Electronic cattle detection method, intelligent detection platform and medium

Abstract

The invention provides an electronic cattle detection method, an intelligent detection platform and a medium, and relates to the technical field of networks. The user abnormal scene comprises a plurality of single-day registration orders of the same user, high access frequency of the same internet protocol IP address, short time for completing appointment registration, high frequency of changing information of a patient by the user, and high number returning frequency; for each user abnormal scene, adopting a corresponding algorithm to screen out an abnormal user, and identifying the screened abnormal user as an electronic cattle; and in response to a received registration confirmation instruction of the electronic cattle, executing an operation of forbidding number locking on the electronic cattle. According to the method, the intelligent detection platform and the medium, the problems that cattle users cannot be comprehensively blocked in the prior art and time delay exists according to feedback report of the users can be solved.

Description

Technical Field

[0001] The present invention relates to the field of network technology, and in particular to an electronic scalper detection method, an intelligent detection platform and a medium. Background Art

[0002] The online appointment booking platform is a convenient platform that combines efficiency and convenience. After logging into the platform, patients can view information such as hospitals, departments, appointments, etc. on demand. The platform provides personalized services such as SMS reminders, medical image inquiries, and test report inquiries based on patient perception.

[0003] Electronic scalpers refer to people who use electronic technology to register for others, grab appointments, and then transfer them. At present, the appointments of key hospitals have been in short supply, and online appointments are even more tense. Key appointments are often snapped up within five minutes after they are released. "Electronic scalpers" take advantage of the tight number of appointments and use illegal means to grab and resell appointments. In response to this phenomenon, major appointment platforms have sorted out the number-grabbing methods of "electronic scalpers" and taken measures to establish a blacklist. Major key hospitals report suspected scalper users, and jointly collect and organize to create a shared blacklist. The platform has adopted a lock-delay operation for blacklist users, but it has not been able to completely block scalper users, and there are still fish that slip through the net. In addition, according to user feedback reports, the platform has not been able to lock scalpers in the first time, and there is still a delay in time. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an electronic scalper detection method, an intelligent detection platform and a medium in view of the above-mentioned deficiencies in the prior art, so as to solve the problem that the prior art cannot completely block scalper users, and there is a time delay in reporting based on user feedback.

[0005] In a first aspect, the present invention provides a method for detecting electronic scalpers, which is applied to an intelligent detection platform, and the method comprises:

[0006] Obtaining pre-sorted abnormal user scenarios, wherein the abnormal user scenarios include a large number of registration orders by the same user on a single day, a high frequency of access by the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of user changes to patient information, and a high frequency of cancellation;

[0007] For each abnormal user scenario, a corresponding algorithm is used to screen out abnormal users, and the screened abnormal users are identified as electronic scalpers;

[0008] In response to receiving the confirmation registration instruction from the electronic scalper, an operation of prohibiting locking the number is performed on the electronic scalper.

[0009] Furthermore, before obtaining the pre-sorted user abnormal scenarios, the method further includes:

[0010] Through the preset optimization algorithm, abnormal operation behaviors of target platform users are collected, sorted and screened to identify abnormal users;

[0011] The abnormal user scenarios are sorted out according to the abnormal operation behaviors of the identified abnormal users.

[0012] Furthermore, for each abnormal user scenario, a corresponding algorithm is used to filter out abnormal users, specifically including:

[0013] In the scenario where the same user has many registration orders in a single day, the PageRank link analysis algorithm is used to screen out abnormal users with many registration orders in a single day;

[0014] For scenarios where the same IP address has a high access frequency, the lonely forest algorithm is used to filter out abnormal users with a high access frequency from the same IP address;

[0015] For scenarios where it takes a short time to complete an appointment, the autoregressive difference moving average model (ARIMA) algorithm is used to screen out abnormal users who take a short time to complete an appointment.

[0016] For scenarios where users frequently change their patient information, the expectation maximization (EM) iterative algorithm based on the Gaussian model is used to screen out abnormal users who frequently change their patient information.

[0017] For scenarios with high cancellation frequency, the support vector machine (SVM) model optimized based on the artificial bee colony algorithm is used to screen out abnormal users with high cancellation frequency.

[0018] Furthermore, after identifying the screened abnormal users as electronic scalpers, the method further includes:

[0019] Add abnormal users identified as e-scalpers to the blacklist;

[0020] The data in the blacklist is refreshed and recorded into the remote dictionary service Redis module.

[0021] Furthermore, in response to receiving the registration confirmation instruction from the electronic scalper, the operation of prohibiting the electronic scalper from locking the number specifically includes:

[0022] Receive the user's confirmation registration instruction;

[0023] Query user information according to the confirmation registration instruction;

[0024] Checking whether the user is a blacklist user through the Redis module according to the user information;

[0025] In response to the user being a blacklist user, an operation of prohibiting number locking is performed on the blacklist user.

