A method, device and equipment for determining a registration priority

Abstract

The application relates to a method, device and equipment for determining a registration priority. The method comprises the following steps: acquiring multi-source data; wherein the multi-source data at least comprises user registration data and a risk coefficient of an external risk control interface; establishing a user three-dimensional label system based on the multi-source data; obtaining an image comprehensive evaluation result based on the multi-source data and the user three-dimensional label system; and determining the registration priority of the user based on the image comprehensive evaluation result and the multi-source data. The application can distinguish the actual medical demand differences of different users, thereby improving the rationality of resource allocation.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and more specifically to a method, apparatus, and device for determining registration priority. Background Technology

[0002] Currently, online appointment registration systems are widely used in various medical institutions. In these systems, the waiting list mechanism allocates appointment slots based solely on the order in which users submit their waiting list applications.

[0003] However, different users have different levels of urgency of illness, different past medical treatment behaviors, and different needs for medical treatment in different places. Simply relying on the order of queuing to allocate and release appointment slots results in poor rationality in the allocation of medical resources.

[0004] Therefore, there is an urgent need for a method to improve the rationality of medical resource allocation. Summary of the Invention

[0005] The technical problem that this invention aims to solve is that relying solely on the order of queuing to allocate and release appointment slots results in a poor rationality in the allocation of medical resources.

[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: a method for determining registration priority, the method comprising: acquiring multi-source data; wherein, the multi-source data includes at least user registration data and risk coefficients of external risk control interfaces; establishing a three-dimensional user tag system based on the multi-source data; obtaining a comprehensive profile evaluation result based on the multi-source data and the user three-dimensional tag system; and determining the user's registration priority based on the comprehensive profile evaluation result and the multi-source data.

[0007] The beneficial effects of this invention are as follows: By acquiring multi-source data such as user registration data and external risk control interface risk coefficients, and constructing a three-dimensional user tag system, and then combining the multi-source data with the tag system to obtain a comprehensive profile evaluation result, the user's registration priority is finally determined based on the comprehensive profile evaluation result and the multi-source data. The overall determination of registration priority is based on the comprehensive evaluation result formed by the integration of multi-dimensional data, rather than simply using the order of submission of waiting list applications as the sole criterion. This breaks away from the single queuing order judgment logic, thereby being able to distinguish the differences in the actual medical needs of different users and improve the problem of unreasonable resource allocation caused by allocating appointment slots solely based on the order of submission.

[0008] Based on the above technical solution, the present invention can be further improved as follows.

[0009] Furthermore, the multi-source data also includes operator data and geographic data. Based on the multi-source data, a three-dimensional user tagging system is established, including: extracting and generating basic attribute tags, health status tags, and platform behavior feature tags based on user registration data; extracting and generating medical treatment preference tags based on user registration data, operator data, and geographic data; and establishing a three-dimensional user tagging system based on basic attribute tags, health status tags, platform behavior feature tags, and medical treatment preference tags.

[0010] The beneficial effect of adopting the above-mentioned further solution is that it extracts and generates basic attribute tags, health status tags, and platform behavior feature tags from user registration data, and then combines user registration data, operator data, and geographic data to extract and generate medical preference tags. By integrating these four types of tags from different dimensions, a three-dimensional user tag system can be established. This system can accurately classify tag categories according to different data sources, fully aggregate user-related information corresponding to various tags, and make the information dimensions of the constructed three-dimensional user tag system clearly defined. It fully integrates user information corresponding to different data sources and achieves comprehensive organization and sorting of user-related information.

[0011] Furthermore, the multi-source data also includes operator data and geographic data. Based on the multi-source data and the user three-dimensional tagging system, a comprehensive profile evaluation result is obtained, including: multi-source feature fusion processing of user registration data, operator data, and geographic data to obtain a fused feature vector; vector construction processing of the fused feature vector to obtain a user feature vector; user group clustering processing of the user feature vector to obtain a first processing result; no-show risk prediction processing of the user feature vector to obtain a second processing result; medical treatment demand prediction processing of the user feature vector to obtain a third processing result; feature association rule mining processing of the user feature vector to obtain a fourth processing result; and a comprehensive profile evaluation result based on the first, second, third, and fourth processing results.

[0012] The beneficial effect of adopting the above-mentioned further scheme is that by performing multi-source feature fusion processing on user registration data, operator data, and geographic data to obtain fused feature vectors, and then performing vector construction processing on the fused feature vectors to obtain user feature vectors, the user feature vectors are then subjected to user group clustering processing, no-show risk prediction processing, medical treatment demand prediction processing, and feature association rule mining processing to obtain four corresponding processing results. Finally, based on the four processing results, a comprehensive profile evaluation result is obtained, which can orderly complete the feature integration and standardized transformation of multiple types of data.

[0013] Furthermore, based on the comprehensive assessment results of the user profile and multi-source data, the user's registration priority is determined, including: obtaining the number of available appointment slots and the number of people on the waiting list corresponding to the user's registration data; determining the appointment slot scarcity coefficient based on the number of available appointment slots and the number of people on the waiting list; determining the user's waiting list order and waiting time based on the user's registration data; determining the user's corresponding first target score based on the user's waiting list order, multi-source data, and the risk coefficient of the external risk control interface; and determining the user's registration priority based on the first target score, waiting time, comprehensive assessment results of the user profile, risk coefficient of the external risk control interface, and appointment slot scarcity coefficient.

[0014] The beneficial effect of adopting the above-mentioned further scheme is that by obtaining the number of available appointment slots and the number of people on the waiting list corresponding to the user's registration data, the scarcity coefficient of appointment slots can be determined. At the same time, the waiting order and waiting time of users can be determined based on the user's registration data. Then, by combining the user's waiting order, multi-source data, and the risk coefficient of the external risk control interface, the first target score corresponding to the user can be determined. Finally, by combining the first target score, waiting time, comprehensive profile evaluation results, risk coefficient of the external risk control interface, and scarcity coefficient of appointment slots, the user's registration priority can be determined. This can integrate the supply and demand of appointment slots, user queuing information, risk control information, and comprehensive user evaluation information to carry out priority determination, making the basis for determining registration priority rich and comprehensive, thereby effectively improving the rationality of medical resource allocation.

[0015] Furthermore, based on the first target score, waiting time, comprehensive profile assessment results, risk coefficient of the external risk control interface, and scarcity coefficient of appointment slots, the user's registration priority is determined, including: if the scarcity coefficient of appointment slots is greater than the scarcity threshold, the preset first weight corresponding to the waiting time is increased to obtain the first target weight, and the preset third weight corresponding to the risk coefficient is decreased to obtain the third target weight; wherein, the first proportion of increasing the preset first weight and the second proportion of decreasing the preset third weight are the same; based on the waiting time and the first target weight, the user's corresponding first target time base score is determined; based on the risk coefficient and the third target weight, the user's corresponding first target risk score is determined; based on the comprehensive profile assessment results and the preset second weight, the profile feature score is determined; based on the first target score, the first target time base score, the profile feature score, and the first target risk score, the user's registration priority is determined; if the scarcity coefficient of appointment slots is not greater than the scarcity threshold, the initial time base score is determined based on the waiting time and the preset first weight; based on the risk coefficient and the preset third weight, the initial risk score is determined; based on the first target score, the initial time base score, the profile feature score, and the initial risk score, the user's registration priority is determined.

[0016] The beneficial effect of adopting the above-mentioned further scheme is that when the scarcity coefficient of the appointment slots is greater than the scarcity threshold, the preset first weight corresponding to the waiting time is increased by the same proportion and the preset third weight corresponding to the risk coefficient is decreased, so as to obtain the first target weight and the third target weight respectively. Then, the first target time base score, the first target risk score and the profile feature score are calculated in sequence, and the user's registration priority is determined by combining the first target score.

[0017] Furthermore, when the scarcity coefficient of appointment slots is not greater than the scarcity threshold, the original preset weights are used to calculate the initial time base score and initial risk score, and the registration priority is determined in combination with the corresponding content. The weight of different scoring dimensions can be flexibly adjusted according to the tightness of appointment slot supply and demand, and the score calculation can be adapted to different appointment slot supply and demand scenarios. This allows the registration priority to be reasonably determined in line with the current tightness of appointment slot resources, thereby effectively improving the rationality of medical resource allocation.

