Cross-domain data anomaly positioning method and device, electronic equipment and storage medium

Abstract

The invention relates to a cross-domain data anomaly positioning method and device, electronic equipment and a storage medium. The method comprises the steps of obtaining cross-domain data generated by a user across business domains; determining a plurality of longitudinal service fields and a core node in each service field according to the cross-domain data, and calculating a node conversion index of each core node; a plurality of service scenes are divided in the transverse direction according to the behavior path of the user, a scene conversion index of each service scene is determined, each service scene corresponds to an independent user conversion path, and each user conversion path comprises a plurality of core nodes; and under the condition that the total conversion index of the cross-domain data abnormally fluctuates, abnormal nodes in the abnormal path are positioned by monitoring each scene conversion index and each node conversion index, and the total conversion index is composed of the sum value of each scene conversion index. According to the invention, the abnormity positioning efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a method, device, electronic device and storage medium for locating anomalies in cross-domain data. Background Art

[0002] With the diversification of corporate business, the business model of multi-scenario mutual diversion is gradually increasing. The business model of multi-scenario mutual diversion refers to an enterprise having multiple business scenarios, and these scenarios use specific strategies and means to achieve mutual conversion or transfer of user traffic. This model is usually used to improve user engagement, increase user stickiness and improve the conversion rate of the overall business. For example, under the guidance of a doctor, a patient uses a specific health application to record his or her condition, and jumps to the insurance platform through the application for reimbursement.

[0003] As the demand for cross-domain data analysis increases, related technologies mainly analyze the cross-domain conversion rate when jumping between multiple scenarios. If abnormal fluctuations occur during cross-domain conversions between multiple scenarios, it is impossible to clearly locate the specific abnormal nodes, resulting in low efficiency in abnormal troubleshooting. Summary of the invention

[0004] The present application provides a method, device, electronic device and storage medium for locating anomalies in cross-domain data to solve the problem of low efficiency in anomaly detection.

[0005] In a first aspect, the present application provides a method for locating anomalies in cross-domain data, the method comprising:

[0006] Acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by the user's behavior actions in each business domain;

[0007] Determine multiple vertical business fields and core nodes in each business field according to the cross-domain data, and calculate the node conversion index of each core node;

[0008] Divide multiple business scenarios horizontally according to the user's behavior path, and determine the scenario conversion index of each business scenario, wherein each business scenario corresponds to an independent user conversion path, and each user conversion path includes multiple core nodes;

[0009] In the case that the total conversion index of the cross-domain data fluctuates abnormally, the abnormal nodes in the abnormal path are located by monitoring each of the scene conversion indicators and each of the node conversion indicators, wherein the total conversion index is composed of the sum of the conversion indicators of each scene.

[0010] Optionally, determining multiple vertical business domains and core nodes in each business domain according to the cross-domain data, and calculating the node conversion index of each core node includes:

[0011] Determining a plurality of business domains associated with the cross-domain data, wherein the business domains are divided vertically;

[0012] Identify the core nodes in each business area;

[0013] A node conversion index of the current node is determined according to the user visits of the current node and the user visits of the previous adjacent node.

[0014] Optionally, determining a scenario conversion index for each of the business scenarios includes:

[0015] Determine multiple core nodes included in the business scenario;

[0016] The scenario conversion index of the business scenario is determined according to the product of the node conversion index of each core node in the business scenario.

[0017] Optionally, when the total conversion index of the cross-domain data fluctuates abnormally, locating the abnormal node in the abnormal path by monitoring each of the scene conversion indexes and each of the node conversion indexes includes:

[0018] Determining a total conversion index of the cross-domain data according to the starting node and the ending node in the serialized business domain;

[0019] When abnormal fluctuations in the total conversion index are detected, each monitored scene conversion index is compared with the corresponding scene index threshold to locate the abnormal path;

[0020] The conversion index of each node in the monitored abnormal path is compared with the corresponding node index threshold to locate the abnormal node.

[0021] Optionally, determining the total conversion index of the cross-domain data according to the starting node and the ending node in the serialized business domain includes:

[0022] Determine the starting and ending nodes in the serialized business domain;

[0023] Determine the first node data of the end node and the second node data of the start node;

[0024] A total conversion index of the cross-domain data is determined according to a ratio of the first node data to the second node data.