[0026] In a second aspect, the present invention provides an intelligent detection platform, comprising:

[0027] An abnormal scenario acquisition module is used to acquire pre-sorted abnormal user scenarios, wherein the abnormal user scenarios include a large number of registration orders for the same user on a single day, a high frequency of access to the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of user changes to the patient information, and a high frequency of cancellation;

[0028] An electronic scalper identification module is connected to the abnormal scenario acquisition module and is used to filter out abnormal users using a corresponding algorithm for each abnormal scenario of the user, and identify the filtered abnormal users as electronic scalpers;

[0029] The number locking prohibition module is connected to the electronic scalper identification module and is used to execute the number locking prohibition operation on the electronic scalper in response to receiving the registration confirmation instruction from the electronic scalper.

[0030] Furthermore, it also includes:

[0031] The abnormal behavior collection module is used to collect, organize and filter abnormal operation behaviors of target platform users through a preset optimization algorithm to identify abnormal users;

[0032] The abnormal scenario sorting module is used to sort out the user abnormal scenarios according to the abnormal operation behaviors of the identified abnormal users.

[0033] Furthermore, the electronic scalper identification module includes:

[0034] The first screening unit is used to screen out abnormal users with many registration orders in a single day by using the PageRank link analysis algorithm in the scenario where the same user has many registration orders in a single day;

[0035] The second screening unit is used to screen out abnormal users with high access frequency from the same IP address by using the lonely forest algorithm in scenarios where the same IP address has high access frequency;

[0036] The third screening unit is used to screen out abnormal users who complete the appointment registration in a short time by using the autoregressive difference moving average model ARIMA algorithm for the scenario where the appointment registration is completed in a short time;

[0037] The fourth screening unit is used to screen out abnormal users who frequently change their patients' information by using an expectation maximization (EM) iterative algorithm based on a Gaussian model in the scenario where users frequently change their patients' information;

[0038] The fifth screening unit is used to screen out abnormal users with high cancellation frequency by optimizing the support vector machine (SVM) model based on the artificial bee colony algorithm in scenarios with high cancellation frequency.

[0039] In a third aspect, the present invention provides an intelligent detection platform, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the electronic scalper detection method described in the first aspect above.

[0040] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the electronic scalper detection method described in the first aspect is implemented.

[0041] The electronic scalper detection method, intelligent detection platform and medium provided by the present invention first obtain pre-sorted user abnormal scenarios by using the intelligent detection platform, wherein the user abnormal scenarios include a large number of registration orders for the same user on a single day, a high frequency of access to the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of users changing the patient information, and a high frequency of cancellation; then, for each of the user abnormal scenarios, a corresponding algorithm is used to screen out abnormal users, and the screened abnormal users are identified as electronic scalpers; and then, in response to receiving a confirmation registration instruction from the electronic scalper, an operation of prohibiting locking the electronic scalper is performed. The present invention pre-sorts and identifies a variety of abnormal user scenarios through an intelligent detection platform, including multiple registration orders for the same user on a single day, high access frequency of the same IP address, short time to complete appointment registration, high frequency of user changing patient information, and high frequency of withdrawal, so that potential electronic scalpers can be identified more accurately and comprehensively. At the same time, for each user abnormal scenario, the platform uses a corresponding algorithm for screening to ensure the accuracy and efficiency of identification. In addition, when the intelligent detection platform receives a confirmation registration instruction from a user identified as an electronic scalper, it immediately executes the operation of prohibiting the lock number, effectively preventing the registration behavior of the scalper user. Through the real-time monitoring and rapid response of the intelligent detection platform, scalper behavior can be discovered and handled in a timely manner, thereby maintaining the fair allocation and normal use order of medical resources. It solves the problem that the existing technology cannot fully block scalper users, and there is a time delay in reporting based on user feedback. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of a method for detecting electronic scalpers according to Embodiment 1 of the present invention;

[0043] Figure 2 A flowchart of an artificial bee colony algorithm according to an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the structure of an intelligent detection platform according to Embodiment 2 of the present invention;

[0045] Figure 4 This is a schematic diagram of the structure of an intelligent detection platform according to Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the technical solution of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0047] It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention rather than to limit the present invention.

[0048] It can be understood that, in the absence of conflict, the various embodiments of the present invention and the various features in the embodiments can be combined with each other.

[0049] It can be understood that, for the convenience of description, the drawings of the present invention only show the parts related to the present invention, while the parts irrelevant to the present invention are not shown in the drawings.

