Intelligent funnel data processing method and device, equipment and medium

Abstract

The application provides a funnel data processing method, device and equipment and a medium. In the method, in the process of analyzing and processing flow data through a funnel system, when it is determined that intelligent analysis of the funnel system is needed, a plurality of analysis dimensions are obtained according to a pre-configured atomic dimension set and a dimension depth by using a dimension priority principle. Data analysis of the plurality of analysis dimensions is respectively performed in each sub-funnel system in the funnel system to obtain node data and a conversion rate corresponding to each analysis dimension. Each sub-funnel system is a sub-funnel generated according to the funnel system, which is consistent with the funnel system process and has different analysis dimensions. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by each sub-funnel system are different. Finally, the funnel steps and the problem causes that occur are analyzed and determined according to the node data and the conversion rate of each analysis dimension, and the specific link and the reason of data loss are determined.

Description

Technical Field

[0001] This invention relates to the field of computers, and more particularly to a method, apparatus, device, and medium for processing intelligent funnel data. Background Technology

[0002] With the development of computer and internet technologies, various internet platforms need to process and analyze more and more data, and the requirements for traffic data analysis are gradually increasing. The funnel is a process-oriented data analysis method, an important analytical model that scientifically reflects user behavior and user conversion rates at each stage from start to finish. Currently, funnels are widely used in user behavior analysis and application (APP) data analysis, including traffic monitoring and product target conversion, in daily data operations and data analysis.

[0003] In existing technologies, data analysis using funnels typically involves at least the following steps: the funnel system cleans the raw logs of the target business to obtain raw behavioral logs; then, the raw behavioral logs are grouped according to computational dimensions; a funnel model is generated based on the funnel configuration data; for each group of raw log behaviors, the corresponding computational operator is invoked via an interface call to perform calculations, obtaining the funnel calculation results; and finally, the funnel analysis results for the target business are obtained. This approach primarily improves funnel processing efficiency by configuring the funnel steps. When multiple behavioral logs require multiple funnel analyses, different computational operators can be invoked through different interfaces to complete the funnel analysis and obtain the analysis results.

[0004] However, the aforementioned solutions cannot determine whether there are risks in the data within the funnel stage, nor can they pinpoint the specific reasons for data loss. Currently, there is no technical solution for intelligently locating and analyzing the causes of funnel analysis results. Summary of the Invention

[0005] This invention provides a method, apparatus, device, and medium for processing intelligent funnel data, and provides a technical solution for intelligent positioning and cause analysis of funnel analysis results.

[0006] In a first aspect, embodiments of the present invention provide a method for processing intelligent funnel data, including:

[0007] Based on the pre-configured set of atomic dimensions and dimension depth, multiple analysis dimensions are obtained using a dimension-first principle.

[0008] In each sub-funnel system of the funnel system, the multi-dimensional data analysis is performed to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different.

[0009] Based on the node data and conversion rate corresponding to each analysis dimension, identify the problematic funnel step and the cause of the problem.

[0010] In one specific embodiment, the method further includes:

[0011] Obtain node data and / or conversion rate of the funnel system for the input flow data;

[0012] If the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required.

[0013] If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

[0014] In one specific implementation, the step of obtaining multiple analysis dimensions based on a pre-configured set of atomic dimensions and dimension depth using a dimension-first principle includes:

[0015] Based on the preset dimension depth, the dimensions in the preset atomic dimension set are arranged and combined according to the dimension priority principle to generate multiple analysis dimensions, wherein each analysis dimension includes one or more dimensions in the dimension set.

[0016] In one specific embodiment, the method further includes:

[0017] In response to the user's operation, the identified historical problem dimensions are added to the multi-analysis dimensions to obtain new multi-analysis dimensions.