[0025] Optionally, obtaining cross-domain data generated by users across business domains includes:

[0026] Acquire business data generated by the user in each business field, wherein the user carries a first identifier and a second identifier, the first identifier is used to indicate the behavior of the user in a session, and the second identifier is used to indicate the behavior of the user in different sessions and different business fields;

[0027] When a user jumps between core nodes, the user's current business domain is determined based on the domain identifier associated with the current core node;

[0028] When the user switches between business areas, the user's behavior data is connected in series through the first identifier and the second identifier.

[0029] Optionally, after locating the abnormal node in the abnormal path, the method further includes:

[0030] Distinguish the user's behavior paths in different business fields according to the field identifier;

[0031] Perform anomaly analysis based on behavioral paths in each business area.

[0032] In a second aspect, the present application provides a device for locating anomalies in cross-domain data, the device comprising:

[0033] An acquisition module, used to acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by the user's behavior actions in each business domain;

[0034] A first determination module is used to determine multiple vertical business fields and core nodes in each business field according to the cross-domain data, and calculate a node conversion index of each core node;

[0035] A second determination module is used to divide a plurality of business scenarios horizontally according to the user's behavior path, and determine a scenario conversion index for each of the business scenarios, wherein each business scenario corresponds to an independent user conversion path, and each of the user conversion paths includes a plurality of the core nodes;

[0036] A positioning module is used to locate abnormal nodes in an abnormal path by monitoring each of the scene conversion indicators and each of the node conversion indicators when an abnormal fluctuation occurs in the total conversion indicator of the cross-domain data, wherein the total conversion indicator is composed of the sum of the conversion indicators of each scene.

[0037] In a third aspect, the present application provides an electronic device comprising: at least one communication interface; at least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; and at least one memory connected to the at least one bus.

[0038] In a fourth aspect, the present application further provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the cross-domain data anomaly locating method described in any one of the above items of the present application.

[0039] The above technical solution provided by the embodiment of the present application has the following advantages compared with the prior art: cross-domain data is analyzed and processed, and is divided into multiple business scenarios horizontally and business fields and core nodes vertically. Since the sum of the conversion indicators of multiple scenarios is the total conversion indicator, when the total conversion indicator fluctuates abnormally, the abnormal business scenario can be quickly located by monitoring the conversion indicators of each business scenario and core node, thereby further locating the abnormal core node. The present application realizes the positioning of specific abnormal nodes and can improve the efficiency of abnormal troubleshooting. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0042] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0043] Figure 1 A flow chart of a method for locating anomalies in cross-domain data provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the process flow of the A platform and B service provided in the embodiment of the present application;

[0045] Figure 3 A schematic diagram of the structure of a device for locating anomalies in cross-domain data provided in an embodiment of the present application;

[0046] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0048] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0049] In order to solve the problems mentioned in the background technology, according to one aspect of the embodiments of the present application, an embodiment of anomaly locating of cross-domain data is provided.

[0050] The following will be combined with a specific implementation method to provide a cross-domain data anomaly location method provided by the present application embodiment in detail, which is applied to a server, such as Figure 1 As shown, the specific steps are as follows:

[0051] Step 101: Acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by user behavior actions in each business domain;

[0052] Step 102: Determine multiple vertical business fields and core nodes in each business field based on the cross-domain data, and calculate the node conversion index of each core node;

[0053] Step 103: Divide multiple business scenarios horizontally according to the user's behavior path, and determine the scenario conversion index of each business scenario, wherein each business scenario corresponds to an independent user conversion path, and each user conversion path includes multiple core nodes;

[0054] Step 104: When the total conversion index of cross-domain data fluctuates abnormally, locate the abnormal node in the abnormal path by monitoring each scene conversion index and each node conversion index, wherein the total conversion index is composed of the sum of the conversion indexes of each scene.

[0055] The present application analyzes cross-domain data in cross-domain scenarios, vertically dividing it into multiple business fields and core nodes in each business field, and horizontally dividing it into multiple user behavior paths (business scenarios). Each user behavior path includes multiple core nodes, so as to realize data division according to business scenarios. In this way, when an abnormality occurs in the total conversion indicator, it can be determined which user behavior path has the abnormality, thereby further locating the abnormal node in the abnormal path.

[0056] This application is applied to cross-domain scenarios, including but not limited to: e-commerce and financial services, health care and insurance services, media and advertising, real estate and home decoration, etc. For example, patients use a specific health application under the guidance of a doctor to record their medical conditions, and through the application jump to the insurance platform for reimbursement processing. For another example, home buyers view house information on a real estate website, and then jump to the service platform to book moving services or jump to the decoration platform for home decoration design consultation according to the link in the website.

