Processing method and device for user breakpoint operation, equipment, medium and product

Abstract

The invention relates to a processing method and device for user breakpoint operation, equipment, a medium and a product. The method comprises the steps of obtaining attribute information and service behavior data of a user for a target service in response to a processing request of the user for the target service, determining service behavior characteristics according to the attribute information and the service behavior data, and determining the target service according to the service behavior characteristics. And determining a breakpoint detection result, determining a breakpoint processing strategy according to the breakpoint detection result under the condition that the breakpoint detection result represents that the user has the breakpoint operation, and processing the breakpoint operation of the user by adopting the breakpoint processing strategy. By adopting the method, the reliability of financial service processing of a user can be ensured.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, medium and product for processing user breakpoint operations. Background Technology

[0002] With the continuous development of the financial sector, in order to ensure the smoothness of user operations when users perform business operations, methods for detecting user breakpoints have emerged. The existing methods generally involve placing data points at key business nodes and analyzing the data points to obtain the results of user breakpoint behavior detection.

[0003] However, since the data used in the relevant technologies is offline data, it is impossible to provide real-time feedback after detecting user breakpoint operations, which reduces the reliability of users' financial transactions. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, equipment, medium, and product for handling user breakpoint operations that can ensure the reliability of users' financial business processing, in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for handling user breakpoint operations, including:

[0006] In response to a user's processing request for a target business, obtain attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0007] Based on attribute information and business behavior data, determine the characteristics of business behavior;

[0008] Determine the breakpoint detection results based on the characteristics of business behavior;

[0009] If the breakpoint detection results indicate that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and the breakpoint handling strategy is used to handle the user's breakpoint operation.

[0010] In one embodiment, business behavior characteristics include static behavior characteristics, dynamic behavior characteristics, and business environment characteristics; determining business behavior characteristics based on attribute information and business behavior data includes:

[0011] Based on business behavior data, determine dynamic behavioral characteristics;

[0012] Based on attribute information and business behavior data, determine static behavioral characteristics and business environment characteristics.

[0013] In one embodiment, dynamic behavioral characteristics include processing repetition characteristics, page dwell characteristics, operation smoothness characteristics, and operation backtracking characteristics; based on business behavior data, dynamic behavioral characteristics are determined, including:

[0014] Based on the number of times the same control is processed repeatedly in the business behavior data, the characteristics of repeated processing are determined;

[0015] Based on the page dwell time of each page in the business behavior data, determine the page dwell characteristics;

[0016] Based on the duration of the operation interval between consecutive interactive operations in the business behavior data, determine the operation smoothness characteristics;

[0017] Based on the backtracking operation data in the business behavior data, determine the operation backtracking characteristics.

[0018] In one embodiment, determining the breakpoint detection result based on business behavior characteristics includes:

[0019] A probabilistic prediction model is used to process business behavior characteristics and obtain breakpoint probabilities.

[0020] If a user's operation is determined to have a breakpoint based on the breakpoint probability, the breakpoint type is determined based on the business behavior characteristics and the breakpoint probability.

[0021] The breakpoint detection result is determined based on the breakpoint probability and breakpoint type.

[0022] In one embodiment, determining that a user has a breakpoint operation based on the breakpoint probability includes:

[0023] Determine the probability threshold based on the business step to which the user's current interaction with the target business belongs within the target business;

[0024] If the probability of a breakpoint is greater than a probability threshold, it is determined that the user has a breakpoint operation.

[0025] In one embodiment, a breakpoint handling strategy is determined based on the breakpoint detection results, including:

[0026] Based on the breakpoint detection results, the user's user identification information, and the user's current interaction behavior with the target business, a target matching vector is generated;

[0027] Using the target matching vector as an index, a query is performed in the strategy library to obtain the breakpoint handling strategy associated with the target matching vector; the strategy library contains the breakpoint handling strategies associated with each candidate matching vector.

[0028] Secondly, this application also provides a processing apparatus for user breakpoint operations, including:

[0029] The data acquisition module is used to respond to user processing requests for the target business and acquire attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0030] The feature determination module is used to determine business behavior features based on attribute information and business behavior data;

[0031] The result determination module is used to determine the breakpoint detection result based on the characteristics of business behavior.

[0032] The breakpoint handling module is used to determine the breakpoint handling strategy based on the breakpoint detection results when the breakpoint detection results indicate that the user has a breakpoint operation, and then use the breakpoint handling strategy to process the user's breakpoint operation.