[0018] In one specific implementation, determining the problematic funnel step and the cause of the problem based on the node data and conversion rate corresponding to each analysis dimension includes:

[0019] Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data map is generated, in which problematic node data and conversion rates are identified;

[0020] Based on the data map, the problematic funnel step and the cause of the problem are determined, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

[0021] Secondly, embodiments of the present invention provide a processing device for intelligent funnel data, comprising:

[0022] The first processing module is used to obtain multiple analysis dimensions based on a pre-configured set of atomic dimensions and dimension depth, using a dimension-first principle.

[0023] The second processing module is used to perform the multi-dimensional data analysis in each sub-funnel system of the funnel system to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different.

[0024] The third processing module is used to determine the problematic funnel step and the cause of the problem based on the node data and conversion rate corresponding to each analysis dimension.

[0025] In one specific embodiment, the device further includes: a fourth processing module, configured to:

[0026] Obtain node data and / or conversion rate of the funnel system for the input flow data;

[0027] If the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required.

[0028] If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

[0029] In one specific implementation, the first processing module is specifically used for:

[0030] Based on the preset dimension depth, the dimensions in the preset atomic dimension set are arranged and combined according to the dimension priority principle to generate multiple analysis dimensions, wherein each analysis dimension includes one or more dimensions in the dimension set.

[0031] In one specific embodiment, the first processing module is further configured to:

[0032] In response to the user's operation, the identified historical problem dimensions are added to the multi-analysis dimensions to obtain new multi-analysis dimensions.

[0033] In one specific implementation, the third processing module is specifically used for:

[0034] Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data map is generated, in which problematic node data and conversion rates are identified;

[0035] Based on the data map, the problematic funnel step and the cause of the problem are determined, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

[0036] Thirdly, embodiments of the present invention provide an electronic device, comprising:

[0037] Processor, memory, and interaction interface;

[0038] The memory is used to store the executable instructions of the processor;

[0039] The processor is configured to execute a method for processing smart funnel data as described in any of the first aspects by executing the executable instructions.

[0040] Fourthly, embodiments of the present invention provide a readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method for processing intelligent funnel data as described in any of the first aspects.

[0041] Fifthly, embodiments of the present invention provide a computer program product, including a computer program, which, when executed by a processor, is used to implement a method for processing intelligent funnel data as described in any of the first aspects.

[0042] The present invention provides a method, apparatus, device, and medium for processing intelligent funnel data. In this method, during the analysis and processing of flow data through a funnel system, when it is determined that intelligent analysis of the funnel system is required, multiple analysis dimensions are obtained based on a pre-configured set of atomic dimensions and dimension depth, using a dimension-first principle. Data analysis of multiple analysis dimensions is performed in each sub-funnel system within the funnel system, obtaining the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system is generated based on the funnel system, sharing the same flow but with different analysis dimensions. Each sub-funnel system performs data analysis on one analysis dimension, and the analysis dimensions performed by each sub-funnel system are different. Finally, based on the node data and conversion rate of each analysis dimension, the problematic funnel steps and causes are analyzed and determined, identifying the specific links and reasons for data loss. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A schematic diagram of the universal funnel data analysis system provided by the present invention;

[0045] Figure 2 A flowchart of an embodiment of the intelligent funnel data processing method provided by the present invention;

[0046] Figure 3 A flowchart of Embodiment 2 of the intelligent funnel data processing method provided by the present invention;

[0047] Figure 4 A flowchart of Embodiment 3 of the intelligent funnel data processing method provided by the present invention;

[0048] Figure 5 A flowchart illustrating an example of intelligent funnel data processing provided by the present invention;

[0049] Figure 6 A schematic diagram of the structure of a first embodiment of the intelligent funnel data processing device provided by the present invention;

[0050] Figure 7 A schematic diagram of the structure of the intelligent funnel data processing device according to Embodiment 2 of the present invention;

[0051] Figure 8 This is a schematic diagram of the structure of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments made by those skilled in the art under the guidance of these embodiments are within the scope of protection of the present invention.