[0057] The system collects user behavior data in each business area through log recording, user behavior tracking and other means. These data include user access time, click behavior, conversion behavior, etc., and integrates data from different business areas into a unified data platform to ensure data consistency and analyzability. For example, in area A: record the user's access time on the business details page, the time when the activity page is clicked, and the payment time on the payment page; in area B: record the time when the user jumps from the payment page to the event landing page, the time when the event is confirmed and clicked, and the time when the number of items consumed, etc.

[0058] The system divides multiple business areas vertically based on cross-domain data, determines the core nodes in each business area, and then calculates the node conversion index of each core node. The core nodes are different for different business areas. The node conversion index includes but is not limited to: conversion rate, transaction rate, exit rate, churn rate, retention rate, etc. The node conversion index indicates the various behavioral performances of users in different business scenarios and paths.

[0059] For example, Figure 2 The process diagram for serving A platform and B shows that the two business areas are vertically divided into A area and B area. The core nodes in A area include but are not limited to: business details page, activity page, payment page, payment intermediate page, echo identity information, order display and order success. The core nodes in B area include but are not limited to: event landing page, additional benefits page, event confirmation click, and number of consumption items. Node conversion indicators include: business details page uv (UniqueVisitor, number of independent visitors), activity page uv, payment page uv, payment intermediate page uv, event landing page uv, additional benefits page uv, event confirmation click uv, etc.

[0060] The system defines multiple independent business scenarios based on the user's behavior path. Each business scenario represents an independent user conversion path. The business scenario includes multiple core nodes, and the scenario conversion index can be calculated based on the multiple core nodes in the business scenario.

[0061] For example, Figure 2 Two scenes can be included, and each scene and the core nodes in each scene are as follows.

[0062] Scenario 1: The user directly enters the activity page from the business details page, and after successful payment, jumps directly to the event landing page across domains, and makes purchases on the event landing page;

[0063] Core nodes: business details page uv → activity page uv → payment page uv → matter landing page uv → additional benefits page uv → matter confirmation click uv → number of consumption items.

[0064] Scenario 2: The user enters the payment intermediate page from the business details page with the identity information path. After the payment is successful, the user automatically jumps to the event landing page and performs the consumption behavior;

[0065] Core nodes: business details page uv→payment middle page uv→echo identity information uv→order display uv→order success uv→item landing page uv→additional benefits page uv→item confirmation click uv→number of consumption items.

[0066] The system monitors the changes in the total conversion index in real time. When the total conversion index fluctuates abnormally, the abnormal troubleshooting process is initiated; the conversion rates of various business scenarios are compared to find the scenarios where the conversion rate drops; for the scenarios where the conversion rate drops, the conversion rate of each core node is further analyzed to find the core nodes where the conversion rate drops, thereby determining the specific problem nodes and analyzing the causes of the problem nodes. When the total conversion index fluctuates abnormally, this application quickly locates the problem points by monitoring the conversion indicators of each business scenario and core node.

[0067] For example, the total conversion rate = the conversion rate of scenario one + the conversion rate of scenario two. In each scenario, there is a separate user conversion funnel, which together constitute the head and tail data values ​​of the total conversion rate. When the total conversion rate drops from 10% to 8%, it is determined that the conversion rate of scenario two drops from 12% to 8%. Further analysis of the core nodes of scenario two shows that the conversion rate of "payment intermediate page to echo identity information" drops from 90% to 70%. By comparing historical data with current data, it is found that the "payment intermediate page" has a slow page loading speed and poor user experience, leading to user loss.

[0068] In this application, the indicator types of core nodes, business scenarios and total conversion indicators are consistent, including but not limited to: conversion rate, transaction rate, exit rate, churn rate, retention rate, etc.

[0069] This application analyzes and processes cross-domain data, divides it into multiple business scenarios horizontally, and divides it into business areas and core nodes vertically. Since the sum of conversion indicators of multiple scenarios is the total conversion indicator, when the total conversion indicator fluctuates abnormally, by monitoring the conversion indicators of each business scenario and core node, the abnormal business scenario can be quickly located, thereby further locating the abnormal core node. This application realizes the location of specific abnormal nodes and can improve the efficiency of abnormal investigation.

[0070] As an optional implementation, determining multiple vertical business fields and core nodes in each business field based on cross-domain data, and calculating the node conversion index of each core node includes: determining multiple business fields associated with cross-domain data, wherein the business fields are divided vertically; determining the core nodes in each business field; and determining the node conversion index of the current node based on the user visits to the current node and the user visits to the previous adjacent node.

