater than the upper limit of the cropping window in this dimension, the corresponding code element is 0_i; otherwise, the corresponding code element is 0; Connect the code elements on all dimensions to obtain an m-bit code. If the code is all 0, the feature point s is completely within the cropping window. If each bit of the code is not 0, the feature point s is completely outside the cropping window. Otherwise, it means that the feature point s intersects with the boundary of the cropping window. For feature points that intersect with the boundary of the cropping window, use the parametric equation to calculate the intersection of the corresponding feature point and the boundary of the cropping window, and split the corresponding feature point into two new points based on the intersection, which are located inside and outside the cropping window respectively; the feature points that are completely within the cropping window are classified into the same area, and for adjacent areas, the key features therein are constructed into vectors, recorded as feature vectors, and for any two areas, the similarity between their feature vectors is calculated, and a similarity threshold is set. If the similarity is greater than or equal to the similarity threshold, the corresponding areas are merged into a new area; the merging is repeated until the merging cannot be continued; the position and size of the cropping window are adjusted, and the iteration is repeated. A new set of areas will be obtained in each iteration until all feature points are assigned to a certain area; the area obtained from the last iteration is output as the final result, that is, n payment areas are obtained.

6. The unified payment cashier processing system for Internet medical care according to claim 5 is characterized in that: The construction method of the effective configuration combination includes: The payment area is regarded as a two-dimensional lattice model, which consists of a series of grids and nodes, and each node represents a payment channel in the payment area; the model parameters of the two-dimensional lattice model are defined, and the model parameters include the jump parameter T, the interaction parameter Y and the chemical potential μ; Define the energy function H, which is used to calculate the total energy of the two-dimensional lattice model; The payment channels in the payment area are randomly assigned to nodes to form the initial state; the initial state is sampled using the Metropolis algorithm; at each step, a node is randomly selected and the state of the node is changed; the difference in the total energy of the two-dimensional lattice model before and after the state change is calculated, recorded as the energy difference ΔE, and whether to accept the new state is determined with probability min(1, exp(-β×ΔE)); where β is the inverse temperature; the total energy is recorded as the number of steps changes, and the value of the inverse temperature is gradually increased to stabilize the two-dimensional lattice model, and this process is repeated until the total energy is the lowest. The two-dimensional lattice model at this time is the ground state configuration; the ground state configuration is mapped to a two-dimensional plane, and the number of payment channels on each node is counted to obtain the overall distribution of the number of payment channels. Based on the distribution of the number of payment channels, the ground state configuration is clustered and analyzed to identify the clustering area of ​​the payment channels; within the clustering area, a fixed payment channel combination pattern is identified, which is the effective configuration combination.

7. The unified payment cashier processing system for Internet medical care according to claim 6 is characterized in that: The formula of the energy function is: H=-T·∑(c + _u·c_v+c + _v·c_u)-Y·∑(n_u·n_v)-μ·∑n_u-z·GH; Among them, c + _u is the creation factor of node u, which acts on node u to create a payment channel on node u. c_v is the elimination factor of node v, which acts on node v to remove the payment channel on node v. + _v is the creation factor of node v. It acts on node v to create a payment channel on node v. c_u is the elimination factor of node u. It acts on node u to remove the payment channel on node u. n_u is the number operator of payment channels on node u. It acts on node u. The eigenvalue of n_u is 0 or 1, corresponding to whether there is a payment channel on node u. z is the distance decay parameter, which takes a value between 0 and 1. GH is the distance decay function. Distance decay function GH = ∑ (u,v) (|r_u-r_v|·n_u·n_v); where r_u is the position vector of node u on the plane where the two-dimensional lattice model is located, and r_v is the position vector of node v on the plane where the two-dimensional lattice model is located, both of which are two-dimensional vectors.

8. The unified payment cashier processing system for Internet medical care according to claim 7 is characterized in that: The method of identifying the fixed payment channel combination pattern includes: For each clustering area, a binary matrix is ​​constructed, which is recorded as the combination pattern matrix; the rows and columns of the combination pattern matrix correspond to the nodes in the clustering area respectively; for any two clustering areas A and B, the pattern similarity between their corresponding combination pattern matrices M_A and M_B is calculated; based on the pattern similarity, a hierarchical clustering analysis is performed on all clustering areas to obtain several categories. For each category, the clustering area with the highest pattern similarity is selected as the representative area, and the combination pattern of payment channels in the representative area is the fixed payment channel combination pattern.

9. The unified payment cashier processing system for Internet medical care according to claim 8, characterized in that: The method of selecting a payment channel from a valid configuration combination corresponding to the payment area for payment processing includes: A payment scenario mapping table is established to map each combination of payment scenario identifiers to the corresponding payment area number. All valid configuration combinations corresponding to the corresponding payment area are loaded according to the payment area number. For each valid configuration combination, its comprehensive evaluation score in the current payment scenario is calculated. According to the comprehensive evaluation score, all valid configuration combinations are sorted in descending order. From the sorted valid configuration combinations, the one with the highest comprehensive evaluation score is selected as the optimal configuration combination. All payment channel IDs and related configuration parameters contained in the optimal configuration combination are extracted from the optimal configuration combination. These payment channels will be used as candidate channels to process the current payment request.

Images