[0050] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0051] It can be understood that the terms "first", "second", etc. in the embodiments of the present invention are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0052] It can be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the drawings.

[0053] It is understood that the flowcharts and block diagrams of the present invention illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the various embodiments of the present invention. Each box in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented by a hardware-based system that implements the specified functions, or may be implemented by a combination of hardware and computer instructions.

[0054] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented by software or hardware. For example, the units and modules can be located in a processor.

[0055] Embodiment 1:

[0056] This embodiment provides a method for detecting electronic scalpers, which is applied to an intelligent detection platform, such as Figure 1 As shown, the method includes:

[0057] Step S101: Obtain pre-sorted user abnormal scenarios, which include a large number of registration orders for the same user on a single day, a high frequency of access to the same IP (Internet Protocol) address, a short time to complete the appointment registration, a high frequency of user changes to patient information, and a high frequency of cancellation.

[0058] In this embodiment, the intelligent detection platform is set on the target platform, which can be an online reservation platform, an online ticket purchase platform, an online training reservation platform, etc. The intelligent detection platform analyzes the behavior of all users in the target platform and conducts in-depth analysis of the user's behavior pattern to obtain a more comprehensive user abnormal scenario, which includes five scenarios: a large number of registration orders per day by the same user, a high frequency of access from the same IP address, a short time to complete the registration, a high frequency of user changes to the patient's information, and a high frequency of cancellation.

[0059] Optionally, before obtaining the pre-sorted user abnormal scenarios, the method further includes:

[0060] Through the preset optimization algorithm, abnormal operation behaviors of target platform users are collected, sorted and screened to identify abnormal users;

[0061] The abnormal user scenarios are sorted out according to the abnormal operation behaviors of the identified abnormal users.

[0062] In this embodiment, the intelligent detection platform collects, organizes, and screens out abnormal users through an optimization algorithm for abnormal operation behaviors of platform users, and organizes abnormal user scenarios based on the abnormal operation behaviors of the identified abnormal users, wherein the optimization algorithm includes but is not limited to machine learning algorithms, deep learning algorithms, statistical algorithms, and the like.

[0063] Step S102: for each abnormal user scenario, a corresponding algorithm is used to screen out abnormal users, and the screened abnormal users are identified as electronic scalpers.

[0064] In this embodiment, the sorted user abnormal scenarios are used, and corresponding algorithms are adopted according to specific circumstances to screen out abnormal users and identify them as electronic scalpers.

[0065] Optionally, for each abnormal user scenario, using a corresponding algorithm to filter out abnormal users specifically includes:

[0066] For the scenario where the same user has many registration orders in a single day, the PageRank link analysis algorithm is used to filter out abnormal users with many registration orders in a single day;

[0067] For scenarios where the same IP address has a high access frequency, the lonely forest algorithm is used to filter out abnormal users with a high access frequency from the same IP address;

[0068] For scenarios where it takes a short time to complete an appointment, the ARIMA (Autoregressive Integrated Moving Average Model) algorithm is used to screen out abnormal users who take a short time to complete an appointment.

[0069] For scenarios where users frequently change their patient information, the EM (Expectation Maximization) iterative algorithm based on the Gaussian model is used to screen out abnormal users who frequently change their patient information.

[0070] For scenarios with high cancellation frequency, the support vector machine (SVM) model optimized based on the artificial bee colony algorithm is used to screen out abnormal users with high cancellation frequency.

[0071] In this embodiment, for the scenario where the same user has many registration orders on a single day, the intelligent detection platform adds a PageRank link analysis algorithm to screen out abnormal users with many registration orders on a single day. By defining a user association graph, PageRank values ​​are assigned to users. According to the graph structure and the transition probability matrix, multiple iterations are performed until the PageRank values ​​converge. Then, based on the converged PageRank values, users with abnormally high PageRank values ​​are screened out. These users may be users with abnormal registration orders on a single day. A reasonable threshold can be set, and users whose PageRank values ​​exceed the threshold are regarded as abnormal users.

[0072] In this embodiment, for the scenario where the same IP address has a high access frequency, the intelligent detection platform develops the IsolationForest algorithm, selects random features as the basis for segmentation, recursively constructs segmentation trees to create an isolation forest, calculates the path length of data points in the isolation forest, and screens out users with a high IP address access frequency (i.e., users whose IP address access frequency exceeds a certain threshold).

[0073] In this embodiment, in order to solve the abnormal scenario where the time taken to complete the appointment registration is short, the intelligent detection platform develops an algorithm based on the Autoregressive Integrated Moving Average Model (ARIMA), which is used to screen out users with abnormal registration time throughout the entire process (i.e., users whose registration time is less than a certain threshold).