[0033] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0034] In response to a user's processing request for a target business, obtain attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0035] Based on attribute information and business behavior data, determine the characteristics of business behavior;

[0036] Determine the breakpoint detection results based on the characteristics of business behavior;

[0037] If the breakpoint detection results indicate that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and the breakpoint handling strategy is used to handle the user's breakpoint operation.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0039] In response to a user's processing request for a target business, obtain attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0040] Based on attribute information and business behavior data, determine the characteristics of business behavior;

[0041] Determine the breakpoint detection results based on the characteristics of business behavior;

[0042] If the breakpoint detection results indicate that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and the breakpoint handling strategy is used to handle the user's breakpoint operation.

[0043] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0044] In response to a user's processing request for a target business, obtain attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0045] Based on attribute information and business behavior data, determine the characteristics of business behavior;

[0046] Determine the breakpoint detection results based on the characteristics of business behavior;

[0047] If the breakpoint detection results indicate that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and the breakpoint handling strategy is used to handle the user's breakpoint operation.

[0048] The aforementioned methods, apparatus, devices, media, and products for handling user breakpoint operations determine business behavior characteristics based on attribute information and user business behavior data related to the target business, and then determine breakpoint detection results based on these characteristics. Subsequently, if the breakpoint detection results indicate that the user has performed a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and this strategy is then applied to handle the user's breakpoint operation. Compared to related technologies that use offline data analysis to identify user breakpoint operations, the above method, by acquiring user business behavior data in real time to determine breakpoint detection results and corresponding breakpoint handling strategies, enables real-time processing of user breakpoint behavior during the user's handling of the target business, thereby improving the reliability of user business operations. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a method for handling user breakpoint operations in one embodiment;

[0051] Figure 2This is a flowchart illustrating the process of determining business behavior characteristics in one embodiment;

[0052] Figure 3 This is a flowchart illustrating the process of determining dynamic behavioral characteristics in one embodiment;

[0053] Figure 4 This is a flowchart illustrating the process of determining breakpoint detection results in one embodiment;

[0054] Figure 5 This is a flowchart illustrating the process of determining a breakpoint handling strategy in one embodiment;

[0055] Figure 6 This is a flowchart illustrating a method for handling user breakpoint operations in another embodiment;

[0056] Figure 7 This is a structural block diagram of a processing device for user breakpoint operations in one embodiment;

[0057] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] With the continuous development of the financial sector, in order to ensure the smoothness of user operations when users perform business operations, methods for detecting user breakpoints have emerged. The existing methods generally involve placing data points at key business nodes and analyzing the data points to obtain the results of user breakpoint behavior detection.

[0060] However, since the data used in the relevant technologies is offline data, it is impossible to provide real-time feedback after detecting user breakpoint operations, which reduces the reliability of users' financial transactions.

[0061] Based on this, in an exemplary embodiment, a method for handling user breakpoint operations is provided. The method is illustrated using an application to a server as an example. Figure 1 As shown, the specific steps include:

[0062] S101, in response to the user's processing request for the target business, obtains attribute information and user business behavior data for the target business.

[0063] The "target business" refers to the financial transaction the user intends to process. The "processing request" is the request sent by the user when executing the target business. The "attribute information" includes user attribute information and business attribute information. User attribute information refers to user-related data, which may include, but is not limited to, the user's location and occupation. Business attribute information refers to attribute data related to the target business, which may include, but is not limited to, the business execution process. Business behavior data refers to the behavioral data generated when the user performs related operations during the execution of the target business.

[0064] Optionally, users can initiate processing requests for target services through mobile devices. After detecting the processing request, attribute information can be obtained, and user business behavior data can be obtained in real time during the user's processing of the target service.

[0065] For example, user interaction data can be collected by embedding high-granularity tracking code (SDK) in the mobile device. Specifically, every user click can be recorded, including but not limited to buttons, links, tabs, radio buttons / checkboxes, and input boxes, and associated with their element identifiers, page addresses, and timestamps. The complete content and corresponding time of user input, deletion, focusing, and defocusing events in each input box (such as name, ID number, income amount) can be recorded. Precise timestamps of user entry and exit from the page can be recorded. Key elements of normal exposure when a user visits the page can be recorded, including the precise status of the results page.

[0066] Business process data is obtained through front-end and back-end transaction logs and front-end specific result tracking events. Specifically, key step transitions when users process target business can be pre-marked, such as: "Application Entry Page" -> "Information Filling Page" -> "Document Signing Page" -> "Application Result Page". Then, the precise timestamps of users entering and leaving each step are recorded to calculate the dwell time within each step, and to explicitly record page back operations performed by users through the back function.

[0067] Business environment data can be obtained directly from the mobile device's configuration file. Specifically, device information (including device model, operating system, system version, etc.), network environment (recording the network type when the user operates), and session identifier (generating a globally unique session ID for each independent access session of the user and associating it with the customer's unique identifier (User ID) to distinguish access from different users and different accesses from the same user) can be obtained from the mobile device's configuration file.