[0053] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0054] As can be seen from the technical solutions provided in the background section, the existing methods for analyzing flow data using funnel systems have at least the following problems:

[0055] (1) It cannot support multi-dimensional funnels;

[0056] (2) It is unable to intelligently determine whether there is a risk in the data of the funnel stage;

[0057] (3) It is impossible to intelligently generate dimension-first N-dimensional sub-funnels based on dimensional consistency to initiate secondary analysis;

[0058] (4) It is impossible to perform intelligent data comparison and predict whether there is any risk in the data;

[0059] (5) It is not possible to intelligently provide the reasons for data loss in the funnel.

[0060] To address the aforementioned problems, this invention provides an intelligent funnel data processing solution. This solution allows for multi-dimensional configuration of the funnel, analysis of the data output by the funnel system, intelligent location of problems within the system, and analysis of the underlying causes, providing reliable technical support for subsequent improvements to the funnel system.

[0061] The inventive concept of this invention is as follows: In the process of analyzing various flow data using a funnel system, efficiency can be improved by manually configuring the funnel to initiate funnel analysis. Simultaneously, based on data evaluation at different funnel stages and prioritizing funnel dimension depth, a multi-dimensional sub-funnel system can be intelligently generated from the main funnel, effectively disassembling the original funnel. Furthermore, based on the data output from the sub-funnel system and the original funnel system, a comprehensive comparison and analysis of the funnel stage data can be performed to quickly identify the root causes of data loss at each stage, thereby improving operational efficiency and enabling rapid problem localization and analysis.

[0062] This solution can be applied to electronic devices such as computers, servers, service platforms, or other computer equipment that are capable of data processing, and this solution does not restrict the specific form of the electronic device.

[0063] The technical solution of the intelligent funnel data processing method provided by the present invention will be described in detail below through several specific embodiments.

[0064] Figure 1 This is a schematic diagram of the universal funnel data analysis system provided by the present invention, as shown below. Figure 1 As shown, in this general funnel data analysis system, the analysis process for traffic data, etc., consists of a unified modeling stage, a process configuration stage, a dynamic parsing stage, and a dynamic execution stage.

[0065] (1) Unified modeling stage: This stage will unify the modeling of user behavior records related topics, such as clicks, traffic, order logs, orders, users, products, etc. The main tasks in the unified modeling process are to ensure the consistency of dimensions (the same field in different tables represents the same meaning) and to distinguish between dimensions and attributes.

[0066] (2) Process Configuration Stage: This stage mainly defines the funnel name, plans and sets the funnel steps (i.e., process path): for example: S1 Homepage -> S2 Activity Page -> S3 Product Details Page -> S4 Shopping Cart -> S5 Checkout Page. Planning and setting the analysis funnel steps requires defining initial entry conditions and step conditions. For example: the initial entry condition for the funnel process is users in Beijing; the condition is that the product is a JD.com self-operated product. Other conditions can be added in each step. Finally, plan and set the dimensions to be analyzed (multiple dimensions are possible), using ABC dimension key-value pairs.

[0067] (3) Dynamic parsing stage: Multiple processes are started according to the number of funnels configured. Each process reads one funnel and supports data processing by a single funnel system.

[0068] After reading the funnel, a self-check is initiated, primarily verifying the configuration conditions and the correctness of the steps. Upon completion of the self-check, rule parsing is performed to generate the process steps, including lexical analysis (lexical analysis: mapping different data tables using custom keywords, e.g., 'p' represents the page table). During parsing, key information such as the maximum number of steps in the funnel (step N>=2) and the order of the process steps can be determined.

[0069] (4) Dynamic execution stage: Based on (3), specific funnel steps can be generated. Each step will generate a step table according to the funnel expression. The next step depends on the previous step, and finally the data is stored in the database.

[0070] Data is aggregated based on the dimensions and metrics of the analysis (UV or frequency, etc.), and finally, the summary data of each funnel is verified and generated and copied to the official table. Failed funnels will be retried and error logs will be entered. The configuration determines whether to clear the intermediate temporary table data. At this point, the funnel system's processing flow for traffic data is completed.