[0018] Furthermore, based on the comprehensive assessment results of the user profile and multi-source data, the user's registration priority is determined, including: determining the user's waiting list order and waiting time based on the user's registration data; determining the user's first target score based on the user's waiting list order, multi-source data, and risk coefficients from external risk control interfaces; if the waiting time exceeds a time threshold, increasing the preset first weight corresponding to the waiting time to obtain the fourth target weight; decreasing the preset second weight corresponding to the comprehensive assessment results to obtain the fifth target weight; decreasing the preset third weight corresponding to the risk coefficient to obtain the sixth target weight; wherein, the third proportion of increasing the preset first weight, the fourth proportion of decreasing the preset second weight, and the fifth proportion of decreasing the preset third weight are the same; determining the second target time base score based on the waiting time and the fourth target weight; determining the target profile feature score based on the comprehensive assessment results and the fifth target weight; determining the user's second target risk score based on the risk coefficient and the sixth target weight; and determining the user's registration priority based on the first target score, the second target time base score, the target profile feature score, and the second target risk score.

[0019] The beneficial effect of adopting the above-mentioned further solution is that it determines the user's waiting order and waiting time based on user registration data. Combining the user's waiting order, multi-source data, and risk coefficients from external risk control interfaces, it determines the user's corresponding first target score. When the waiting time exceeds the time threshold, the preset first weight corresponding to the waiting time is increased by the same proportion, while the preset second weight corresponding to the comprehensive profile assessment result and the preset third weight corresponding to the risk coefficient are simultaneously decreased. After obtaining the weights of each target in sequence, the corresponding second target time base score, target profile feature score, and second target risk score are calculated. Then, based on the various scores, the user's registration priority is determined. This approach can uniformly and standardizedly adjust the weights of each dimension for users who have been on the waiting list for a long time, keeping the weight increase and decrease ratios consistent. It can reasonably tilt the weights towards users with longer waiting times. The score calculation is completed based on the adjusted weights, so that the final determined registration priority can take into account the user's actual waiting time, thereby improving the rationality of medical resource allocation.

[0020] Furthermore, the method also includes: when it is detected that a user's registration slot is released, the waiting queue is reordered based on the target registration priority, and the user with the highest target registration priority is selected to complete the registration slot allocation; wherein, the target registration priority is the registration priority after updating the user's registration priority, and the waiting queue is the queue obtained by sorting the registration priorities in descending order.

[0021] The beneficial effect of adopting the above-mentioned further scheme is that when a user's appointment slot is detected to be released, the waiting queue is reordered based on the target registration priority. This waiting queue itself is formed based on the registration priority in descending order. Then, the user with the highest target registration priority is selected to complete the appointment slot allocation. This can update the queue order according to priority as soon as the appointment slot is released, ensuring that the appointment slot allocation is always based on the priority, allowing users with higher priority to obtain appointment slots first, ensuring the rationality and orderliness of appointment slot allocation, and avoiding chaos in appointment slot allocation.

[0022] Furthermore, the method also includes: determining whether a target event has been detected; wherein the target event includes at least one of the following: a number release event, a user profile change event, and an external risk control score update event; if the target event is detected, recalculating the registration priority of the user corresponding to the target event to obtain the target registration priority; and / or, based on a preset time window, updating the user's waiting time, the preset first weight corresponding to the waiting time, the preset second weight corresponding to the profile comprehensive evaluation result, and the preset third weight corresponding to the risk coefficient; determining the target registration priority based on the updated first weight, the updated second weight, and the updated third weight; wherein the target registration priority is the registration priority after updating the user's registration priority.

[0023] The beneficial effect of adopting the above-mentioned further solution is that by determining whether target events including number release events, user profile change events, and external risk control score update events are detected, the registration priority of the user corresponding to the target event is recalculated when the corresponding target event is detected to obtain the target registration priority. At the same time, the user waiting time and the preset weights corresponding to the three different dimensions can be uniformly updated according to the preset time window, and the target registration priority is determined according to the updated weights. This can realize an operation mode that combines event-triggered targeted priority recalculation with timed batch parameter updates, and can timely follow various changes to synchronously correct the user registration priority, ensuring that the obtained target registration priority has real-time performance and accuracy.

[0024] Furthermore, the present invention provides a device for determining registration priority, the device comprising: an acquisition module for acquiring multi-source data; wherein the multi-source data includes at least user registration data and risk coefficients of external risk control interfaces; an establishment module for establishing a user three-dimensional tag system based on the multi-source data; a first determination module for obtaining a comprehensive profile evaluation result based on the multi-source data and the user three-dimensional tag system; and a second determination module for determining the user's registration priority based on the comprehensive profile evaluation result and the multi-source data.

[0025] Furthermore, this application provides a computer device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the registration priority determination method of the first aspect or any corresponding embodiment described above.

[0026] Furthermore, this application provides a computer-readable storage medium storing computer instructions for causing a computer to execute the registration priority determination method of the first aspect or any corresponding embodiment described above.

[0027] Furthermore, this application provides a computer program product, including computer instructions, which are used to cause a computer to execute the registration priority determination method of the first aspect described above or any corresponding embodiment thereof. Attached Figure Description

[0028] Figure 1 A flowchart illustrating the method for determining registration priority provided by the present invention; Figure 2 A flowchart illustrating another method for determining registration priority provided by the present invention; Figure 3 A flowchart illustrating another method for determining registration priority provided by the present invention; Figure 4A flowchart illustrating the method for determining registration priority provided by the present invention; Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0029] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0030] Currently, online appointment registration systems are widely used in various medical institutions. In these systems, the waiting list mechanism allocates appointment slots based solely on the order in which users submit their waiting list applications.

[0031] However, different users have different levels of urgency of illness, different past medical treatment behaviors, and different needs for medical treatment in different places. Simply relying on the order of queuing to allocate and release appointment slots results in poor rationality in the allocation of medical resources.

[0032] Based on this, the present invention provides a method for determining registration priority. By acquiring multi-source data such as user registration data and risk coefficients from external risk control interfaces, a three-dimensional user tag system is constructed. Then, by combining the multi-source data with the tag system, a comprehensive profile evaluation result is obtained. Finally, the user's registration priority is determined by the comprehensive profile evaluation result and the multi-source data. The overall determination of registration priority is based on the comprehensive evaluation result formed by the integration of multi-dimensional data, rather than simply using the order of submission of waiting list applications as the sole criterion. This breaks away from the single queuing order determination logic, thereby distinguishing the differences in the actual medical needs of different users and improving the problem of unreasonable resource allocation caused by allocating appointment slots solely based on the order of submission.

[0033] According to an embodiment of the present invention, a method for determining registration priority is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0034] This embodiment provides a method for determining registration priority, which can be used in computer devices such as computers and servers. Figure 1 This is a flowchart of a method for determining registration priority according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps: Step S101: Obtain multi-source data; wherein, the multi-source data includes at least user registration data and risk coefficients from external risk control interfaces.

[0035] Multi-source data can be understood as various business data and interface data related to user registration services. This multi-source data includes at least user registration data and risk coefficients from external risk control interfaces.

[0036] Optionally, multi-source data may also include carrier data, geographic data, etc.

[0037] The user registration data can include appointment records (such as the hospital, department, doctor, time slot, and appointment channel), registration information (such as registration type, appointment level, and payment method), medical feedback (such as completion rate, no-show records, and platform rating), and user account data (such as registration time, login frequency, and type of bound document), which are used to depict the patient's health status and medical needs.

[0038] Operator data can include user location trajectories, APP usage behavior, device characteristics, communication frequency, and network activity periods, which can be used to help analyze the correlation between patients' lifestyle patterns and health.

[0039] Geographic data can include residential coordinates, workplace coordinates, activity radius, regional environmental exposure, accessibility to medical resources, and local disease prevalence characteristics, which can be used to empower spatial dimension profiling.

[0040] The risk coefficient of the external risk control interface can include the risk probability output by the third-party scalper identification model, which can be used as an input parameter for priority assessment.

[0041] Specifically, it can connect to the hospital's registration business database, the operator's data storage terminal, the geographic information data terminal, and the external risk control service interface to collect multi-source data related to user registration in batches. The collected data includes at least user registration data and risk coefficients output by the external risk control interface, while simultaneously acquiring operator data and geographic data.