[0068] Understandably, since business behavior data is multi-source, preprocessing is necessary to ensure the rationality of subsequent processing. Specifically, this involves filtering invalid data, i.e., automatically filtering invalid or duplicate data packets caused by front-end code errors or network fluctuations. Outliers in the business behavior data are handled, i.e., identifying and processing outliers based on statistical methods and the business rules of the target business. For example, data where the total review time for a single document exceeds 24 hours, or the dwell time on a single page exceeds a preset threshold, will be marked and excluded from the dataset. Logical consistency checks are performed on the business behavior data, i.e., verifying the logical rationality of the data, for example, ensuring that the timestamp of leaving a page is not earlier than the timestamp of entering that page.

[0069] Furthermore, session segmentation is performed on the preprocessed business behavior data. Specifically, based on the session ID and combined with a business timeout threshold, the continuous stream of user behavior events is segmented. If the time interval between two adjacent events exceeds this threshold, it is considered the end of one session and the beginning of a new session, thus organizing the data into independent request sessions in units of "times".

[0070] The business behavior data is processed into a behavior serialization sequence. First, based on the actual business process sequence (e.g., the application process must precede document signing), the status of the same customer in the current business process is distinguished, and a user process-dimensional process event sequence S1={D1,D2,...,Dn} is constructed, where each process event D contains the event type and time information of the customer's complete process start and end. At the same time, within each independent session, all collected events (clicks, inputs, jumps, etc.) are strictly sorted according to their timestamps to construct a user behavior event sequence S2={E1,E2,...,En}, where each event E contains the event type, attributes, and time information of a single customer behavior operation.

[0071] In addition, data standardization can be performed. For example, time fields can be standardized by converting all timestamps to the same time zone (e.g., UTC time) and format. Categorical data can be encoded, such as non-numerical data like device type and browser type, converting them into numerical form. Numerical data can be normalized, such as dwell time and number of characters entered, to eliminate the influence of units and accelerate model convergence. Each behavioral event can be abstracted into a unified structured object, such as: {session ID, user ID, event type, event timestamp, page address, interactive element ID, additional attributes}.

[0072] S102, Determine business behavior characteristics based on attribute information and business behavior data.

[0073] Among them, the so-called business behavior characteristics are feature vectors that characterize user behavior.

[0074] Optionally, the preprocessed attribute information and business behavior data can be input into a trained feature extraction model, which will then output business behavior features based on the attribute information and business behavior data.

[0075] For example, the preprocessed process event sequence S1 and user behavior event sequence S2, along with attribute information, can be input into a trained feature extraction model. The feature extraction model then outputs business behavior features based on the process event sequence S1, user behavior event sequence S2, and attribute information.

[0076] S103, Determine the breakpoint detection results based on business behavior characteristics.

[0077] The so-called breakpoint detection result is the detection result of whether the user has performed a breakpoint operation.

[0078] Optionally, business behavior characteristics can be analyzed to determine whether there are behavioral characteristics corresponding to user breakpoint operations, thereby obtaining breakpoint detection results. Alternatively, business behavior characteristics can be input into the breakpoint detection model, which will then output breakpoint detection results based on the business behavior characteristics.

[0079] S104. If the breakpoint detection result indicates that the user has a breakpoint operation, determine the breakpoint handling strategy based on the breakpoint detection result, and use the breakpoint handling strategy to handle the user's breakpoint operation.

[0080] Among them, the so-called breakpoint handling strategy refers to the relevant strategies for handling user breakpoint operations.

[0081] Optionally, if the breakpoint detection result indicates that the user has not performed any breakpoint operations, user behavior data can continue to be monitored. If the breakpoint detection result indicates that the user has performed any breakpoint operations, an appropriate breakpoint handling strategy can be selected based on the severity of the breakpoints indicated by the breakpoint detection result. Then, the breakpoint handling strategy can be used to call each breakpoint response unit to process the user's breakpoint operations.

[0082] For example, a breakpoint response unit may include a front-end response unit and a back-end response unit. The front-end response unit can be used for intelligent pop-ups and notifications (e.g., non-intrusive pop-up notifications, promotional announcements, etc. within the application interface), dynamic content loading (dynamically loading and displaying preset content, such as "Frequently Asked Questions" or "Highly Recommended" tags, in specific areas of the page), and integration with online customer service systems (e.g., automatically activating the online customer service chat window and pushing the user's current page and potential problems as contextual information to customer service personnel for precise assistance).

[0083] The backend response unit can be used for SMS outreach (e.g., sending preset SMS messages to high-value or high-probability users within a short period of time after a user session is interrupted), intelligent outbound calling systems (e.g., for users identified as "high-risk abandonment", AI voice or human customer service can follow up to understand the reasons for abandonment and provide assistance), and push notifications (e.g., sending reminder messages in the application platform).