[0071] Based on the aforementioned general funnel data analysis process for processing flow data, this invention provides a specific scheme for intelligent analysis of the processing results of a funnel system. Several specific embodiments are described below.

[0072] Figure 2 The flowchart of Embodiment 1 of the intelligent funnel data processing method provided by the present invention is as follows: Figure 2 As shown, the data processing method for this intelligent funnel specifically includes the following steps:

[0073] S101: Based on the pre-configured set of atomic dimensions and dimension depth, multiple analysis dimensions are obtained using the dimension-first principle.

[0074] After analyzing the flow data, the intelligent funnel system generates processed data and indicators such as conversion rate. The electronic equipment needs to judge the processed data and conversion rate according to preset conditions to determine whether the data meets the requirements or whether the conversion rate has reached the expectation. If the conditions are not met, it is determined that further intelligent analysis of the funnel system is needed; if the conditions are met, intelligent analysis of the funnel system is not required.

[0075] In this step, when the electronic device determines that further intelligent analysis of the funnel system is needed, it first generates multiple analysis dimensions based on the standard atomic dimensions in the pre-configured set of atomic dimensions (also known as the dimension pool) and the preset dimension depth, following a dimension depth-first principle. Any analysis dimension in these multiple analysis dimensions includes one or more standard atomic dimensions. That is, an analysis dimension is composed of one or more standard atomic dimensions. If an analysis dimension consists of multiple standard atomic dimensions, the number of standard atomic dimensions in that analysis dimension cannot exceed the aforementioned dimension depth.

[0076] In a specific implementation, the dimension depth priority principle refers to the fact that, when the number of atomic standard dimensions does not exceed the dimension depth, all possible combinations can be obtained from the set of atomic dimensions, such as one atomic standard dimension, two atomic standard dimensions, and so on, up to the number of atomic standard dimensions corresponding to the dimension depth, thus obtaining the aforementioned multi-analysis dimensions.

[0077] S102: Perform multi-dimensional data analysis in each sub-funnel system of the funnel system to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different.

[0078] In this step, once the electronic device acquires the multiple analytical dimensions for intelligent analysis of the funnel system, it needs to return to the original funnel system for further data processing. Unlike before, the intelligent analysis process requires breaking down the funnel system into multiple sub-funnel systems for data processing. Each sub-funnel system includes a sub-funnel, and its processing flow is consistent with the original funnel system. These sub-funnel systems correspond one-to-one with the aforementioned multiple analytical dimensions. That is, each sub-funnel system uses one analytical dimension for traffic data processing. After processing the traffic data, each sub-funnel system will obtain corresponding node data and conversion rates. These node data and conversion rates can be used to analyze the steps and atomic standard dimensions within the corresponding sub-funnel system.

[0079] S103: Based on the node data and conversion rate corresponding to each analysis dimension, determine the problematic funnel step and the cause of the problem.

[0080] In this step, after obtaining the execution results of different analysis dimensions output by each sub-funnel system, which are a large amount of node data and conversion rates, the output node data and conversion rates can be compared with the pre-configured data thresholds and conversion rate thresholds for different analysis dimensions to determine the specific analysis dimension with the problem. Further, the sub-funnel system with the problem can be identified.

[0081] From the perspective of the entire funnel system, once the sub-funnel system is identified, the node data can be further compared to determine the specific funnel steps. By analyzing the atomic dimensions corresponding to the problematic funnel steps, the cause of the problem can be determined.

[0082] The intelligent funnel data processing method provided in this embodiment obtains multiple different analysis dimensions by combining standard atomic dimensions. The original funnel system is decomposed into multiple sub-funnel systems that perform different analysis dimensions, thereby obtaining the processing results of traffic data processing under different analysis dimensions. Based on these processing results and comparing them with preset index thresholds, the problematic steps and causes in the funnel system can be determined. Compared with existing technologies, this method can more efficiently and accurately identify the problematic links in the funnel process and also determine the reasons for data loss.