[0071] The system divides the business areas vertically according to the logical order of user behavior, and integrates data from different business areas into a unified data platform to ensure data consistency and analyzability. In each business area, the core nodes of user behavior are determined to ensure that the data of each node can be accurately recorded and analyzed, such as click behavior, form submission, page visit, etc. The user visits of the current node and the user visits of the previous adjacent node are counted, and the calculation formula of the node conversion index is:

[0072] Node conversion rate = (user visits to the current node / user visits to the previous adjacent node)*100%.

[0073] For example, from the time a user enters the business details page to the time the user completes the purchase, the entire process is divided into area A and area B. The core nodes in area A include the business details page, activity page, payment page, and payment intermediate page, and the core nodes in area B include the event landing page, additional benefits page, event confirmation click, etc. If the node conversion rate from the business details page to the activity page is required, assuming that the number of users who visit the activity page = 1000 and the number of users who visit the business details page = 2000, then the node conversion rate of the activity page = (1000 / 2000)*100% = 50%.

[0074] As an optional implementation, determining the scenario conversion index of each business scenario includes: determining multiple core nodes included in the business scenario; and determining the scenario conversion index of the business scenario based on the product of the node conversion index of each core node in the business scenario.

[0075] The system clarifies the starting and end points of each business scenario, as well as the main path of the user in the scenario. In each business scenario, the system clarifies the definition and judgment criteria of each node, such as click behavior, form submission, page visit, etc., so as to determine the core nodes of user behavior. By calculating the product of the node conversion indicators of all core nodes in each business scenario, the overall conversion rate of the business scenario is determined.

[0076] against Figure 2 , the calculation formula of the scene conversion index in scene 1 is as follows:

[0077] Conversion rate of scenario one = number of consumption brought by scenario one / business details page uv*100% = conversion rate of business details page uv to activity page uv * conversion rate of activity page uv to payment page uv * conversion rate of payment page uv to event landing page uv * conversion rate of event landing page uv to additional benefits page * conversion rate of additional benefits page uv to event confirmation click uv * conversion rate of event confirmation click uv to number of consumption.

[0078] The calculation formula for the scene conversion index in Scenario 2 is similar to that in Scenario 1 and will not be repeated here.

[0079] This application can more accurately calculate the overall conversion rate of the business scenario by multiplying the node conversion rate of each core node. This method takes into account the conversion effect of each node and avoids the impact of single node data deviation on the overall conversion rate; by calculating the node conversion rate of each core node, it is easier to find which node has a lower conversion rate, so as to quickly locate the problem point.

[0080] As an optional implementation, when the total conversion index of cross-domain data fluctuates abnormally, locating abnormal nodes in the abnormal path by monitoring each scene conversion index and each node conversion index includes: determining the total conversion index of the cross-domain data based on the starting node and the ending node in the serial business field; when abnormal fluctuations in the total conversion index are detected, comparing the monitored conversion index of each scene with the corresponding scene index threshold to locate the abnormal path; comparing the monitored conversion index of each node in the abnormal path with the corresponding node index threshold to locate the abnormal node.

[0081] The system determines the starting node and the ending node in the serial business domain, determines the first node data of the ending node and the second node data of the starting node; according to the ratio of the first node data to the second node data, determines the total conversion index of the cross-domain data. Then monitor the changes in the total conversion index in real time. When the total conversion rate fluctuates abnormally, start the abnormal investigation process. Calculate the conversion rate of each business scenario, and set a reasonable conversion rate threshold for each business scenario. Compare the conversion rate of each scenario with the corresponding scenario indicator threshold to find out the scenarios with conversion rates below the threshold. In the abnormal path, determine the core nodes of user behavior, set a reasonable conversion rate threshold for each core node, compare the conversion rate of each node with the corresponding node indicator threshold, and find out the nodes with conversion rates below the threshold.

[0082] For example, in platform A and service B, the node data of the starting node is the business details page uv, and the node data of the ending node is the number of consumption items. Then the total conversion index = number of consumption items / business details page uv*100%. The higher the conversion rate, the higher the value of a single visit uv, and vice versa. If the total conversion rate is monitored to drop from 10% to 8%, the scene conversion rate of scenario one is 12%, the scene conversion rate of scenario two is 8%, the scene index threshold is 10%, and the conversion rate of scenario two is 8% lower than the threshold of 10%, and scenario two is located as an abnormal path. The conversion rate of core node 1 in scenario two is 90%, the conversion rate of core node 1 is 80%, and the conversion rate threshold is 85%, so node 2 is located as an abnormal node.