[0084] The above-described method for handling user breakpoint operations determines business behavior characteristics based on attribute information and user business behavior data related to the target business, and then determines breakpoint detection results based on these characteristics. Subsequently, if the breakpoint detection results indicate that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection results, and this strategy is then applied to handle the user's breakpoint operation. Compared to related technologies that use offline data analysis to identify user breakpoint operations, this method, by acquiring user business behavior data in real time to determine breakpoint detection results and corresponding breakpoint handling strategies, enables real-time processing of user breakpoint behavior during the user's handling of the target business, thereby improving the reliability of user business operations.

[0085] Based on the above embodiments, in this application embodiment, business behavior characteristics include static behavior characteristics, dynamic behavior characteristics, and business environment characteristics. Furthermore, an optional method for determining business behavior characteristics is provided, such as... Figure 2 As shown, the specific steps include:

[0086] S201, Determine dynamic behavioral characteristics based on business behavior data.

[0087] Among these, dynamic behavioral characteristics refer to behavioral characteristics under the dimension of dynamic user operations. Business behavioral data includes user interaction data, business process data, and business environment data. Furthermore, user interaction data refers to the interaction data between the user and the information display page corresponding to the target business; business process data refers to the processing data related to each business process when the user handles the target business; and business environment data refers to data related to the environment in which the target business is processed.

[0088] Optionally, business behavior data can be input into a trained dynamic feature determination model, which then outputs dynamic behavior features based on the business behavior data. Alternatively, for business behavior data across different dimensions, feature extraction methods specific to that dimension can be used to determine the dynamic behavior features for that dimension.

[0089] S202, Based on attribute information and business behavior data, determine static behavior characteristics and business environment characteristics.

[0090] Among these, the so-called business environment characteristics refer to the relevant characteristics of the user's operating environment. The so-called static behavioral characteristics refer to the behavioral characteristics of users under static statistical dimensions. Furthermore, the so-called static behavioral characteristics include basic attribute characteristics, business completion characteristics, and total time consumption characteristics, etc.

[0091] Optionally, fixed parameters can be extracted from user attribute information and business attribute information as basic attribute features. For example, fixed parameters could be form information filled out by the user, such as user region, occupation, and applied resource information. Based on the total number of steps in the target business process in the business attribute information and the number of currently completed steps in the business behavior data, the business completion characteristic is determined. Then, based on the time when the user started executing the target business and the current time in the business data, the total processing time of the target business is determined, thereby determining the total time consumption characteristic.

[0092] Business processing time periods and user sources can be obtained from business behavior data, and the device type and operating system used can be obtained from attribute information. Then, by combining the business processing time period, user source, device type and operating system, business environment characteristics can be generated.

[0093] After determining the static behavior characteristics, dynamic behavior characteristics, and business environment characteristics, these characteristics can be normalized to obtain the business behavior characteristic V_user.

[0094] In this embodiment of the application, by determining static behavior characteristics, dynamic behavior characteristics, and business environment characteristics based on attribute information and the business behavior data, the comprehensiveness of the characteristic determination can be guaranteed.

[0095] Based on the above embodiments, in this application embodiment, the dynamic behavior characteristics include processing repetition characteristics, page dwell characteristics, operation smoothness characteristics, and operation backtracking characteristics. Furthermore, an optional method for determining dynamic behavior characteristics is provided, such as... Figure 3 As shown, the specific steps include:

[0096] S301, determine the repetitive processing characteristic based on the number of times the same control is processed repeatedly in the business behavior data.

[0097] The number of repeated processing operations refers to the number of times a user performs the same operation on the same control. The repeated processing characteristic is used to characterize the degree of hesitation in the user's processing.

[0098] Optionally, the number of loops in each interactive control can be obtained from the user behavior sequence corresponding to the business behavior data. For example, the number of loops of "input-delete-re-input" in an input box can be counted. By analyzing user behavior data that abandons the target business in historical time periods, a threshold for the number of loops can be determined (using the quantile method, in this scenario, the 95th percentile is used).

[0099] When the number of iterations is greater than or equal to the threshold, the repetitive processing characteristic is determined to be obvious. When the number of iterations is less than the threshold, the repetitive processing characteristic is determined to be insignificant.

[0100] S302, determine page dwell characteristics based on the page dwell time of each page in the business behavior data.

[0101] Among these, page dwell time refers to the amount of time a user spends on the displayed page. Page dwell characteristics are used to characterize the degree to which a user remains on the page.

[0102] Optionally, the dwell time corresponding to the interactive page (e.g., non-information reading page) that requires interactive operation can be obtained from business behavior data. Then, it can be determined whether the dwell time exceeds the preset average dwell time threshold of the page (e.g., the historical data mean plus two standard deviations), and the degree of exceeding it can be quantified to obtain the page dwell characteristics.