[0042] As an example, an event-triggered data collection method can be adopted, that is, whenever a user submits a registration waiting list application, modifies medical information, or cancels a registration order, the full retrieval of multi-source data is immediately triggered synchronously, and the acquisition of full-dimensional data of a single user is completed in real time.

[0043] Step S102: Establish a three-dimensional user tagging system based on multi-source data.

[0044] The user three-dimensional tagging system can be understood as a multi-dimensional user feature tagging architecture that integrates four types of tags: basic attribute tags, health status tags, platform behavior feature tags, and medical treatment preference tags, based on multi-source data.

[0045] Specifically, after determining the multi-source data, a three-dimensional user tagging system can be established based on the multi-source data.

[0046] As an example, all preprocessed multi-source data is split into data fields and effective features are accurately extracted. Based on user registration data, basic attribute tags, health status tags, and platform behavior feature tags are extracted and generated for users. Then, user registration data, collected operator data, and geographic data are combined to extract user medical preference tags. The extracted basic attribute tags, health status tags, platform behavior feature tags, and medical preference tags are uniformly summarized and integrated, and arranged in an orderly manner according to tag categories to obtain a three-dimensional user tag system.

[0047] As an example, data from multiple sources, such as user registration data, operator data, and geographic data, are integrated to form a "data puzzle" covering the entire process of patient medical treatment, providing comprehensive support for profile construction. This data is then synchronized to the profile system in real time through the platform's API interface to form the data foundation for user profiles. Based on this data foundation, a three-dimensional user tagging system is then established.

[0048] Step S103: Based on multi-source data and a three-dimensional user tagging system, a comprehensive evaluation result of the profile is obtained.

[0049] The comprehensive assessment result of the profile can be understood as a comprehensive evaluation result obtained after data processing and feature analysis, which can comprehensively assess the user's medical needs, medical qualifications, and performance.

[0050] Specifically, after determining the multi-source data and the user three-dimensional tagging system, the comprehensive evaluation results of the user profile can be further determined.

[0051] As an example, the collected user registration data, operator data, and geographic data are retrieved and processed for multi-source feature fusion to generate corresponding fused feature vectors. Then, the fused feature vectors are processed for standardized vector construction to obtain user feature vectors in a standard format. The user feature vectors are then processed sequentially to obtain the first processing result: user group clustering, no-show risk prediction, medical treatment demand prediction, and feature association rule mining. Finally, the first, second, third, and fourth processing results are integrated and summarized to obtain the comprehensive profile evaluation result.

[0052] As an example, a horizontal benchmarking comparison method is used to compare the current user tag data, multi-source data, and standard medical user data within the platform, and generate corresponding profile comprehensive evaluation results based on the degree of difference in the comparison.

[0053] Step S104: Based on the comprehensive assessment results of the user profile and multi-source data, determine the user's registration priority.

[0054] A user's registration priority can be understood as a quantitative evaluation score used to distinguish the order in which all users in the waiting queue obtain and release appointment slots.

[0055] Specifically, the generated profile comprehensive evaluation results are retrieved, and combined with the information on the supply and demand of appointment slots, user queuing data, risk control data, etc. from all the retained multi-source data, quantitative scoring is completed according to the preset scoring rules and weight allocation rules, and finally the registration priority corresponding to each candidate user is calculated.

[0056] As an example, the comprehensive assessment results of the profile are divided into different level intervals with multi-source data, and the registration priority of the corresponding level is directly matched according to the interval.

[0057] As an example, a basic priority is initially generated based on two types of data, and then the back-end administrators fine-tune and correct it according to the actual medical treatment scenario to determine the final registration priority.

[0058] In a feasible scenario, for the release of appointment slots after the daily outpatient appointment slots are released and within 24 hours before the appointment (such as users canceling their registrations, doctors adding appointment slots, and the system releasing idle appointment slots), the system can accurately sort waiting list users and intelligently allocate appointment slots. Specific feasible scenarios are as follows: The system connects to the hospital's registration database, carrier data interfaces, geographic information interfaces, and external risk control interfaces to obtain registration data, carrier data (including users' recent travel history, call records, and related medical information), geographic data (including the distance between the user's current location and the hospital, and transportation convenience), and risk coefficients from the external risk control interfaces for all waiting users. This process involves collecting, cleaning, and formatting the multi-source data. Subsequently, based on the collected multi-source data, the system extracts basic user attribute tags (age, gender, patient identity), health status tags (past medical history, urgency of the visit), and platform behavioral characteristic tags (past registration frequency, no-shows). The system records user information and medical preference tags (frequently visited departments, preferred appointment times), building a complete three-dimensional user tagging system. Next, it performs feature fusion and vector construction on multi-source data to generate user feature vectors. This is followed by user group clustering (differentiating between chronic disease follow-up visits, emergency general visits, and routine visits), no-show risk prediction, medical demand prediction, and feature association rule mining. The results of these four processing methods are integrated to obtain a comprehensive user profile assessment. Based on the comprehensive profile assessment and multi-source data, the user's registration priority is determined. The waiting list is then reordered based on real-time registration priority, and the user with the highest priority is selected for appointment allocation.

[0059] The method for determining registration priority provided by this invention acquires multi-source data such as user registration data and risk coefficients from external risk control interfaces, and constructs a three-dimensional user tag system. Then, it combines the multi-source data with the tag system to obtain a comprehensive profile evaluation result. Finally, it relies on the comprehensive profile evaluation result and multi-source data to jointly determine the user's registration priority. The overall determination of registration priority is based on the comprehensive evaluation result formed by the integration of multi-dimensional data, rather than simply using the order of submission of waiting list applications as the sole criterion. This breaks away from the single queuing order determination logic, thereby being able to distinguish the differences in the actual medical needs of different users and improve the problem of unreasonable resource allocation caused by allocating appointment slots solely based on the order of submission.

[0060] The aforementioned multi-source data also includes operator data and geographic data. Based on step S102 of this embodiment, a user three-dimensional tagging system is established based on the multi-source data, such as... Figure 2 Implementation of steps S2021 to S2023: Step S2021: Based on user registration data, extract and generate basic attribute tags, health status tags, and platform behavior feature tags.

[0061] Basic attribute tags can be understood as tags that represent basic user information extracted from user registration data. These tags may include age, gender, marital status, whether the user sought medical treatment outside the local area, and whether they have medical insurance.

[0062] Health status tags can be understood as tags extracted from user registration data that reflect the user's health status and the urgency of their medical needs. These tags can include disease characteristics (tumor, respiratory issues, etc.); health status (physical examination, chronic disease, pregnant women, elderly, children); and urgency indicators (automatically assigned based on the department of registration or historical diagnostic tags).

[0063] Platform behavioral characteristic tags can be understood as tags extracted from user registration data that reflect users' usage habits and fulfillment of obligations on the registration platform. These tags may include appointment frequency, user-patient relationship, number of days prior to appointment, and frequency of follow-up visits.

[0064] Specifically, user registration data is retrieved, and the data is split into fields and filtered for effective features. Three types of core tags are extracted from the user registration data. When extracting basic attribute tags, the user's age, gender, ID number, and related household registration information, as well as the patient's identity (general patient, chronic disease patient, elderly patient, etc.), are extracted from the user registration data to generate corresponding basic attribute tags. When extracting health status tags, tags reflecting the user's health status (such as chronic disease follow-up, emergency, general visit, severe case follow-up, etc.) are generated based on the user's past medical records, medical history information, departments visited (such as emergency department, chronic disease department), and symptom descriptions from the user registration data. When extracting platform behavior feature tags, corresponding platform behavior feature tags are generated based on the user's past registration frequency, no-show records, number of canceled registrations, registration time period selection habits, and frequently visited departments from the user registration data.

[0065] Step S2022: Based on user registration data, operator data, and geographic data, extract and generate medical treatment preference tags.

[0066] Healthcare preference tags can be understood as labels that reflect a user's healthcare selection tendencies, extracted by combining user registration data, carrier data, and geographic data. These tags may include frequently used hospital types, frequently used departments, preferences for specialist appointments, and acceptance of online consultations.