[0083] This application can more accurately locate abnormal paths and abnormal nodes through multi-level monitoring of total conversion indicators, scene conversion indicators and node conversion indicators; by real-time monitoring of total conversion indicators, the abnormality troubleshooting process can be started immediately when a problem occurs, reducing delays in problem discovery.

[0084] As an optional implementation, obtaining cross-domain data generated by users across business fields includes: obtaining business data generated by users in each business field, wherein the user carries a first identifier and a second identifier, the first identifier is used to indicate the user's behavior in a session, and the second identifier is used to indicate the user's behavior in different sessions and different business fields; when the user jumps between core nodes, the business field in which the user is currently located is determined based on the field identifier associated with the current core node; when the user jumps between business fields, the user's behavior data is concatenated through the first identifier and the second identifier.

[0085] 1. Data collection of core nodes.

[0086] Determine the key behavior nodes of users in each business scenario, for example, from the business details page to the activity page, the payment page to the event landing page, etc. Set up data collection points at each core node to record the user's access behavior, operation behavior, and conversion behavior. For example, record the user's access time on the business details page, click behavior on the activity page, behavior on the event landing page, etc., so that data of core nodes can be collected in the user behavior path to ensure that the behavior of each node can be accurately recorded.

[0087] 2. Solve the problem of user cross-domain disconnection by connecting SSID and UnionID in series.

[0088] A unique first identifier ssid (Session ID) and a second identifier unionid (Union ID) are generated for each user, where ssid is used to identify the user's behavior in a session, and unionid is used to identify the user's behavior in different sessions and different domains. When a user jumps from one business area to another, the user's behavior data is linked together through ssid and unionid. For example, when a user jumps from a payment page to an event landing page, the ssid and unionid parameters are carried to ensure the continuity of data between different domains, which can ensure that the user's behavior data between different business areas can be continuously recorded and solve the problem of cross-domain disconnection.

[0089] 3. Differentiate the node problems that users pass through through different scenario values ​​(RC).

[0090] Define a unique scenario value (RC) for each business scenario to identify the business scenario the user is in. For example, the RC value of scenario one is 1, the RC value of scenario two is 2, and so on. At each core node, record the user's current scenario value (RC) to ensure that the data of each node carries the corresponding scenario identifier. For example, when the user jumps from the business details page to the activity page, record the RC value as 1. Use scenario values ​​(RC) to distinguish the user's behavior paths in different business scenarios to ensure the accuracy and completeness of the data. For example, data with an RC value of 1 can be used to analyze the user's behavior path in scenario one. It is possible to distinguish the user's behavior paths in different business scenarios through different scenario values ​​(RC) to ensure the accuracy and completeness of the data.

[0091] For example, assume that the user's conversion path from the business details page to the number of consumption items is as follows:

[0092] 1. Business details page:

[0093] When a user visits the business details page, the access time, SSID, and UnionID are recorded.

[0094] Scenario value (RC): 1.

[0095] 2. Activity page:

[0096] The user clicks to enter the activity page, and the click time, ssid, and unionid are recorded.

[0097] Scenario value (RC): 1.

[0098] 3. Payment page:

[0099] The user completes the payment and records the payment success time and ssid and unionid.

[0100] Scenario value (RC): 1.

[0101] 4. Event landing page:

[0102] The user jumps to the event landing page, and the jump time, ssid, and unionid are recorded.

[0103] Scenario value (RC): 2.

[0104] 5. Additional benefits page:

[0105] The user views additional benefits and records the viewing time, SSID, and UnionID.

[0106] Scenario value (RC): 2.

[0107] 6. Click to confirm:

[0108] User confirmation items, record confirmation time and ssid, unionid.

[0109] Scenario value (RC): 2.

[0110] 7. Number of pieces consumed:

[0111] The user completes the order and records the order success time and ssid and unionid.

[0112] Scenario value (RC): 2.

[0113] Through data collection and splitting of core nodes, combined with the use of SSID and UnionID concatenation and different scenario values ​​(RC), it can ensure that the user's behavior data in different business scenarios and paths can be accurately recorded and analyzed. This process helps solve the problem of cross-domain disconnection, and ensures the continuous recording of user behavior data between different business areas through SSID and UnionID. It can also distinguish user paths and distinguish user behavior paths in different business scenarios through scenario values ​​(RC) to ensure the accuracy and completeness of the data. It can also optimize the user experience. Through detailed data analysis, it can find problem points in user behavior, optimize business processes and improve user experience.