[0083] Figure 3 The flowchart of Embodiment 2 of the intelligent funnel data processing method provided by the present invention is as follows: Figure 3 As shown, based on the above embodiments, the method for processing smart funnel data further includes the following steps:

[0084] S201: Obtain node data and / or conversion rate of the funnel system for the input flow data.

[0085] In this step, the funnel system processes the input flow data and outputs the corresponding processing results, which generally include node data from different steps and the final conversion rate.

[0086] S202: If the node data is less than the preset node data threshold or the conversion rate is less than the preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required.

[0087] S203: If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

[0088] In the two steps above, before using the funnel system for traffic data analysis, analysts or R&D personnel can set indicator thresholds based on actual needs or historical data analysis, such as node data thresholds and conversion rate thresholds.

[0089] After obtaining the node data and conversion rate output by the funnel system, they are compared with the corresponding node data thresholds and conversion rate thresholds to determine whether intelligent analysis of the funnel system is required.

[0090] In this scheme, if at least one of the node data or conversion rate fails to reach a preset threshold, further analysis is required to determine the cause of the problem. In other words, if the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, it indicates that further intelligent analysis of the funnel system is needed.

[0091] If both node data and conversion rate reach the set thresholds, then the conversion rate and other indicators have met the expected results, and there is no longer a need for intelligent analysis of the funnel system.

[0092] Figure 4 The flowchart of Embodiment 3 of the intelligent funnel data processing method provided by the present invention is as follows: Figure 4 As shown, based on the above embodiment, step S103 determines the problematic funnel step and the cause of the problem according to the node data and conversion rate corresponding to each analysis dimension, specifically including the following steps:

[0093] S1031: Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data graph is generated, in which problematic node data and conversion rates are identified.

[0094] In this step, after obtaining the node data and conversion rates from the different sub-funnel systems, a data graph representing the processing results of each sub-funnel system can be generated based on this output data. This data graph must at least represent the node data of each sub-funnel system. Furthermore, the data graph needs to show the data processed by the original funnel system and the nodes where the conversion rate is problematic, and these nodes need to be marked in the graph.

[0095] S1032: Based on the data map, determine the problematic funnel step and the cause of the problem, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

[0096] In this step, based on the data presented in the data graph, the sub-funnel system closest to the problematic data and conversion rate node in the original funnel system is selected from multiple sub-funnel systems. This sub-funnel system can include one or more sub-funnel systems, and one or more funnel steps within these sub-funnel systems can be identified as the steps most likely to cause problems. The electronic device can then identify the funnel steps in these types of sub-funnel systems as the problematic funnel steps. Furthermore, these problematic funnel steps and node data can be analyzed to determine the cause of the problem.

[0097] Figure 5 A flowchart illustrating an example of processing intelligent funnel data provided by the present invention is shown below. Figure 5 As shown, in the specific implementation of this scheme, it should be understood that the intelligent analysis process is a pluggable process of the funnel system process. The intelligent analysis process mainly adds intelligent threshold checking based on singularities, depth-first dimension intelligent decomposition, data comparison, and other processes to the original funnel system. This intelligent funnel analysis process includes at least the funnel recursive system, as well as the subsequent decision node stage, intelligent dimension stage, data comparison stage, and result output stage. The specific process steps are as follows:

[0098] S1: Funnel Recursive System Stage

[0099] S1-1, the funnel recursive system mainly relies on the original funnel system. Due to the activation of intelligent analysis, a hidden sub-funnel system is generated. This sub-funnel system is mainly obtained by continuously generating sub-funnels with the same process as the original funnel but different analysis dimensions.

[0100] S2: Decision-making stage

[0101] S2-1: After the funnel system completes its execution, it will obtain node data and conversion rates, among other results. During the decision-making phase, it will intrude to check the data status of each process node. It can also check the final conversion rate.

[0102] S2-2: If the node data meets the user-configured node data threshold or the conversion rate meets the conversion rate threshold, the process ends and no further analysis is needed. If the node data or conversion rate is less than the set threshold, then further analysis is initiated.

[0103] S2-3, if analysis needs to be restarted, proceed to the intelligent dimension stage.