[0067] Specifically, user registration data, carrier data, and geographic data are fused and correlated to jointly extract and generate user medical treatment preference tags. By combining information such as frequently visited departments, appointment times, and past hospital choices from user registration data, recent travel patterns and frequently used activity areas from carrier data, and distance and accessibility from geographic data to hospitals, user medical treatment preferences are comprehensively analyzed to obtain these tags.

[0068] For example, by combining the distance between a user's residence and the hospital and their travel trajectory, we can determine whether the user prefers to seek medical treatment nearby; by combining the user's previous registered departments and time periods, we can determine the user's preferred departments and time of visit (such as weekday mornings and weekends); by combining the medical-related locations in the user's travel trajectory, we can determine the user's preference for hospital level and type, thus obtaining medical treatment preference tags.

[0069] Step S2023: Establish a three-dimensional user tag system based on basic attribute tags, health status tags, platform behavior feature tags, and medical treatment preference tags.

[0070] The four categories of tags are systematically categorized and arranged, clarifying the subordinate and related relationships of each category. The tags are standardized, unifying the naming conventions and value ranges to ensure the standardization and uniformity of the tag system. At the same time, a tag association verification mechanism is established to ensure that there are no conflicts and that the information is consistent between tags of different dimensions, resulting in a comprehensive tag system for users.

[0071] The method for determining registration priority provided by this invention extracts and generates basic attribute tags, health status tags, and platform behavior feature tags from user registration data. Then, it combines user registration data, operator data, and geographic data to extract and generate medical preference tags. By integrating these four types of tags from different dimensions, a three-dimensional user tag system is established. This system can accurately classify tags according to different data sources, fully aggregate user-related information corresponding to various tags, and clearly define the information dimensions of the constructed three-dimensional user tag system. It fully integrates user information from different data sources and achieves comprehensive organization and sorting of user-related information.

[0072] The aforementioned multi-source data also includes operator data and geographic data. Based on step S103 of this embodiment, which uses multi-source data and a user three-dimensional tagging system, a comprehensive profile evaluation result is obtained as follows: Figure 3 The implementation of steps S3031 to S3036 in the above steps: Step S3031: Perform vector construction processing on the fused feature vector to obtain the user feature vector.

[0073] The fusion feature vector can be understood as a standardized set of features obtained after multi-source feature fusion processing of user registration data, operator data, and geographic data.

[0074] User feature vectors can be understood as feature vectors obtained by performing vector construction processing on the fused feature vectors.

[0075] Specifically, a fused feature vector is obtained by performing multi-source feature fusion processing on user registration data, operator data, and geographic data, and then a vector construction process is performed on the fused feature vector.

[0076] As an example, the fused feature vector undergoes feature cleaning to remove redundant and duplicate feature information and fill in missing feature values ​​to ensure the integrity of the feature vector. Then, the feature vector is standardized to convert feature values ​​of different dimensions and magnitudes into the same value range, eliminating the influence of dimensions. Finally, the processed feature vector is dimension-normalized to fix the vector dimensions and feature arrangement order, transforming the fused feature vector into a user feature vector with a unified format that can be directly used for various subsequent data processing.

[0077] As an example, the fused feature vectors are first sorted by feature importance to select core features that are highly relevant to user profile assessment and registration priority determination, while irrelevant features are removed. Then, a dimensionality compression algorithm is used to reduce the dimension of the feature vectors while retaining the core feature information, simplifying the subsequent data processing flow, completing the vector construction, and obtaining the user feature vector.

[0078] Step S3032: Perform user group clustering processing on the user feature vector to obtain the first processing result.

[0079] The first processing result can be understood as the result obtained after performing user group clustering processing on the user feature vector.

[0080] Specifically, user group clustering is performed on the user feature vectors of all candidate users.

[0081] As an example, based on the similarity of core features such as health status, medical preferences, and platform behavior in user feature vectors, a clustering algorithm is used to group users. Users with similar features are divided into the same group, while users with significant differences in features are divided into different groups. After clustering, the core features of each group are extracted and summarized to identify the common features of various groups (such as chronic disease follow-up visits, low no-show risk, and fixed medical departments), thus obtaining the first processing result.

[0082] Step S3033: Perform no-show risk prediction processing on the user feature vector to obtain the second processing result.

[0083] The second processing result can be understood as the result obtained after performing a no-show risk prediction process on the user feature vector.

[0084] Specifically, based on the core features related to user performance in the user feature vector (such as past no-show records, registration frequency, number of cancellations, and completion rate of medical visits), the no-show risk prediction process is performed on the user feature vector to obtain the second processing result.

[0085] As an example, the probability of no-show for each user is quantified by a pre-set prediction model. The prediction results are then verified and corrected by combining the platform behavior feature tags in the user three-dimensional tag system. Finally, the user no-show risk is divided into different levels (such as high no-show risk, medium no-show risk, and low no-show risk) or a specific no-show probability value is output, forming the second processing result.

[0086] Step S3034: Perform medical demand prediction processing on the user feature vector to obtain the third processing result.

[0087] The third processing result can be understood as the result obtained after performing medical demand prediction processing on the user feature vector.

[0088] Specifically, based on the core features related to user health and medical treatment in the user feature vector (such as health status tags, past medical records, symptom descriptions, and selection of medical departments), the user feature vector is processed to predict medical treatment needs, resulting in the third processing result.

[0089] As an example, by analyzing a user's health status and past medical visit patterns, the urgency and intensity of the user's current medical needs can be assessed. Combined with the health status tags in the user's three-dimensional tagging system, the prediction results can be optimized and adjusted. Finally, the user's medical needs can be divided into different levels (such as urgent, general, and ordinary) or specific demand intensity scores can be output to form a third processing result.

[0090] Step S3035: Perform feature association rule mining on the user feature vector to obtain the fourth processing result.

[0091] The fourth processing result can be understood as the result obtained after performing feature association rule mining on the user feature vector.

[0092] Specifically, feature association rule mining is performed on user feature vectors to uncover the inherent relationships between different user features and the correlation patterns between features and user medical behavior. The focus is on mining the correlations between features such as health status, platform behavior, and medical preferences (e.g., the correlation between elderly patients, chronic disease follow-up visits and low no-show risk, and the correlation between preference for nearby medical care and high-frequency registration). The mined association rules are then screened and validated, eliminating invalid and meaningless associations and retaining those with practical reference value, forming the fourth processing result.

[0093] Step S3036: Based on the first processing result, the second processing result, the third processing result, and the fourth processing result, a comprehensive portrait evaluation result is obtained.

[0094] After determining the first, second, third, and fourth processing results, a weighted fusion calculation is performed on the four types of processing results, assigning different weights to different processing results (e.g., the medical demand prediction result has the highest weight, followed by the no-show risk prediction result), and then a comprehensive profile evaluation result is obtained.

[0095] The registration priority determination method provided by this invention obtains a fused feature vector by performing multi-source feature fusion processing on user registration data, operator data, and geographic data. Then, vector construction processing is performed on the fused feature vector to obtain a user feature vector. The user feature vector is then subjected to user group clustering processing, no-show risk prediction processing, medical demand prediction processing, and feature association rule mining processing to obtain four corresponding processing results. Finally, a comprehensive profile evaluation result is obtained based on the four processing results, which can orderly complete the feature integration and standardized transformation of multiple types of data.

[0096] The aforementioned multi-source data also includes operator data and geographic data. Based on the comprehensive evaluation results of the user profile and the multi-source data in step S104 of this embodiment, the user's registration priority is determined as follows: Figure 4 The implementation of steps S4041 to S4045 is as follows: Step S4041: Obtain the number of available appointment slots and the number of people on the waiting list corresponding to the user registration data.

[0097] The number of available appointment slots can be understood as the total number of appointment slots that are available for waiting users to book again during the current consultation period due to reasons such as user cancellation, doctors adding appointment slots, and the system releasing idle appointment slots.

[0098] The number of people on the waiting list can be understood as the total number of users who have submitted waiting list applications and are in a valid waiting list status under the same department or doctor, based on user registration data statistics.

[0099] Specifically, the number of available appointment slots during the current consultation period can be obtained through the hospital's registration database. This includes appointment slots released by users who actively cancel their registrations, additional appointment slots added by doctors, and automatically released idle and unused appointment slots. By statistically summarizing all types of released appointment slots, the accurate total number of released appointment slots can be obtained.