[0103] In addition, for key decision-making pages (such as product selection pages and contract confirmation pages), the deviation of the dwell time from the average dwell time of the page (Z-score standardization) can be used to determine whether there are abnormal dwell times on the key pages.

[0104] S303, determine the smoothness characteristics of operation based on the operation interval duration between consecutive interactive operations in the business behavior data.

[0105] Among them, continuous interactive operation refers to two interactive operations executed consecutively. The so-called operation fluency feature is used to characterize the smoothness of user operation.

[0106] Optionally, the variance of the time interval sequence between two consecutive interactive operations (such as switching between form fields or clicking "next") can be calculated based on business behavior data, and the smoothness characteristics of the operation can be determined based on the variance. A larger variance indicates a less smooth user operation.

[0107] It is worth noting that abnormal "long interval - short-term intensive operation" patterns can be identified based on the smoothness of operation. For example, the sudden and rapid completion of multiple fields after a long period of stillness, as well as identifiable abnormal pop-up exposure information, may indicate that the user has been interfered with by external factors or that the operation was not performed by the user, and an abnormal prompt should be given.

[0108] S304. Determine the operation backtracking characteristics based on the backtracking operation data in the business behavior data.

[0109] In this context, backtracking refers to returning to and executing a previous operation. Operation backtracking characteristics are used to characterize the degree of hesitation a user exhibits during an operation. That is, high-frequency, deep backtracking usually indicates user confusion or uncertainty about previously entered information.

[0110] Optionally, based on business behavior data, the frequency (number of times / minute) and depth (maximum number of backtracking steps) of users returning from subsequent steps to previous steps can be determined, thereby identifying the operation backtracking characteristics.

[0111] In this embodiment of the application, by determining the processing repetition feature, page dwell feature, operation smoothness feature and operation backtracking feature based on business behavior data, the comprehensiveness of the determination of dynamic behavior features can be guaranteed.

[0112] Based on the above embodiments, this application provides an optional method for determining breakpoint detection results, such as... Figure 4 As shown, the specific steps include:

[0113] S401 uses a probabilistic prediction model to process business behavior characteristics and obtain breakpoint probabilities.

[0114] The probabilistic prediction model is used to predict the probability of a user's breakpoint. It can be a trained Gradient Boosting Decision Tree (GBDT) model or a Long Short-Term Memory (LSTM) network. For example, a large-scale historical behavior data containing labels such as "completed business processing" or "abandoned processing" can be used to supervise the initial GBDT and LSTM to obtain the probabilistic prediction model. The breakpoint probability P_break is the probability that the user will perform a breakpoint operation.

[0115] Optionally, business behavior features can be input into a trained probabilistic prediction model, which will then output breakpoint probabilities based on the business behavior features and model parameters.

[0116] For example, static behavioral features and business environment features can be input into a trained GBDT, and dynamic behavioral features can be input into a trained LSTM. The GBDT and LSTM will then process the data together to output the breakpoint probability.

[0117] S402, if it is determined that a user has a breakpoint operation based on the breakpoint probability, determine the breakpoint type based on the business behavior characteristics and the breakpoint probability.

[0118] The breakpoint type refers to the type of exception that occurs when a user performs a breakpoint operation, which may include, but is not limited to, hesitant breakpoints, confused breakpoints, and abandoned breakpoints.

[0119] Optionally, if it is determined that a user has a breakpoint operation based on the breakpoint probability, the Shapley Additive Explanations (SHAP) analysis method can be used to analyze the business behavior feature V_user, obtain the contribution degree of each behavior component in the business behavior feature V_user, and then determine the breakpoint type based on the feature with the greater contribution.

[0120] For example, the characteristics of hesitant breakpoints are mainly high hesitation (repeated editing, comparison of optional elements) and high dwell time (long periods of time spent on the decision-making page). The characteristics of confused breakpoints are mainly frequent backtracking behavior, low operational fluency (large variance in operation intervals), and abnormal page exposure. The characteristics of abandoned breakpoints are mainly long periods of inactivity (abnormally high dwell time, while other dynamic characteristics are zero), ultimately leading to the direct closure of the page / application. Corresponding marketing strategies involve attempting to re-engage users upon their next active session through push notifications and SMS messages, using more attractive gimmicks (such as "Your exclusive quota has been approved").

[0121] S403, determine the breakpoint detection result based on the breakpoint probability and breakpoint type.

[0122] Optionally, breakpoint detection results can be obtained by analyzing breakpoint probabilities and breakpoint types. For example, breakpoint probabilities and breakpoint types can be mapped to a detection template to generate breakpoint detection results.