[0074] In this embodiment, for scenarios where users frequently change their patient information, the intelligent detection platform uses the EM (Expectation Maximization) iterative algorithm based on the Gaussian model to screen out users who frequently change their patients (i.e., users who change their patients more than a certain threshold number of times). Specifically, for scenarios where the frequency of changing patient information is high, the EM iterative algorithm based on the Gaussian model screens out users who frequently change their patients. The EM iterative algorithm based on the Gaussian model is generally used to process the maximum likelihood estimation of probability model parameters containing latent variables. In the scenario of screening out users who frequently change their patients, we can regard the user's behavior pattern (such as whether they frequently change their patients) as a latent variable, and estimate the model parameters through the Gaussian mixture model (GMM) and the EM algorithm, and then identify users who frequently change their patients.

[0075] In this embodiment, for scenarios with high cancellation frequency, the intelligent detection platform uses a support vector machine (SVM) model optimized based on an artificial bee colony algorithm, which has high processing efficiency. Using this model, the intelligent detection platform can accurately screen out abnormal users with high cancellation frequency (i.e., users whose cancellation frequency, number of times, or proportion exceeds a preset threshold).

[0076] In an optional embodiment, the abnormal users with high cancellation frequency can be screened out based on the artificial bee colony algorithm to optimize the support vector machine (SVM) model, which may include:

[0077] (1) Collect data:

[0078] Obtain the user's basic information and registration records, including user ID, user IP, registration time, department, doctor, registration status (whether to cancel), number of registrations, number of cancellations, and other information.

[0079] (2) Data preprocessing:

[0080] Data cleaning: Remove missing values, outliers, and irrelevant data to ensure data quality.

[0081] The data is normalized to improve the training efficiency of the SVM model.

[0082] (3) Building the SVM model

[0083] Select features related to cancellation behavior, such as the number of registrations, the number of cancellations, the popularity of the registration department, etc.

[0084] The collected data set is used to train the SVM model to distinguish normal users from abnormal users (users with high account cancellation frequency).

[0085] Preliminary settings are made for the parameters of the SVM model (such as penalty coefficient C, kernel function parameters, etc.).

[0086] (4) Artificial bee colony algorithm optimizes the SVM model

[0087] Initialize the bee colony: Randomly initialize a set of solutions in the parameter space of the SVM model as the initial positions of the bees.

[0088] Iterative optimization: Bees search for new SVM parameter combinations near their location and calculate fitness (such as classification accuracy).

[0089] The follower bee selects the object to follow according to the fitness of the collecting bee and performs a neighborhood search near it.

[0090] If a solution has not been updated after multiple iterations, the corresponding bee is transformed into a scout bee to randomly search for new solutions in the solution space.

[0091] Termination condition: When the maximum number of iterations is reached or other termination conditions are met, the iteration stops.

[0092] Finally, the optimal SVM parameter combination is output.

[0093] (5) Screening abnormal users

[0094] Apply the optimized SVM model to new registration data to classify and predict the user's cancellation behavior. Set a threshold based on actual needs, such as the ratio of cancellation times to total registration times, and filter out users predicted to be abnormal users whose cancellation ratio exceeds the threshold to obtain abnormal users with high cancellation frequency.

[0095] Optionally, after identifying the screened abnormal users as electronic scalpers, the method further includes:

[0096] Add abnormal users identified as e-scalpers to the blacklist;

[0097] The data in the blacklist is refreshed and entered into the Redis (remote dictionary service) module.

[0098] In this embodiment, in order to ensure that the blacklist information stored in the Redis module is up to date, once a new e-scalper is identified, the blacklist user data in the Redis module is instantly updated by refreshing the entry.

[0099] Step S103: in response to receiving the registration confirmation instruction from the electronic scalper, executing an operation of prohibiting the electronic scalper from locking the number.

[0100] In this embodiment, after the user confirms the registration, if it is determined that the user is an electronic scalper, the intelligent detection platform will execute a prohibition operation on the user's registration request, that is, the reserved number source will not be locked, thereby preventing the user from further abusing the number source.

[0101] Optionally, in response to receiving the registration confirmation instruction from the electronic scalper, performing an operation of prohibiting the electronic scalper from locking the number, specifically includes:

[0102] Receive the user's confirmation registration instruction;

[0103] Query user information according to the confirmation registration instruction;

[0104] Checking whether the user is a blacklist user through the Redis module according to the user information;

[0105] In response to the user being a blacklist user, an operation of prohibiting number locking is performed on the blacklist user.