[0100] By analyzing user registration data, we can identify users who have submitted waitlist applications and are currently in a valid state (not canceled from the waitlist and not allocated an appointment) within the same department, doctor, or time slot. We can then count these users to obtain the corresponding number of people on the waitlist.

[0101] Step S4042: Determine the scarcity coefficient of available appointment slots based on the number of available slots and the number of people on the waiting list.

[0102] The scarcity coefficient of available appointment slots can be understood as a quantitative parameter calculated based on the number of available slots and the number of people on the waiting list, used to characterize the current supply and demand tension of available slots.

[0103] Based on the number of available appointment slots and the number of people on the waiting list, the two sets of data are calculated using preset calculation rules to determine the scarcity coefficient of appointment slots.

[0104] As an example, the basic calculation is performed by the ratio of the number of available numbers to the number of people on the waiting list. When the number of people on the waiting list is not 0, the number of available numbers is equal to the number of available numbers divided by the number of people on the waiting list. When the number of people on the waiting list is 0, the number of available numbers is set to a preset maximum value (indicating that there are currently no waiting users and the number of available numbers is not scarce). After the calculation is completed, the obtained number of available numbers is standardized and converted into a quantitative parameter within a preset value range to obtain the number of available numbers.

[0105] Step S4043: Based on user registration data, determine the user waiting order and waiting time.

[0106] The user waiting list order can be understood as the order in which users are queued in the waiting list, determined based on information such as the submission time of their waiting list application and their historical performance in the user registration data.

[0107] The waiting time for standby registration can be understood as the cumulative waiting time from the time the user submits the standby application to the current moment, calculated based on the submission time of the standby application in the user's registration data.

[0108] Specifically, user registration data is retrieved, and relevant information such as the submission time of each valid waiting list user's application and historical fulfillment records are extracted to determine the user's waiting order and waiting time. The cumulative waiting time for each user from the submission time of their waiting list application to the current time is calculated to obtain the waiting time.

[0109] Step S4044: Based on the user candidate order, multi-source data, and risk coefficient of the external risk control interface, determine the first target score corresponding to the user.

[0110] The primary target score can be understood as a basic user score derived through quantitative calculation, combining user candidate order, multi-source data, and risk coefficients from external risk control interfaces.

[0111] Specifically, based on the determined user candidate order, multi-source data collected in the early stage, and risk coefficients pushed by external risk control interfaces, quantitative scoring is carried out according to preset scoring rules to determine the user's corresponding first target score.

[0112] As an example, a base score is assigned based on the user's waiting list order, with higher waiting lists resulting in higher base scores. Secondly, the base score is adjusted by adding or subtracting information from multi-source data, such as the user's historical performance records and registration frequency (e.g., adding points for no-no-show records and adding points for high-frequency compliant registrations). Finally, the score is corrected by incorporating the risk coefficient from the external risk control interface; users with low-risk coefficients receive additional points, while users with high-risk coefficients have their points deducted. The combined results of these three calculations yield the primary target score for each user.

[0113] Step S4045: Based on the first target score, the waiting time, the comprehensive assessment results of the profile, the risk coefficient of the external risk control interface, and the scarcity coefficient of the appointment slots, determine the user's registration priority.

[0114] Specifically, after determining the primary target score, waiting time, comprehensive profile assessment results, risk coefficient of external risk control interface, and scarcity coefficient of appointment slots, the user's appointment priority can be further determined.

[0115] Specifically, step S4045 above may include the following steps: Step a1: If the scarcity coefficient of the number source is greater than the scarcity threshold, increase the preset first weight corresponding to the waiting time to obtain the first target weight, and decrease the preset third weight corresponding to the risk coefficient to obtain the third target weight; wherein, the first proportion of increasing the preset first weight and the second proportion of decreasing the preset third weight are the same.

[0116] The scarcity threshold can be understood as a preset critical value used to determine the degree of scarcity of available phone numbers.

[0117] The preset first weight can be understood as an initial weight value that is preset in advance and corresponds to the waiting time of the candidate.

[0118] The first target weight can be understood as the final weight obtained by adding a preset first weight when the scarcity coefficient of the number source is greater than the scarcity threshold.

[0119] The first ratio can be understood as a preset ratio used to increase the preset first weight, which is exactly the same as the second ratio for reducing the preset third weight, ensuring the balance of weight adjustment.

[0120] The risk coefficient can be understood as the risk coefficient of the external risk control interface, which is pushed synchronously by the third-party risk control service interface. It is used to characterize the degree of risk of user's failure to fulfill registration obligations, no-shows, malicious registration, etc.

[0121] The preset third weight can be understood as the initial weight value corresponding to the risk coefficient that is preset in advance.

[0122] The third target weight can be understood as the final weight obtained by reducing the preset third weight when the scarcity coefficient of the number source is greater than the scarcity threshold.

[0123] Specifically, the calculated scarcity coefficient of available appointment slots is compared with a preset scarcity threshold. If the scarcity coefficient is greater than the threshold, the current availability of appointment slots is determined to be tight, and a dynamic weight adjustment process is initiated. This means increasing the preset first weight corresponding to the waiting time to obtain the first target weight, and decreasing the preset third weight corresponding to the risk coefficient to obtain the third target weight.

[0124] For example, first, retrieve the preset first proportion (e.g., 10%), and increase the preset first weight (e.g., 20%) corresponding to the waiting time by the first proportion to calculate the first target weight (20%×(1+10%)=22%). At the same time, retrieve the second proportion (10%) which is the same as the first proportion, and decrease the preset third weight (e.g., 15%) corresponding to the risk coefficient by the second proportion to calculate the third target weight (15%×(1-10%)=13.5%).

[0125] As an example, multiple first ratios (consistent with the second ratio) are preset. Based on the extent to which the scarcity coefficient of the available appointment slots exceeds the scarcity threshold, the corresponding ratio is matched (e.g., if the exceedance is 0-0.1, the ratio is 10%; if the exceedance is 0.1-0.2, the ratio is 20%). The more the scarcity coefficient exceeds the threshold, the larger the adjustment ratio becomes, so that the weight adjustment is more in line with the scarcity of available appointment slots.

[0126] Step a2: Based on the waiting time for the candidate and the weight of the first target, determine the base score of the first target time for the user.

[0127] The first target time base score can be understood as a score related to the waiting time, calculated through quantitative calculation based on the waiting time and the weight of the first target when there is a shortage of available slots.

[0128] Specifically, based on the weight of the first target and the user's waiting time, the base score of the first target time is calculated according to the preset calculation rules.

[0129] As an example, the waiting time for the backup is first standardized (e.g., the waiting time is converted into minutes, and the waiting time is mapped to a base score of 0-30 minutes according to the rule that the longer the waiting time, the higher the base score); then the standardized base score is multiplied by the first target weight to obtain the base score of the first target time for each user (e.g., base score of 25 points × first target weight of 22% = 5.5 points).

[0130] Step a3: Based on the risk coefficient and the third objective weight, determine the first objective risk score corresponding to the user.

[0131] The first target risk score can be understood as a score related to user risk, calculated through quantitative calculation based on the risk coefficient and the weight of the third target when appointment slots are scarce.

[0132] Specifically, based on the weight of the third objective and the risk coefficient of each user pushed by the external risk control interface, the risk score of the first objective is calculated according to the preset calculation rules.

[0133] As an example, the risk coefficient is standardized (e.g., the risk coefficient of 0-1 is mapped to a basic risk score of 0-10, and the lower the risk coefficient, the higher the basic risk score); then the standardized basic risk score is multiplied by the third target weight to obtain the first target risk score for each user (e.g., basic risk score of 8 points × third target weight of 13.5% = 1.08 points).

[0134] Step a4: Determine the portrait feature score based on the comprehensive portrait evaluation results and the preset second weight.

[0135] The profile feature score can be understood as a score that reflects the user's medical needs and overall status, obtained through quantitative calculation based on the comprehensive assessment results of the profile and a preset second weight.

[0136] The preset second weight can be understood as a fixed weight value that is preset in advance and corresponds to the comprehensive evaluation result of the profile.

[0137] Specifically, based on the comprehensive evaluation results of each user's profile and the preset second weight, the profile feature score is calculated according to the preset calculation rules.