[0123] In this embodiment of the application, a probability prediction model is used to process the business behavior characteristics to obtain the breakpoint probability. Based on the business behavior characteristics and the breakpoint probability, the breakpoint type is determined, thereby obtaining the breakpoint detection result, which can ensure the accuracy of the breakpoint detection result determination.

[0124] Based on the above embodiments, this application provides an optional method for determining breakpoint operations. Specifically, a probability threshold is determined based on the business step to which the user's current interaction with the target business belongs in the target business; if the breakpoint probability is greater than the probability threshold, it is determined that the user has a breakpoint operation.

[0125] The so-called current interaction behavior refers to the user's interaction behavior with the target business at the current moment. The so-called probability threshold is a value used to measure whether a breakpoint operation exists.

[0126] Optionally, the business steps within the target business can be used as an index to query among candidate business steps and candidate probability thresholds to obtain the probability threshold corresponding to the current business step. The later the business step, the higher its importance and the higher its probability threshold. Furthermore, the probability threshold can be optimized based on the types of breakpoints that may exist.

[0127] Furthermore, the breakpoint probability can be compared with a probability threshold. If the breakpoint probability is greater than the probability threshold, it is determined that the user has a breakpoint operation; if the breakpoint probability is less than or equal to the probability threshold, it is determined that the user has no breakpoint operation.

[0128] In this embodiment of the application, by determining the probability threshold based on the business steps of the current interaction behavior in the target business, it is possible to determine whether there is a breakpoint operation, thus ensuring the rationality and flexibility of breakpoint operation identification.

[0129] Based on the above embodiments, this application provides an optional method for determining breakpoint handling strategies, such as... Figure 5 As shown, the specific steps include:

[0130] S501, based on the breakpoint detection results, the user's user identification information, and the user's current interaction behavior with the target service, generate a target matching vector.

[0131] The user identification information is used to identify user types. The target matching vector is used to match breakpoint handling strategies.

[0132] Optionally, the current interaction behavior, breakpoint detection results, and user identification information can be concatenated to obtain a target matching vector. For example, the current interaction behavior, breakpoint detection results, and user identification information can be mapped to a preset vector template to obtain the target matching vector.

[0133] For example, by concatenating the breakpoint type (hesitant, confused, abandoned, etc.), breakpoint probability P_break, breakpoint threshold T, the current business process step corresponding to the current interaction behavior (such as "amount input page" or "identity information filling page") and the execution action (which defines the specific marketing or intervention instructions that the system should trigger when the conditions are met), and user identification information (such as "high-value potential customer" or "new user"), the target matching vector can be obtained.

[0134] S502: Using the target matching vector as an index, query the strategy library to obtain the breakpoint handling strategy associated with the target matching vector.

[0135] The so-called strategy library contains breakpoint handling strategies associated with each candidate matching vector. For example, breakpoint handling strategies can be pre-configured for various types of candidate matching vectors, thereby generating the strategy library. Candidate matching vectors are simply the various possible matching vectors.

[0136] Optionally, the target matching vector can be used as an index. An efficient rule matching algorithm matches each candidate matching vector in the strategy library to obtain the candidate matching vector with the highest matching degree. The breakpoint handling strategy associated with this candidate matching vector is then used as the breakpoint handling strategy associated with the target matching vector.

[0137] For example, for hesitant breakpoints, processing strategies such as pushing "discounted resource access" or "product activity push" can be configured to facilitate decision-making; for confused breakpoints, processing strategies such as triggering intelligent customer service to intervene proactively or highlighting "Frequently Asked Questions" and "Operation Guide" can be configured; for abandoned breakpoints, processing strategies such as pushing through the interface or sending SMS can be configured when the user becomes active again.

[0138] It's worth noting that after handling user breakpoint actions, the processing results can be tracked. Specifically, a processing ID can be generated for each breakpoint handling strategy, and all its context information can be persistently recorded, including: the triggering rule, the reach channel, the user's Session_ID, and the processing content. Then, the user's subsequent behavior after receiving the processing is continuously tracked, and the conversion result is recorded. This result is quantified into a binary label, for example: conversion success = 1 (the user completed the business processing in the current session or within a specified subsequent time); conversion failure = 0 (the user did not complete the business processing within the time threshold).

[0139] Furthermore, processing records and conversion results over a period of time can be periodically (e.g., weekly or monthly) compiled into a strategy effectiveness dataset. Based on this dataset, comparative testing and analysis can be conducted, for example, comparing the conversion effects of different rules, different copywriting, and different channels on the same type of breakpoint.

[0140] Furthermore, the policy library can be optimized; for example, inefficient policies can be removed and new efficient policies can be enabled or created based on data reports. Policy performance datasets can also be used as training data to automatically optimize conditional thresholds and action selections in the policy library using machine learning algorithms (such as reinforcement learning).