[0106] In this embodiment, in order to efficiently complete the information verification, the user information is compared with the data stored in the Rdeis module to verify whether the user is a blacklist user. If the user is confirmed to be a blacklist user, the platform will prohibit the user's number locking operation. This means that even if the user successfully reserves a number source, the system will not lock it, but directly reject its reservation, thereby preventing the blacklist user from continuing to abuse the number source, thereby protecting the rights and interests of other users.

[0107] It should be noted that the electronic scalper detection method provided by the present invention can efficiently and low-costly detect the user's operating behavior in real time, and take multiple prevention and control measures. It innovatively integrates multiple algorithm technologies to organize, collect, and filter abnormal user behaviors, and detects from various key nodes of user operations, including rapid locking of number source operations, repeated operations of appointment and cancellation operations, and analysis of abnormal operations such as no correlation between the information of bound patients and regional dispersion, lock the electronic scalpers, and then take blacklist ban measures. Among them, the Rdeis module stores data for comparison, and can efficiently complete information verification. Once it meets the configuration information, it will be banned immediately.

[0108] In a specific embodiment, the electronic scalper detection method is applied to an online reservation platform. For online reservation number sources, the platform includes a Mobile module, a Web background management module, a configuration center, a scheduled task, a Mysql module, a Redis (remote dictionary service) module, and a message queue module. The online reservation platform includes an intelligent detection platform for sorting and analyzing the behavior of 30 million users on the platform, conducting in-depth analysis of user behavior patterns, and combining advanced data mining technology and modeling methods to accurately assist in locating "electronic scalpers", and blocking the platform number and other rules for such users.

[0109] The specific implementation steps include:

[0110] S1. The intelligent detection platform collects and organizes abnormal operation behaviors of platform users through optimization algorithms and screens out abnormal users.

[0111] Among them, the optimization algorithm includes but is not limited to machine learning algorithm, deep learning algorithm, statistical algorithm, etc.

[0112] S2. Organize the abnormal scenarios of users, take corresponding measures to screen out abnormal users according to the specific situation, and add them to the blacklist for blacklisting / delayed account locking.

[0113] S3. In response to the abnormal situation where the same user has many registration orders in a single day, the intelligent detection platform adds the PageRank link analysis algorithm to screen out abnormal users with many registration orders in a single day. By defining the user association graph, the PageRank value is assigned to the user. According to the graph structure and the transition probability matrix, multiple iterations are performed until the PageRank value converges, thereby improving the screening rate of abnormal users.

[0114] Specifically, we first conduct model research, then develop the PageRank algorithm, continuously perform parameter optimization and performance testing, and finally deploy it into the system. We count the number of users who pay for registration orders that exceed the appointment rule limit every day, and use the user association graph established by the platform's patients to assign PageRank values ​​to users. Based on the graph structure and transition probability matrix, we iterate multiple times until the PageRank value converges.

[0115] It should be noted that by analyzing all the order data of the platform, adding PageRank links, and calculating offline rather than during search, the query efficiency is greatly improved, and then the data with a large number of registered orders from the same user on a single day are filtered out for blacklist verification.

[0116] S4. For abnormal scenarios with high access frequency from the same IP address, the intelligent detection platform develops the IsolationForest algorithm, selects random features as the basis for segmentation, recursively constructs segmentation trees to create an isolation forest, calculates the path length of data points in the isolation forest, and screens out users with high IP address access frequency.

[0117] Specifically, the access IP addresses of platform users are detected, and the high access frequency features of IP addresses are used as the basis for segmentation. The segmentation tree is recursively constructed to create an isolation forest, and the path length indicates the difference between the data.

[0118] S5. For the abnormal scenario of short time to complete the appointment registration, the intelligent detection platform develops an algorithm based on the Autoregressive Integrated Moving Average Model (ARIMA), which screens out users with abnormal registration time throughout the whole process. The formula of ARIMA is as follows:

[0119]

[0120] Where Y t is the current value, c is the constant term, q is the order, θ i is the correlation coefficient, ∈ t is the error, and ∈ t Must conform to normal distribution.

[0121] S6. For scenarios where users frequently change their patient information, the intelligent detection platform uses the EM (Expectation Maximization) iterative algorithm based on the Gaussian model to screen out users who frequently change their patient information.

[0122] Specifically, for scenarios where the frequency of changing patient information is high, the EM iterative algorithm based on the Gaussian model is used to screen users who frequently change their patients. The EM iterative algorithm based on the Gaussian model is usually used to process the maximum likelihood estimation of probability model parameters containing latent variables. In the scenario of screening users who frequently change their patients, we can regard the user's behavior pattern (such as whether they frequently change their patients) as a latent variable, and estimate the model parameters through the Gaussian mixture model (GMM) and the EM algorithm, and then identify users who frequently change their patients.