[0138] As an example, the comprehensive assessment results of the profile are quantified (e.g., the assessment results are divided into quantitative scores of 0-100 points, with higher scores for more urgent medical needs and better performance records); then the quantified comprehensive assessment score of the profile is multiplied by the preset second weight to obtain the profile feature score for each user (e.g., profile quantification score of 90 points × preset second weight of 45% = 40.5 points).

[0139] Step a5: Based on the first target score, the first target time base score, the profile feature score, and the first target risk score, determine the user's registration priority.

[0140] Specifically, based on the first target score, the first target time base score, the profile feature score, and the first target risk score, the priority total score for each user is calculated according to the rule that the total score = first target score + first target time base score + profile feature score - first target risk score. After the calculation is completed, all candidate users are sorted in descending order of priority total score. The higher the total score, the higher the registration priority. Finally, the registration priority of each user is determined.

[0141] Step a6: If the scarcity coefficient of the number source is not greater than the scarcity threshold, determine the initial time base score based on the waiting time of the candidate and the preset first weight.

[0142] The initial time base score can be understood as a score related to the waiting time, calculated through quantitative calculation based on the waiting time and a preset first weight when there are sufficient available slots.

[0143] Specifically, if the scarcity coefficient of the number source is not greater than the scarcity threshold, it is determined that the current number source is in a sufficient state. There is no need to adjust the weight. The preset first weight (e.g., 20%) and the waiting time of each user are directly retrieved. According to the standardization rules consistent with step a2, the waiting time is mapped to the base time score. Then the base time score is multiplied by the preset first weight to obtain the initial time base score corresponding to each user.

[0144] Step a7: Determine the initial risk score based on the risk coefficient and the preset third weight.

[0145] The initial risk score can be understood as a score related to user risk, obtained through quantitative calculation based on risk coefficients and preset third weights when there are sufficient available appointment slots.

[0146] Specifically, when there are sufficient numbers available, the risk coefficients of each user are directly retrieved from the preset third weight and the external risk control interface. Following the same standardization rules as step a3, the risk coefficients are mapped to the basic risk score. Then, the basic risk score is multiplied by the preset third weight to obtain the initial risk score for each user.

[0147] Step a8: Based on the first target score, initial time base score, profile feature score, and initial risk score, determine the user's registration priority.

[0148] Specifically, based on the first target score, initial time base score, profile feature score, and initial risk score, the total score is calculated for each user according to the same total score calculation rule as in step a5 (total score = first target score + initial time base score + profile feature score - initial risk score). Then, all candidate users are sorted in descending order of total score to determine the registration priority for each user.

[0149] The method for determining registration priority provided by this invention obtains the number of available appointment slots and the number of people on the waiting list corresponding to the user's registration data to determine the scarcity coefficient of appointment slots. Simultaneously, it determines the user's waiting order and waiting time based on the user's registration data. Then, it combines the user's waiting order, multi-source data, and the risk coefficient of the external risk control interface to determine the user's corresponding first target score. Finally, it combines the first target score, waiting time, comprehensive profile assessment results, the risk coefficient of the external risk control interface, and the scarcity coefficient of appointment slots to determine the user's registration priority. This method integrates appointment supply and demand, user queuing information, risk control information, and comprehensive user assessment information to determine priority, making the basis for determining registration priority rich and comprehensive, thereby effectively improving the rationality of medical resource allocation.

[0150] In addition, when the scarcity coefficient of appointment slots is greater than the scarcity threshold, the preset first weight corresponding to the waiting time is increased by the same proportion and the preset third weight corresponding to the risk coefficient is decreased. The first target weight and the third target weight are obtained respectively. Then, the first target time base score, the first target risk score and the profile feature score are calculated in sequence. The user's registration priority is determined by combining the first target score.

[0151] Furthermore, when the scarcity coefficient of appointment slots is not greater than the scarcity threshold, the original preset weights are used to calculate the initial time base score and initial risk score, and the registration priority is determined in combination with the corresponding content. The weight of different scoring dimensions can be flexibly adjusted according to the tightness of appointment slot supply and demand, and the score calculation can be adapted to different appointment slot supply and demand scenarios. This allows the registration priority to be reasonably determined in line with the current tightness of appointment slot resources, thereby effectively improving the rationality of medical resource allocation.

[0152] In one possible implementation, step S104 above may also include the following steps: Step b1: Based on user registration data, determine the user waiting list order and waiting time. Please refer to step S4043 above for details; further explanation is omitted here.

[0153] Step b2 determines the primary target score for each user based on their candidate order, multi-source data, and risk coefficients from external risk control interfaces. Please refer to step S4044 above for details; further elaboration is omitted here.

[0154] Step b3: If the waiting time for the candidate exceeds the time threshold, increase the preset first weight corresponding to the waiting time for the candidate to obtain the fourth target weight.

[0155] The duration threshold can be understood as a preset critical value used to determine whether the waiting time for candidates is too long.

[0156] The fourth objective weight can be understood as the final weight obtained by adding a preset first weight when the waiting time for the candidate exceeds the time threshold.

[0157] Specifically, based on user registration data, the core information of each valid waiting user (who has not canceled their waiting list and has not been allocated a number) is extracted to determine the user's waiting order and waiting time.

[0158] Specifically, a numerical comparison is performed based on each user's waiting time and a preset time threshold (e.g., 50 minutes). If a user's waiting time exceeds the time threshold, the user is identified as a long-waiting user, and a weight adjustment process is initiated. A preset third ratio is retrieved, and the preset first weight corresponding to the waiting time is increased by the third ratio to calculate the fourth target weight.

[0159] Step b4: Reduce the preset second weight corresponding to the comprehensive portrait evaluation result to obtain the fifth target weight.

[0160] The fifth objective weight can be understood as the final weight obtained by reducing the preset second weight when the waiting time for the candidate exceeds the time threshold.

[0161] Specifically, under the premise of initiating weight adjustment in step b3, the preset fourth ratio is retrieved, and the preset second weight corresponding to the comprehensive assessment result of the portrait is reduced by the fourth ratio to calculate the fifth target weight; wherein, the fourth ratio is consistent with the third ratio to avoid the overall weight imbalance after the weight adjustment.

[0162] Step b5: Reduce the preset third weight corresponding to the risk coefficient to obtain the sixth target weight; wherein, the third proportion of increasing the preset first weight, the fourth proportion of decreasing the preset second weight, and the fifth proportion of decreasing the preset third weight are the same.

[0163] The sixth objective weight can be understood as the final weight obtained by reducing the preset third weight when the waiting time for the candidate exceeds the time threshold.

[0164] Specifically, under the premise of initiating weight adjustment in step b3, the preset fifth ratio is retrieved, and the preset third weight corresponding to the comprehensive assessment result of the portrait is reduced by the fifth ratio to calculate the sixth target weight; wherein, the fifth ratio, the fourth ratio and the third ratio are consistent.

[0165] Step b6: Determine the base score for the second objective based on the waiting time for the candidate and the weight of the fourth objective.

[0166] The second objective time base score can be understood as a score related to the waiting time, calculated by quantifying the waiting time and the weight of the fourth objective when the waiting time exceeds the time threshold.

[0167] Specifically, after determining the waiting time for the alternates and the weight of the fourth objective, the base score for the second objective time can be determined.

[0168] As an example, the waiting time for standby is standardized and mapped to a base time score of 0-30 points (the longer the waiting time, the higher the base time score, such as a base time score of 28 points for a 90-minute wait and 20 points for a 70-minute wait). Then, the standardized base time score is multiplied by the fourth objective weight to obtain the second objective time base score for each long-waiting user (e.g., base time score of 28 points × fourth objective weight of 23% = 6.44 points).

[0169] Step b7: Based on the comprehensive assessment results of the portrait and the weight of the fifth objective, determine the feature score of the target portrait.

[0170] The target profile feature score can be understood as a score reflecting the user's overall medical status, obtained through quantitative calculation based on the comprehensive profile evaluation results and the weight of the fifth target when the waiting time exceeds the time threshold.

[0171] Specifically, after determining the comprehensive assessment results of the portrait and the weight of the fifth objective, the feature scores of the target portrait can be further determined.

[0172] As an example, the evaluation results are divided into quantitative scores of 0-100 (the more urgent the medical needs and the better the performance record, the higher the quantitative score, such as 95 points for emergency patients and 75 points for regular follow-up patients); then the quantitative profile comprehensive evaluation score is multiplied by the fifth target weight to obtain the target profile feature score for each long-waiting user (such as 95 points × fifth target weight 38.25% = 36.34 points).