[0114] This application defines a unique scenario value for each business scenario, which is used to identify the business scenario in which the user is located. At each core node, the user's current scenario value (RC) is recorded to ensure that the data of each node carries the corresponding scenario identifier. Through different RC values, the data of different business scenarios are isolated to prevent data confusion. For example, data with an RC value of 1 only reflects the user's behavior path in scenario one, and data with an RC value of 2 only reflects the user's behavior path in scenario two. Through the RC value, the data of each business scenario can be verified to ensure the integrity and accuracy of the data. For example, during the data audit process, the RC value can be used to quickly locate the data of a specific business scenario for detailed review and verification.

[0115] This application provides an overall process of a method for locating anomalies in cross-domain data, including the following steps:

[0116] Step 1. Obtain cross-domain data generated by users across business areas.

[0117] Objective: Collect user behavior data in different business areas and ensure the integrity and accuracy of the data.

[0118] 1) Data collection: Through log recording, user behavior tracking and other means, collect user behavior data in each business area, including visit time, click behavior, conversion behavior, etc.

[0119] 2) Data integration: Integrate data from different business areas into a unified data platform to ensure data consistency and analyzability.

[0120] 3) Data tagging: Mark each business area with a unique identifier (such as RC value) to ensure data traceability.

[0121] Step 2. Identify multiple vertical business areas and core nodes in each business area, and calculate the node conversion index of each core node.

[0122] Objective: Identify the core nodes of each business area in the user behavior path and calculate the conversion indicators of each node.

[0123] 1) Determine the business area: clarify the main paths of users in different business areas and divide them vertically.

[0124] 2) Identify the core nodes: In each business area, identify the core nodes of user behavior.

[0125] 3) Data collection: Set up data collection points at each core node to record users' access behavior, operation behavior and conversion behavior.

[0126] 4) Calculate node conversion index: Calculate the conversion rate of each core node based on the user visits to the current node and the user visits to the previous adjacent node.

[0127] Step 3. Divide multiple business scenarios horizontally according to the user's behavior path, and determine the scenario conversion indicators for each business scenario.

[0128] Objective: Divide the user's behavior path into multiple independent business scenarios and calculate the conversion indicators for each business scenario.

[0129] 1) Define business scenarios: Based on the user's behavior path, define multiple independent business scenarios. Each business scenario represents an independent user conversion path.

[0130] 2) Identify the core nodes: In each business scenario, identify the core nodes of user behavior.

[0131] 3) Calculate the scenario conversion index: Determine the overall conversion rate of the business scenario based on the product of the node conversion index of each core node in the business scenario.

[0132] Step 4. When the total conversion index of cross-domain data fluctuates abnormally, locate the abnormal nodes in the abnormal path by monitoring the conversion index of each scene and the conversion index of each node.

[0133] Objective: When the total conversion index fluctuates abnormally, quickly locate the problem point by monitoring the conversion index of each business scenario and core node.

[0134] 1) Monitor the total conversion index: monitor the changes in the total conversion rate in real time. When the total conversion rate fluctuates abnormally, start the abnormal troubleshooting process.

[0135] 2) Compare scenario conversion indicators: Compare the conversion rate of each business scenario with the corresponding scenario indicator threshold, and find out the scenarios with conversion rates lower than the threshold.

[0136] 3) Locating abnormal paths: After determining the abnormal path, further analyze the conversion rate of each core node in the path to find out the nodes whose conversion rate is lower than the node indicator threshold.

[0137] 4) Problem location: Determine the specific problem nodes through detailed data analysis and analyze the causes of the problem nodes.

[0138] Step 5. Distinguish the user's behavior paths in different business areas based on the domain identifier and perform anomaly analysis.

[0139] Objective: To distinguish the user's behavior paths in different business areas through domain identifiers (such as RC values) and conduct detailed anomaly analysis.

[0140] 1) Define domain identification: Define a unique identifier (such as RC value) for each business domain.

[0141] 2) Record the domain identifier: At each core node, record the user's current domain identifier to ensure that the data of each node carries the corresponding domain identifier.

[0142] 3) Data isolation: Group user behavior data according to domain identifiers to ensure that data in each business area is independent and analyzable.

[0143] 4) Abnormal analysis: In each business area, detailed abnormal analysis is performed based on the user's behavior path to identify problem points and make optimization suggestions.