[0123] S7. For abnormal users with high account cancellation frequency, the intelligent detection platform uses the support vector machine (SVM) model optimized by the artificial bee colony algorithm, which has high processing efficiency. Using this model, the intelligent detection platform can accurately screen out abnormal users with high account cancellation frequency.

[0124] Among them, the steps of the artificial bee colony algorithm are as follows Figure 2 As shown, it includes: initializing parameters and generating an initial position; finding the initial nectar source, leading the bees to explore and mark the position; following the bees to search and change the nectar source mark, and judging whether there are scout bees. If so, generating a new nectar source to replace the old nectar source and the current generation nectar source. If not, judging whether the termination condition is met. If so, outputting the optimized nectar source and ending the algorithm.

[0125] S8. Integrate and analyze the measures in S3-S7, screen out suspected scalpers, add them to the blacklist, operate the Qschedule task scheduling module to refresh the data and enter it into the Redis module. When the user places an order, the blacklist is checked and compared with the blacklist information stored in Redis. After matching information is detected, the blacklist is banned and the account cannot be locked.

[0126] Specifically, for the following types of users: those who have many registration orders in a single day, those who have a high frequency of access from the same IP address, those who take a short time to complete appointment registration, those who change patient information frequently, and those who cancel their appointments frequently, the platform will extract the user's information, lock the mobile phone number and IP information, and enter the user into the blacklist.

[0127] Specifically, the blocking process includes: after the user confirms the registration, the intelligent detection platform queries the user information and uses the redis module to verify the blacklist user. If yes, the number is prohibited from being locked, otherwise, the appointment is successful.

[0128] It should be noted that prohibiting number locking means that when the system confirms the registration behavior, it will query the user information and verify whether the user is a blacklist user through the redis module. If the user is confirmed as a blacklist user, the platform will prohibit the number locking operation for the user. This means that even if the user successfully reserves a number, the system will not lock it, but directly reject its reservation, thereby preventing blacklist users from continuing to abuse the number, thereby protecting the rights and interests of other users.

[0129] It should be noted that the electronic scalper detection method provided by the present invention can form a complete intelligent detection platform based on the user operation model, screen the user behavior data of the platform, perform background detection without the user's knowledge, accurately match scalper users, and configure a blacklist. The core technical point of the present invention is not only to detect abnormal user behavior, but also to increase prevention and control measures for such users, which has a significant effect on preventing scalpers. The average accuracy rate of suspected scalpers has risen to 84%, and the results have been recognized by users. In addition, the "large number of abnormal registration and cancellation records" and "large number of abnormal patients bound" have been reduced after being banned by the blacklist, and no new problems have been generated. The improvement is effective.

[0130] The electronic scalper detection method provided in the embodiment of the present invention first obtains pre-sorted user abnormal scenarios by using an intelligent detection platform, wherein the user abnormal scenarios include a large number of registration orders for the same user on a single day, a high frequency of access to the same Internet Protocol IP address, a short time to complete appointment registration, a high frequency of users changing patient information, and a high frequency of cancellation; then, for each of the user abnormal scenarios, a corresponding algorithm is used to screen out abnormal users, and the screened out abnormal users are identified as electronic scalpers; and then, in response to receiving a registration confirmation instruction from the electronic scalper, an operation of prohibiting locking the electronic scalper is performed. The present invention pre-sorts and identifies a variety of abnormal user scenarios through an intelligent detection platform, including multiple registration orders for the same user on a single day, high access frequency of the same IP address, short time to complete appointment registration, high frequency of user changing patient information, and high frequency of withdrawal, so that potential electronic scalpers can be identified more accurately and comprehensively. At the same time, for each user abnormal scenario, the platform uses a corresponding algorithm for screening to ensure the accuracy and efficiency of identification. In addition, when the intelligent detection platform receives a confirmation registration instruction from a user identified as an electronic scalper, it immediately executes the operation of prohibiting the lock number, effectively preventing the registration behavior of the scalper user. Through the real-time monitoring and rapid response of the intelligent detection platform, scalper behavior can be discovered and handled in a timely manner, thereby maintaining the fair allocation and normal use order of medical resources. It solves the problem that the existing technology cannot fully block scalper users, and there is a time delay in reporting based on user feedback.

[0131] Embodiment 2:

[0132] like Figure 3 As shown, this embodiment provides an intelligent detection platform for executing the above-mentioned electronic scalper detection method, including:

[0133] The abnormal scenario acquisition module 11 is used to acquire pre-sorted abnormal user scenarios, wherein the abnormal user scenarios include a large number of registration orders for the same user on a single day, a high frequency of access to the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of user changes to the patient information, and a high frequency of cancellation;

[0134] The electronic scalper identification module 12 is connected to the abnormal scenario acquisition module 11, and is used to filter out abnormal users using a corresponding algorithm for each abnormal scenario of the user, and identify the filtered abnormal users as electronic scalpers;

[0135] The number locking prohibition module 13 is connected to the electronic scalper identification module 12, and is used for executing the number locking prohibition operation on the electronic scalper in response to receiving the registration confirmation instruction from the electronic scalper.