[0173] Step b8: Based on the risk coefficient and the sixth objective weight, determine the second objective risk score corresponding to the user.

[0174] The second target risk score can be understood as a score related to user risk, calculated by quantifying the risk coefficient of the external risk control interface and the weight of the sixth target when the waiting time exceeds the time threshold.

[0175] Specifically, after determining the risk coefficient and the weight of the sixth objective, the user's corresponding second objective risk score can be further determined.

[0176] As an example, the risk coefficient is standardized, mapping a risk coefficient of 0-1 to a base risk score of 0-10 (the lower the risk coefficient, the higher the base risk score; for example, a risk coefficient of 0.1 corresponds to a base risk score of 10, and a risk coefficient of 0.7 corresponds to a base risk score of 3). Then, the standardized base risk score is multiplied by the sixth target weight to obtain the second target risk score for each long-waiting user (e.g., base risk score of 10 × sixth target weight 12.75% = 1.28).

[0177] Step b9: Determine the user's registration priority based on the first target score, the second target time base score, the target profile feature score, and the second target risk score.

[0178] After determining the base score for the second target time, the target profile feature score, and the second target risk score, the total priority score for each long-waiting user is calculated according to the rule: Total Priority Score = First Target Score + Second Target Base Score + Target Profile Feature Score - Second Target Risk Score. After calculation, all long-waiting candidates are sorted in descending order of total priority score, with higher total scores indicating higher registration priority. At the same time, the overall priority of long-waiting users is set higher than that of users whose waiting time does not exceed the time threshold, thus determining the registration priority of all candidates.

[0179] The registration priority determination method provided by this invention reorders the waiting queue based on the target registration priority when a user's registration slot is released. The waiting queue itself is formed by sorting the registration priorities from highest to lowest. Then, the user with the highest target registration priority is selected to complete the allocation of the registration slot. This method can update the queue order according to priority as soon as the registration slot is released, ensuring that the allocation of registration slots is always based on priority, allowing users with higher priority to obtain registration slots first, ensuring the rationality and orderliness of the allocation of registration slots, and avoiding chaos in the allocation of registration slots.

[0180] In one possible implementation, the method for determining the registration priority described above may also include the following steps: Step c1: Determine whether the target event has been detected; the target event includes at least one of the following: number release event, user profile change event, and external risk control score update event.

[0181] The target event may include at least one of the following: the release of registration slots, changes in user profiles, and updates to external risk control scores.

[0182] Specifically, a real-time monitoring mechanism is activated, connecting to the hospital's registration database, user profile management system, and third-party external risk control interfaces to capture various target events.

[0183] As an example, the system monitors for three events: First, it monitors for events related to the release of appointment slots. This involves real-time synchronization with the hospital's registration database. When a user cancels an appointment, a doctor adds an appointment slot, or the system releases an idle appointment slot, the system determines that an appointment slot release event has been detected. Second, it monitors for events related to changes in user profiles. This involves real-time monitoring of changes in the comprehensive evaluation results of user profiles. When a user adds a new medical record, has a no-show record, experiences a change in health status, or adjusts their medical preferences, resulting in an update to the profile evaluation results, the system determines that a user profile change event has been detected. Finally, it monitors for events related to updates in external risk control scores. This involves receiving risk coefficient data pushed by a third-party risk control interface. When a fluctuation (increase or decrease) in a user's risk coefficient is detected, the system determines that an external risk control score update event has been detected. The system performs real-time checks on these three types of events. If any one of these events is detected, step c2 is executed.

[0184] Step c2: If the target event is detected, recalculate the registration priority of the user corresponding to the target event to obtain the target registration priority.

[0185] Once any target event is detected, the registration priority recalculation process is immediately initiated, focusing on the relevant users corresponding to the target event to obtain the target registration priority.

[0186] As an example, the process of locating the target user for a specific event is as follows: If a number release event is detected, all currently valid waiting list users are located; if a user profile change event is detected, the specific user whose profile has changed is located; if an external risk control score update event is detected, the specific user whose risk coefficient has changed is located. Next, the latest relevant data for this type of user is retrieved (including current waiting time, updated profile comprehensive assessment results, updated risk coefficient, and various preset weights). Subsequently, according to preset priority calculation rules, the registration priority for this type of user is recalculated to obtain the target registration priority.

[0187] The registration priority determination method provided by this invention determines whether a target event, including a number release event, a user profile change event, and an external risk control score update event, is detected. When a corresponding target event is detected, the registration priority of the user corresponding to the target event is recalculated to obtain the target registration priority. This method can promptly adjust the registration priority according to various changes, ensuring that the obtained target registration priority is real-time and accurate.

[0188] In one possible implementation, the method for determining the registration priority described above may also include the following steps: Step c3: Based on the preset time window, update the user's wait time, the preset first weight corresponding to the wait time, the preset second weight corresponding to the comprehensive profile evaluation result, and the preset third weight corresponding to the risk coefficient.

[0189] Specifically, a preset time window is started (e.g., a preset time window of 30 minutes). At the end of each time window, the update process of various parameters is automatically triggered.

[0190] As an example, the system updates the waiting time for users on the waiting list: It retrieves the submission time of each valid waiting list user's application, subtracts the submission time from the system time at the end of the current time window, recalculates the waiting time, removes invalid waiting times, and converts the data to minutes, overwriting the original waiting time data. Secondly, it updates the preset first weight: Based on the supply and demand of appointment slots within the current time window (e.g., the scarcity of slots, the number of users with long waiting times), the preset first weight is dynamically adjusted (e.g., when slots are scarce, the preset first weight is adjusted from 20% to 25%; when slots are plentiful, it is adjusted to 18%). Thirdly, ... Update the preset second weight: Based on the overall changes in the comprehensive evaluation results of user profiles within the current time window (such as changes in the proportion of high-demand users), fine-tune the preset second weight (e.g., if the proportion of high-demand users increases, adjust the preset second weight from 45% to 48%; if the proportion decreases, adjust it to 42%). Finally, update the preset third weight: Based on the overall distribution of risk coefficients pushed by external risk control interfaces within the current time window (such as changes in the proportion of high-risk users), adjust the preset third weight (e.g., if the proportion of high-risk users increases, adjust the preset third weight from 15% to 18%; if the proportion decreases, adjust it to 12%).

[0191] Step c4: Determine the target registration priority based on the updated first weight, the updated second weight, and the updated third weight; wherein, the target registration priority is the registration priority updated from the user's registration priority.

[0192] Based on the updated first weight, the updated second weight, and the updated third weight, and simultaneously retrieving the corresponding data (waiting time for candidates, comprehensive assessment results of the profile, and risk coefficient), the target registration priority is calculated and determined step by step according to the preset calculation rules.

[0193] As an example, the process involves calculating a base time score based on the updated first weight and the updated waiting time; secondly, calculating a profile feature score based on the updated second weight and the updated profile comprehensive evaluation result; thirdly, calculating a risk score based on the updated third weight and the updated risk coefficient; subsequently, calculating each user's total priority score according to the rule: total priority score = base time score + profile feature score - risk score; finally, ranking all waiting users in descending order of total priority score, with higher total scores indicating higher target registration priority, thus determining the target registration priority for each user.

[0194] The registration priority determination method provided by this invention updates the user waiting time and the preset weights corresponding to three different dimensions according to a preset time window, and determines the target registration priority based on the updated weights. It can realize an operation mode that combines event-triggered targeted priority recalculation with timed batch parameter updates, and can promptly follow various changes to synchronously correct the user registration priority, ensuring that the obtained target registration priority has real-time performance and accuracy.

[0195] In one possible implementation, the above method also includes: When a user's appointment slot is detected to be released, the waiting queue is reordered based on the target registration priority, and the user with the highest target registration priority is selected to complete the appointment slot allocation. The target registration priority is the updated registration priority of the user, and the waiting queue is the queue obtained by sorting the target registration priorities in descending order.

[0196] The target registration priority indicator is the updated registration priority after the user's registration priority has been applied. For specific update methods, please refer to steps c1 to c4 above; they will not be elaborated upon here.