[0144] Suppose an enterprise conducts user conversion analysis between platform A and service B. Through log recording and user behavior tracking, it collects user behavior data in platform A and service B and integrates them into a unified data platform. In platform A, determine the core nodes from the business details page to the activity page, payment page, etc., and calculate the conversion rate of each node. In service B, determine the core nodes from the event landing page to the additional rights page, event confirmation click, etc., and calculate the conversion rate of each node. The enterprise divides the user's behavior path into multiple business scenarios, such as scenario one and scenario two, and calculates the overall conversion rate of each scenario. When the total conversion rate is monitored to have abnormal fluctuations, by comparing the conversion rate of each scenario, it is found that the conversion rate of scenario two has decreased. Further analysis of the conversion rate of each core node in scenario two shows that the conversion rate from the payment intermediate page to the echo identity information is low. Through detailed data analysis, the problem points of the node are determined, and optimization suggestions are put forward, such as optimizing page design, increasing loading speed, etc., so as to improve the overall conversion rate.

[0145] This application implements a multi-scenario user cross-domain conversion analysis method to achieve cross-domain monitoring and analysis of the user's entire path, and comprehensively consider the utilization and value assessment of traffic resources. This application integrates the company's internal business data to complete the behavior analysis of users in different business scenarios from a data level, thereby obtaining the user's cross-domain conversion rate and behavior trend. By dividing and monitoring the core nodes of each business scenario in detail, the system can quickly locate abnormal fluctuations in the conversion path and improve the efficiency of abnormal investigation. In addition, this method can also help companies optimize user paths, improve overall conversion rates, and ensure that traffic resources in each business area are fully utilized.

[0146] This application provides a device for locating anomalies in cross-domain data, such as Figure 3 As shown, the device comprises:

[0147] An acquisition module 301 is used to acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by the user's behavior actions in each business domain;

[0148] A first determination module 302 is used to determine multiple vertical business domains and core nodes in each business domain according to the cross-domain data, and calculate a node conversion index of each core node;

[0149] The second determination module 303 is used to divide multiple business scenarios horizontally according to the user's behavior path, and determine the scenario conversion index of each business scenario, wherein each business scenario corresponds to an independent user conversion path, and each user conversion path includes multiple core nodes;

[0150] The positioning module 304 is used to locate abnormal nodes in the abnormal path by monitoring each scene conversion index and each node conversion index when the total conversion index of the cross-domain data fluctuates abnormally, wherein the total conversion index is composed of the sum of the conversion indexes of each scene.

[0151] Optionally, the second determining module 303 is used to:

[0152] Identify multiple business areas associated with cross-domain data, where the business areas are divided vertically;

[0153] Identify the core nodes in each business area;

[0154] The node conversion index of the current node is determined according to the user visits of the current node and the user visits of the previous adjacent node.

[0155] Optionally, the second determining module 303 is used to:

[0156] Determine the multiple core nodes included in the business scenario;

[0157] The scenario conversion index of the business scenario is determined based on the product of the node conversion index of each core node in the business scenario.

[0158] Optionally, the positioning module 304 is used to:

[0159] Determine the total conversion index of cross-domain data based on the starting and ending nodes in the serialized business domains;

[0160] When abnormal fluctuations in the total conversion index are detected, the conversion index of each monitored scene is compared with the corresponding scene index threshold to locate the abnormal path;

[0161] Compare the conversion index of each node in the monitored abnormal path with the corresponding node index threshold to locate the abnormal node.

[0162] Optionally, the positioning module 304 is used to:

[0163] Determine the starting and ending nodes in the serialized business domain;

[0164] Determine the first node data of the end node and the second node data of the start node;

[0165] The total conversion index of the cross-domain data is determined according to the ratio of the first node data to the second node data.

[0166] Optionally, the acquisition module 301 is used to:

[0167] Acquire business data generated by the user in each business field, wherein the user carries a first identifier and a second identifier, the first identifier is used to indicate the behavior of the user in a session, and the second identifier is used to indicate the behavior of the user in different sessions and different business fields;

[0168] When a user jumps between core nodes, the user's current business domain is determined based on the domain identifier associated with the current core node;

[0169] When the user switches between business areas, the user's behavior data is connected in series through the first identifier and the second identifier.

[0170] Optionally, the device is also used for:

[0171] Based on the domain identifier, distinguish the user's behavior paths in different business areas;

[0172] Perform anomaly analysis based on behavioral paths in each business area.