[0136] Optionally, it also includes:

[0137] The abnormal behavior collection module is used to collect, organize and filter abnormal operation behaviors of target platform users through a preset optimization algorithm to identify abnormal users;

[0138] The abnormal scenario sorting module is used to sort out the user abnormal scenarios according to the abnormal operation behaviors of the identified abnormal users.

[0139] Optionally, the electronic scalper identification module 12 includes:

[0140] The first screening unit is used to screen out abnormal users with many registration orders in a single day by using the PageRank link analysis algorithm in the scenario where the same user has many registration orders in a single day;

[0141] The second screening unit is used to screen out abnormal users with high access frequency from the same IP address by using the lonely forest algorithm in scenarios where the same IP address has high access frequency;

[0142] The third screening unit is used to screen out abnormal users who complete the appointment registration in a short time by using the autoregressive difference moving average model ARIMA algorithm for the scenario where the appointment registration is completed in a short time;

[0143] The fourth screening unit is used to screen out abnormal users who frequently change their patients' information by using an expectation maximization (EM) iterative algorithm based on a Gaussian model in the scenario where users frequently change their patients' information;

[0144] The fifth screening unit is used to screen out abnormal users with high cancellation frequency by optimizing the support vector machine (SVM) model based on the artificial bee colony algorithm in scenarios with high cancellation frequency.

[0145] Optionally, it also includes:

[0146] A blacklist adding module is used to add abnormal users identified as e-scalpers to the blacklist;

[0147] The refresh entry module is used to refresh the data in the blacklist and enter it into the remote dictionary service Redis module.

[0148] Optionally, the number locking prohibition module 13 includes:

[0149] An instruction receiving unit, used for receiving a user's confirmation registration instruction;

[0150] A user information query unit, used to query user information according to the confirmation registration instruction;

[0151] A blacklist user verification unit, used to verify whether the user is a blacklist user through the Redis module according to the user information;

[0152] The number locking prohibiting unit is used for executing a number locking prohibiting operation on the blacklist user in response to the user being a blacklist user.

[0153] Embodiment 3:

[0154] refer to Figure 4 This embodiment provides an intelligent detection platform, including a memory 21 and a processor 22. The memory 21 stores a computer program, and the processor 22 is configured to run the computer program to execute the electronic scalper detection method in Example 1.

[0155] The memory 21 is connected to the processor 22. The memory 21 may be a flash memory, a read-only memory or other memory. The processor 22 may be a central processing unit or a single-chip microcomputer.

[0156] Embodiment 4:

[0157] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the electronic scalper detection method in the above-mentioned embodiment 1 is implemented.

[0158] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0159] To summarize, the electronic scalper detection method, intelligent detection platform and medium provided in the embodiments of the present invention utilize the intelligent detection platform to first obtain pre-sorted user abnormal scenarios, wherein the user abnormal scenarios include a large number of registration orders per day by the same user, a high frequency of access to the same Internet Protocol IP address, a short time to complete appointment registration, a high frequency of users changing patient information, and a high frequency of cancellation; then, for each of the user abnormal scenarios, a corresponding algorithm is used to screen out abnormal users, and the screened out abnormal users are identified as electronic scalpers; and then, in response to receiving a registration confirmation instruction from the electronic scalper, an operation of prohibiting locking the electronic scalper is performed. The present invention pre-sorts and identifies a variety of abnormal user scenarios through an intelligent detection platform, including multiple registration orders for the same user on a single day, high access frequency of the same IP address, short time to complete appointment registration, high frequency of user changing patient information, and high frequency of withdrawal, so that potential electronic scalpers can be identified more accurately and comprehensively. At the same time, for each user abnormal scenario, the platform uses a corresponding algorithm for screening to ensure the accuracy and efficiency of identification. In addition, when the intelligent detection platform receives a confirmation registration instruction from a user identified as an electronic scalper, it immediately executes the operation of prohibiting the lock number, effectively preventing the registration behavior of the scalper user. Through the real-time monitoring and rapid response of the intelligent detection platform, scalper behavior can be discovered and handled in a timely manner, thereby maintaining the fair allocation and normal use order of medical resources. It solves the problem that the existing technology cannot fully block scalper users, and there is a time delay in reporting based on user feedback.