[0141] In this embodiment of the application, by using the target matching vector as an index to query the strategy library, the breakpoint handling strategy associated with the target matching vector is obtained, which can ensure the accuracy of the breakpoint handling strategy determination.

[0142] Figure 6 This is a flowchart illustrating a method for handling user breakpoint operations in another embodiment. Based on the above embodiments, this embodiment provides an optional example of a method for handling user breakpoint operations. (Combined with...) Figure 6 The specific implementation process is as follows:

[0143] S601, in response to a user's processing request for a target business, obtains attribute information and user business behavior data for the target business.

[0144] The attribute information includes user attribute information and target business attribute information.

[0145] S602, based on business behavior data, determine dynamic behavior characteristics, and based on attribute information and business behavior data, determine static behavior characteristics and business environment characteristics.

[0146] Among them, dynamic behavioral characteristics include processing repetitive characteristics, page dwell characteristics, operation smoothness characteristics, and operation backtracking characteristics.

[0147] Optionally, the following features can be determined: processing repetition feature based on the number of times the same control is processed repeatedly in the business behavior data; page dwell time feature based on the page dwell time of each page in the business behavior data; operation smoothness feature based on the operation interval between consecutive interactive operations in the business behavior data; and operation backtracking feature based on the backtracking operation data in the business behavior data.

[0148] S603 uses a probabilistic prediction model to process dynamic behavior characteristics, static behavior characteristics, and business environment characteristics to obtain breakpoint probabilities.

[0149] S604: If it is determined that a user has a breakpoint operation based on the breakpoint probability, the breakpoint type is determined based on the business behavior characteristics and the breakpoint probability, and the breakpoint detection result is determined based on the breakpoint probability and the breakpoint type.

[0150] S605: If the breakpoint detection result indicates that the user has a breakpoint operation, generate a target matching vector based on the breakpoint detection result, the user's user identification information, and the user's current interaction behavior with the target business. Then, use the target matching vector as an index to query the policy library to obtain the breakpoint handling policy associated with the target matching vector.

[0151] The strategy library contains breakpoint handling strategies associated with each candidate matching vector.

[0152] Optionally, a probability threshold is determined based on the business step to which the user's current interaction with the target business belongs within the target business; if the breakpoint probability is greater than the probability threshold, it is determined that the user has a breakpoint operation.

[0153] S606 employs a breakpoint handling strategy to process user breakpoint operations.

[0154] The specific processes of S601-S606 described above can be found in the description of the above method embodiments. Their implementation principles and technical effects are similar, and will not be repeated here.

[0155] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0156] Based on the same inventive concept, this application also provides a processing apparatus for user breakpoint operations to implement the above-described processing method for user breakpoint operations. The solution provided by this apparatus is similar to the implementation described in the above-described method. Therefore, the specific limitations of one or more processing apparatus embodiments for user breakpoint operations provided below can be found in the limitations of the processing method for user breakpoint operations described above, and will not be repeated here.

[0157] In one exemplary embodiment, such as Figure 7 As shown, a processing device 1 for user breakpoint operations is provided, including: a data acquisition module 10, a feature determination module 20, a result determination module 30, and a breakpoint processing module 40, wherein:

[0158] The data acquisition module 10 is used to respond to the user's processing request for the target business and acquire attribute information and user business behavior data for the target business; wherein, the attribute information includes user attribute information and business attribute information of the target business;

[0159] Feature determination module 20 is used to determine business behavior features based on attribute information and business behavior data;

[0160] The result determination module 30 is used to determine the breakpoint detection result based on the characteristics of business behavior.

[0161] The breakpoint handling module 40 is used to determine the breakpoint handling strategy based on the breakpoint detection result when the breakpoint detection result indicates that the user has a breakpoint operation, and to process the user's breakpoint operation using the breakpoint handling strategy.

[0162] In an exemplary embodiment, business behavior characteristics include static behavior characteristics, dynamic behavior characteristics, and business environment characteristics; the characteristic determination module 20 is specifically used for:

[0163] Based on business behavior data, determine dynamic behavior characteristics; based on attribute information and business behavior data, determine static behavior characteristics and business environment characteristics.

[0164] In one exemplary embodiment, the dynamic behavior features include processing repetition features, page dwell features, operation smoothness features, and operation backtracking features; the feature determination module 20 is further configured to:

[0165] Based on the number of times the same control is repeatedly processed in the business behavior data, the characteristics of repeated processing are determined; based on the page dwell time of each page in the business behavior data, the characteristics of page dwell time are determined; based on the operation interval between consecutive interactive operations in the business behavior data, the characteristics of operation smoothness are determined; based on the backtracking operation data in the business behavior data, the characteristics of operation backtracking are determined.