[0173] like Figure 4 As shown, an embodiment of the present application provides an electronic device, including a processor 401, a communication interface 402, a memory 403 and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404.

[0174] The memory 403 is used to store computer programs.

[0175] In one embodiment of the present application, the processor 401 is used to implement the cross-domain data anomaly locating method provided by any one of the aforementioned method embodiments when executing the program stored in the memory 403.

[0176] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for locating anomalies in cross-domain data provided in any one of the aforementioned method embodiments are implemented.

[0177] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0178] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0179] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0180] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for locating anomalies in cross-domain data, characterized in that: The method comprises: Acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by the user's behavior actions in each business domain; Determine multiple vertical business fields and core nodes in each business field according to the cross-domain data, and calculate the node conversion index of each core node; Divide multiple business scenarios horizontally according to the user's behavior path, and determine the scenario conversion index of each business scenario, wherein each business scenario corresponds to an independent user conversion path, and each user conversion path includes multiple core nodes; In the case that the total conversion index of the cross-domain data fluctuates abnormally, the abnormal nodes in the abnormal path are located by monitoring each of the scene conversion indicators and each of the node conversion indicators, wherein the total conversion index is composed of the sum of the conversion indicators of each scene.

2. The method according to claim 1, characterized in that Determining multiple vertical business fields and core nodes in each business field based on the cross-domain data, and calculating the node conversion index of each core node includes: Determining a plurality of business domains associated with the cross-domain data, wherein the business domains are divided vertically; Identify the core nodes in each business area; A node conversion index of the current node is determined according to the user visits of the current node and the user visits of the previous adjacent node.

3. The method according to claim 2, characterized in that Determine the scenario conversion indicators for each of the business scenarios including: Determine multiple core nodes included in the business scenario; The scenario conversion index of the business scenario is determined according to the product of the node conversion index of each core node in the business scenario.

4. The method according to claim 1, characterized in that: In the case where the total conversion index of the cross-domain data fluctuates abnormally, locating the abnormal node in the abnormal path by monitoring each of the scene conversion indexes and each of the node conversion indexes includes: Determining a total conversion index of the cross-domain data according to the starting node and the ending node in the serialized business domain; When abnormal fluctuations in the total conversion index are detected, each monitored scene conversion index is compared with the corresponding scene index threshold to locate the abnormal path; The conversion index of each node in the monitored abnormal path is compared with the corresponding node index threshold to locate the abnormal node.

5. The method according to claim 4, characterized in that Determining the total conversion index of the cross-domain data based on the starting node and the ending node in the serialized business domain includes: Determine the starting and ending nodes in the serialized business domain; Determine the first node data of the end node and the second node data of the start node; A total conversion index of the cross-domain data is determined according to a ratio of the first node data to the second node data.

6. The method according to claim 1, characterized in that Obtaining cross-domain data generated by users across business areas includes: Acquire business data generated by the user in each business field, wherein the user carries a first identifier and a second identifier, the first identifier is used to indicate the behavior of the user in a session, and the second identifier is used to indicate the behavior of the user in different sessions and different business fields; When a user jumps between core nodes, the user's current business domain is determined based on the domain identifier associated with the current core node; When the user switches between business areas, the user's behavior data is connected in series through the first identifier and the second identifier.

7. The method according to claim 6, characterized in that After locating the abnormal node in the abnormal path, the method further includes: Distinguish the user's behavior paths in different business fields according to the field identifier; Perform anomaly analysis based on behavioral paths in each business area.

8. A device for locating anomalies in cross-domain data, characterized in that: The device comprises: An acquisition module, used to acquire cross-domain data generated by users across business domains, wherein the cross-domain data is used to indicate data generated by the user's behavior actions in each business domain; A first determination module is used to determine multiple vertical business fields and core nodes in each business field according to the cross-domain data, and calculate a node conversion index of each core node; A second determination module is used to divide a plurality of business scenarios horizontally according to the user's behavior path, and determine a scenario conversion index for each of the business scenarios, wherein each business scenario corresponds to an independent user conversion path, and each of the user conversion paths includes a plurality of the core nodes; A positioning module is used to locate abnormal nodes in an abnormal path by monitoring each of the scene conversion indicators and each of the node conversion indicators when an abnormal fluctuation occurs in the total conversion indicator of the cross-domain data, wherein the total conversion indicator is composed of the sum of the conversion indicators of each scene.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-7 when executing a program stored in a memory.

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 method according to any one of claims 1 to 7 is implemented.

Images