[0160] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for detecting electronic scalpers, characterized in that: Applied to an intelligent detection platform, the method comprises: Obtaining pre-sorted abnormal user scenarios, wherein the abnormal user scenarios include a large number of registration orders by the same user on a single day, a high frequency of access by the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of user changes to patient information, and a high frequency of cancellation; For each abnormal user scenario, a corresponding algorithm is used to screen out abnormal users, and the screened abnormal users are identified as electronic scalpers; In response to receiving the confirmation registration instruction from the electronic scalper, an operation of prohibiting locking the number is performed on the electronic scalper.

2. The method according to claim 1, characterized in that Before obtaining the pre-sorted user abnormal scenarios, the method further includes: Through the preset optimization algorithm, abnormal operation behaviors of target platform users are collected, sorted and screened to identify abnormal users; The abnormal user scenarios are sorted out according to the abnormal operation behaviors of the identified abnormal users.

3. The method according to claim 1, characterized in that For each abnormal user scenario, a corresponding algorithm is used to filter out abnormal users, specifically including: In the scenario where the same user has many registration orders in a single day, the PageRank link analysis algorithm is used to screen out abnormal users with many registration orders in a single day; For scenarios where the same IP address has a high access frequency, the lonely forest algorithm is used to filter out abnormal users with a high access frequency from the same IP address; For scenarios where it takes a short time to complete an appointment, the autoregressive difference moving average model (ARIMA) algorithm is used to screen out abnormal users who take a short time to complete an appointment. For scenarios where users frequently change their patient information, the expectation maximization (EM) iterative algorithm based on the Gaussian model is used to screen out abnormal users who frequently change their patient information. For scenarios with high cancellation frequency, the support vector machine (SVM) model optimized based on the artificial bee colony algorithm is used to screen out abnormal users with high cancellation frequency.

4. The method according to claim 1, characterized in that: After identifying the screened abnormal users as electronic scalpers, the method further includes: Add abnormal users identified as e-scalpers to the blacklist; The data in the blacklist is refreshed and recorded into the remote dictionary service Redis module.

5. The method according to claim 4, characterized in that In response to receiving the registration confirmation instruction from the electronic scalper, the operation of prohibiting the electronic scalper from locking the number specifically includes: Receive the user's confirmation registration instruction; Query user information according to the confirmation registration instruction; Checking whether the user is a blacklist user through the Redis module according to the user information; In response to the user being a blacklist user, an operation of prohibiting number locking is performed on the blacklist user.

6. An intelligent detection platform, characterized in that: include: An abnormal scenario acquisition module is used to acquire pre-sorted abnormal user scenarios, wherein the abnormal user scenarios include a large number of registration orders for the same user on a single day, a high frequency of access to the same Internet Protocol IP address, a short time to complete the appointment registration, a high frequency of user changes to the patient information, and a high frequency of cancellation; An electronic scalper identification module is connected to the abnormal scenario acquisition module and is used to filter out abnormal users using a corresponding algorithm for each abnormal scenario of the user, and identify the filtered abnormal users as electronic scalpers; The number locking prohibition module is connected to the electronic scalper identification module and is used to execute the number locking prohibition operation on the electronic scalper in response to receiving the registration confirmation instruction from the electronic scalper.

7. The intelligent detection platform according to claim 6, characterized in that: Also includes: The abnormal behavior collection module is used to collect, organize and filter abnormal operation behaviors of target platform users through a preset optimization algorithm to identify abnormal users; The abnormal scenario sorting module is used to sort out the user abnormal scenarios according to the abnormal operation behaviors of the identified abnormal users.

8. The intelligent detection platform according to claim 6, characterized in that: The electronic scalper identification module comprises: The first screening unit is used to screen out abnormal users with many registration orders in a single day by using the PageRank link analysis algorithm in the scenario where the same user has many registration orders in a single day; The second screening unit is used to screen out abnormal users with high access frequency from the same IP address by using the lonely forest algorithm in scenarios where the same IP address has high access frequency; The third screening unit is used to screen out abnormal users who complete the appointment registration in a short time by using the autoregressive difference moving average model ARIMA algorithm for the scenario where the appointment registration is completed in a short time; The fourth screening unit is used to screen out abnormal users who frequently change their patients' information by using an expectation maximization (EM) iterative algorithm based on a Gaussian model in the scenario where users frequently change their patients' information; The fifth screening unit is used to screen out abnormal users with high cancellation frequency by optimizing the support vector machine (SVM) model based on the artificial bee colony algorithm in scenarios with high cancellation frequency.

9. An intelligent detection platform, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the electronic scalper detection method as described in any one of claims 1-5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the electronic scalper detection method according to any one of claims 1 to 5 is implemented.

Images