[0197] Specifically, when it is detected that a user's appointment slot has been released, the waiting queue needs to be reordered to select the user with the highest priority for registration to complete the appointment slot allocation.

[0198] In some embodiments, the registration priority determination device of the present invention can be implemented in a combination of hardware and software. As an example, the registration priority determination device of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the registration priority determination method of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0199] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0200] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned methods for determining registration priority. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the method for determining registration priority shown in any embodiment of the present invention by calling the computer program.

[0201] In one alternative embodiment, an electronic device is provided, such as Figure 5 As shown, Figure 5 The illustrated electronic device 5000 includes a processor 5001 and a memory 5003. The processor 5001 and the memory 5003 are connected, for example, via a bus 5002. Optionally, the electronic device 5000 may further include a transceiver 5004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 5004 is not limited to one type, and the structure of the electronic device 5000 does not constitute a limitation on the embodiments of the present invention.

[0202] Processor 5001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 5001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0203] Bus 5002 may include a path for transmitting information between the aforementioned components. Bus 5002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 5002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5The bus 5002 is represented by only one thick line, but this does not mean that there is only one bus or one type of bus.

[0204] The memory 5003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0205] The memory 5003 stores application code (computer program) for executing the present invention, and its execution is controlled by the processor 5001. The processor 5001 executes the application code stored in the memory 5003 to implement the content shown in the foregoing method embodiments.

[0206] Among them, electronic devices can also be terminal devices. Terminal devices can be any devices that can install applications, including at least two of the following: smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0207] It should be noted that, Figure 5 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0208] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned methods for determining registration priority.

[0209] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0210] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the aforementioned registration priority determination method.

[0211] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0212] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0213] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EEPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0214] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0215] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0216] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0217] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a circuit, module, or system. Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0218] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for determining registration priority, characterized in that, The method includes: Acquire multi-source data; wherein, the multi-source data includes at least user registration data and risk coefficients from external risk control interfaces; Based on the aforementioned multi-source data, a three-dimensional user tagging system is established; Based on the multi-source data and the user three-dimensional tag system, a comprehensive profile evaluation result is obtained; Based on the comprehensive evaluation results of the aforementioned profile and the multi-source data, the user's registration priority is determined.

2. The method for determining registration priority according to claim 1, characterized in that, The multi-source data also includes operator data and geographic data. The establishment of a three-dimensional user tagging system based on the multi-source data includes: Based on the user registration data, basic attribute tags, health status tags, and platform behavior feature tags are extracted and generated. Based on the user registration data, operator data, and geographic data, medical preference tags are extracted and generated. A three-dimensional user tagging system is established based on the aforementioned basic attribute tags, health status tags, platform behavior feature tags, and medical treatment preference tags.

3. The method for determining registration priority according to claim 1, characterized in that, The multi-source data also includes operator data and geographic data. Based on the multi-source data and the user three-dimensional tagging system, the comprehensive profile evaluation result includes: The user registration data, operator data, and geographic data are subjected to multi-source feature fusion processing to obtain a fused feature vector; The fused feature vector is processed to obtain the user feature vector; The user feature vectors are subjected to user group clustering processing to obtain the first processing result; Perform no-show risk prediction processing on the user feature vector to obtain the second processing result; The user feature vector is subjected to medical demand prediction processing to obtain a third processing result; The user feature vector is subjected to feature association rule mining processing to obtain the fourth processing result; Based on the first processing result, the second processing result, the third processing result, and the fourth processing result, the comprehensive evaluation result of the portrait is obtained.

4. The method for determining registration priority according to claim 1, characterized in that, The process of determining a user's registration priority based on the comprehensive evaluation results of the user profile and the multi-source data includes: Get the number of available appointment slots and the number of people on the waiting list corresponding to user registration data; Based on the number of available appointment slots and the number of people on the waiting list, a scarcity coefficient for appointment slots is determined; Based on the user registration data, the user waiting order and waiting time are determined; Based on the user candidate order, the multi-source data, and the risk coefficient of the external risk control interface, the first target score corresponding to the user is determined; Based on the first target score, the waiting time for the backup, the comprehensive assessment result of the profile, the risk coefficient of the external risk control interface, and the scarcity coefficient of the appointment slots, the user's registration priority is determined.

5. The method for determining registration priority according to claim 4, characterized in that, The process of determining a user's registration priority based on the first target score, the waiting time for the backup appointment, the comprehensive assessment result of the user profile, the risk coefficient of the external risk control interface, and the scarcity coefficient of appointment slots includes: If the scarcity coefficient of the available slots is greater than the scarcity threshold, the preset first weight corresponding to the waiting time for the backup slots is increased to obtain the first target weight, and the preset third weight corresponding to the risk coefficient is decreased to obtain the third target weight; wherein, the first proportion of increasing the preset first weight and the second proportion of decreasing the preset third weight are the same; Based on the waiting time and the first target weight, the first target time base score corresponding to the user is determined; Based on the risk coefficient and the third target weight, the first target risk score corresponding to the user is determined; Based on the comprehensive evaluation results of the portrait and the preset second weight, the portrait feature score is determined; Based on the first target score, the first target time base score, the profile feature score, and the first target risk score, the user's registration priority is determined. If the scarcity coefficient of the number source is not greater than the scarcity threshold, the initial time base score is determined based on the waiting time of the candidates and the preset first weight. Based on the aforementioned risk coefficient and the preset third weight, an initial risk score is determined; Based on the first target score, the initial time base score, the profile feature score, and the initial risk score, the user's registration priority is determined.

6. The method for determining registration priority according to claim 1, characterized in that, The process of determining a user's registration priority based on the comprehensive evaluation results of the user profile and the multi-source data includes: Based on the user registration data, the user waiting order and waiting time are determined; Based on the user candidate order, the multi-source data, and the risk coefficient of the external risk control interface, the first target score corresponding to the user is determined; If the waiting time for the candidate exceeds the time threshold, the preset first weight corresponding to the waiting time for the candidate is increased to obtain the fourth target weight; The fifth objective weight is obtained by reducing the preset second weight corresponding to the comprehensive assessment result of the portrait; The sixth target weight is obtained by reducing the preset third weight corresponding to the risk coefficient; wherein, the third proportion of increasing the preset first weight, the fourth proportion of decreasing the preset second weight, and the fifth proportion of decreasing the preset third weight are the same. Based on the candidate waiting time and the fourth target weight, the second target time base score is determined; Based on the comprehensive evaluation results of the portrait and the weight of the fifth target, the feature score of the target portrait is determined; Based on the risk coefficient and the sixth objective weight, the second objective risk score corresponding to the user is determined; Based on the first target score, the second target time base score, the target profile feature score, and the second target risk score, the user's registration priority is determined.

7. The method for determining registration priority according to any one of claims 1-6, characterized in that, The method further includes: When a user's appointment slot is detected to be released, the waiting queue is reordered based on the target registration priority, and the user with the highest target registration priority is selected to complete the appointment slot allocation; wherein, the target registration priority is the registration priority after updating the user's registration priority, and the waiting queue is the queue obtained by sorting the target registration priority in descending order.

8. The method for determining registration priority according to claim 1, characterized in that, The method further includes: Determine whether the target event has been detected; wherein the target event includes at least one of the following: number release event, user profile change event, and external risk control score update event; If the target event is detected, the registration priority of the user corresponding to the target event is recalculated to obtain the target registration priority; and / or, Based on a preset time window, update the user's wait time, the preset first weight corresponding to the wait time, the preset second weight corresponding to the comprehensive profile evaluation result, and the preset third weight corresponding to the risk coefficient. The target registration priority is determined based on the updated first weight, the updated second weight, and the updated third weight; wherein the target registration priority is the registration priority updated from the user's registration priority.

9. A device for determining registration priority, characterized in that, The device includes: The acquisition module is used to acquire multi-source data; wherein, the multi-source data includes at least user registration data and risk coefficients from external risk control interfaces; A module is established to build a three-dimensional user tagging system based on the multi-source data; The first determining module is used to obtain a comprehensive assessment result of the profile based on the multi-source data and the user three-dimensional tag system; The second determining module is used to determine the user's registration priority based on the comprehensive evaluation results of the profile and the multi-source data.

10. An electronic device, characterized in that, include: A processor and a memory connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to cause the processor to perform the registration priority determination method as described in any one of claims 1-8.

Images