[0166] In one exemplary embodiment, the result determination module 30 is specifically used for:

[0167] A probabilistic prediction model is used to process business behavior characteristics to obtain breakpoint probabilities. If a breakpoint operation is determined based on the breakpoint probabilities, the breakpoint type is determined based on the business behavior characteristics and breakpoint probabilities. Finally, the breakpoint detection result is determined based on the breakpoint probabilities and breakpoint types.

[0168] In one exemplary embodiment, the breakpoint handling module 40 is specifically used for:

[0169] Based on the current interaction behavior between the user and the target business, determine the business step to which the user belongs in the target business; if the probability of a breakpoint is greater than the probability threshold, determine that the user has a breakpoint operation.

[0170] In one exemplary embodiment, the breakpoint handling module 40 is further configured to:

[0171] Based on the breakpoint detection results, the user's user identification information, and the user's current interaction behavior with the target service, a target matching vector is generated. The target matching vector is used as an index to query the strategy library to obtain the breakpoint handling strategy associated with the target matching vector. The strategy library contains the breakpoint handling strategies associated with each candidate matching vector.

[0172] The modules in the aforementioned processing device for user breakpoint operations can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0173] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores user behavior data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for handling user breakpoint operations.

[0174] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0175] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0176] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0177] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0178] It should be noted that the user information (including but not limited to user attribute information) and data (including but not limited to user behavior data) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0179] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0180] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0181] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for handling user breakpoint operations, characterized in that, The method includes: In response to a user's processing request for a target service, attribute information and the user's business behavior data for the target service are obtained; wherein, the attribute information includes user attribute information and business attribute information of the target service; Based on the attribute information and the business behavior data, the business behavior characteristics are determined; Based on the aforementioned business behavior characteristics, determine the breakpoint detection results; If the breakpoint detection result indicates that the user has a breakpoint operation, a breakpoint handling strategy is determined based on the breakpoint detection result, and the breakpoint handling strategy is used to handle the user's breakpoint operation.

2. The method according to claim 1, characterized in that, The business behavior characteristics include static behavior characteristics, dynamic behavior characteristics, and business environment characteristics; The step of determining business behavior characteristics based on the attribute information and the business behavior data includes: Based on the aforementioned business behavior data, dynamic behavioral characteristics are determined; Based on the attribute information and the business behavior data, static behavior characteristics and business environment characteristics are determined.

3. The method according to claim 2, characterized in that, The dynamic behavior features include processing repetition features, page dwell features, operation smoothness features, and operation backtracking features; The step of determining dynamic behavior characteristics based on the business behavior data includes: Based on the number of times the same control is repeatedly processed in the business behavior data, the repetitive processing characteristics are determined; Based on the page dwell time of each page in the business behavior data, determine the page dwell characteristics; Based on the duration of the operation interval between consecutive interactive operations in the business behavior data, the operation smoothness characteristics are determined; Based on the backtracking operation data in the business behavior data, determine the operation backtracking characteristics.

4. The method according to claim 1, characterized in that, The step of determining the breakpoint detection result based on the business behavior characteristics includes: A probabilistic prediction model is used to process the business behavior characteristics to obtain the breakpoint probability; If it is determined that the user has a breakpoint operation based on the breakpoint probability, the breakpoint type is determined based on the business behavior characteristics and the breakpoint probability. The breakpoint detection result is determined based on the breakpoint probability and the breakpoint type.

5. The method according to claim 4, characterized in that, The step of determining that the user has a breakpoint operation based on the breakpoint probability includes: A probability threshold is determined based on the business step to which the user's current interaction with the target service belongs within the target service; If the breakpoint probability is greater than the probability threshold, it is determined that the user has a breakpoint operation.

6. The method according to claim 1, characterized in that, The step of determining the breakpoint handling strategy based on the breakpoint detection results includes: Based on the breakpoint detection results, the user's user identification information, and the user's current interaction behavior with the target service, a target matching vector is generated; Using the target matching vector as an index, a query is performed in the strategy library to obtain the breakpoint handling strategy associated with the target matching vector; wherein, the strategy library contains breakpoint handling strategies associated with each candidate matching vector.

7. A processing device for user breakpoint operations, characterized in that, The device includes: The data acquisition module is used to respond to a user's processing request for a target service and acquire attribute information and the user's business behavior data for the target service; wherein, the attribute information includes user attribute information and business attribute information of the target service; The feature determination module is used to determine business behavior features based on the attribute information and the business behavior data; The result determination module is used to determine the breakpoint detection result based on the business behavior characteristics. The breakpoint handling module is used to determine a breakpoint handling strategy based on the breakpoint detection result when the breakpoint detection result indicates that the user has a breakpoint operation, and to process the user's breakpoint operation using the breakpoint handling strategy.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images