[0104] S3: Intelligent Dimension Stage

[0105] S3-1, if the funnel system needs to be re-analyzed, the program will generate multiple analysis dimensions based on the standard atomic dimensions in the atomic dimension set (i.e., the dimension pool), using a dimension depth-first principle. For example, if the dimension priority is set to N (>=1), this will generate one-dimensional or multi-dimensional dimensions, and the dimension depth is set to 3 (the standard atomic dimensions in the set include A, B, C, D, E, F). At this stage, a maximum of three types of analysis dimensions will be recursively generated. Specifically, an analysis dimension composed of one standard atomic dimension can be generated, i.e., A, B, C, D, E, F; an analysis dimension composed of two standard atomic dimensions can be generated, i.e., AB, AC, AD, AE, AF, BC... and so on up to EF; an analysis dimension composed of three standard atomic dimensions can be generated, i.e., ABC, ABD, ABE, ABF, BCD, BCE, etc., generated according to mathematical permutations and combinations. Finally, multiple analysis dimensions are obtained, and then the generated analysis dimension combinations are returned to the funnel recursive stage for flow data processing.

[0106] S3-2, In the process of generating multiple analysis dimensions with a focus on dimension depth, intrusive analysis dimension configuration can be performed based on historical problem dimensions. In other words, analysis dimensions can be added manually. For example, if the funnel system previously identified data problems in the funnel due to ABC dimension analysis, then in this intelligent analysis process, the ABC dimension can be added as an additional analysis dimension, and the analysis of that dimension can be configured to be started first.

[0107] In step S3-3, the obtained multi-dimensional analysis is used again to process the traffic data. These various dimensions generate a large dataset, including at least node data and related conversion rates. The intelligent analysis process then enters the data comparison phase.

[0108] S4: Data Comparison Phase

[0109] S4-1: After entering this stage, based on the large amount of node data generated in S3, the conversion rate, and the threshold inflection point problem data in S1, the steps of each sub-funnel system are aligned, and the node indicators are processed according to the set indicator thresholds. (There are some analysis dimensions with no steps or output indicators that need to be processed).

[0110] S4-2: Based on the funnel steps generated in S4-1 and the node data output by each sub-funnel system, a data graph is formed. This data graph displays the data of the problem node and the data from the sub-funnels. Then, a clustering algorithm (this process can be manually configured with other analysis methods) is used to process the data graph. The initial point is the location of the problem data. The sub-funnel system closest to the location of the problem data is found. The steps in this sub-funnel are the steps most likely to cause problems. Therefore, these sub-funnels can be grouped together to output a funnel list. This list can include one or more sub-funnel systems. Furthermore, the dimensions and steps where problems occur can be determined based on these sub-funnel systems.

[0111] S5: Results Output Stage

[0112] The list of problematic funnels can be presented to operations staff through the graphical user interface of an electronic device, or it can be output through other external devices for operations staff to confirm.

[0113] The intelligent funnel data processing method provided in the embodiments of this application, based on the output data and conversion rate of the funnel system, splits the original funnel by generating multiple analysis dimensions at the dimensional stage, resulting in multiple sub-funnel systems that perform data analysis processing on different analysis dimensions to obtain node data and conversion rate data. Based on the execution results of the original funnel system and the execution results of each sub-funnel system, a comprehensive analysis and comparison is performed to quickly locate the steps where the funnel malfunctions and determine the cause of the problem, thereby solving the problem that the original technology cannot perform intelligent location and cause analysis.

[0114] Figure 6 This is a schematic diagram of the structure of an embodiment of the intelligent funnel data processing device provided by the present invention; as shown below. Figure 6 As shown, the intelligent funnel data processing device 10 includes:

[0115] The first processing module 11 is used to obtain multiple analysis dimensions based on a pre-configured set of atomic dimensions and dimension depth, using a dimension-first principle.

[0116] The second processing module 12 is used to perform the multi-dimensional data analysis in each sub-funnel system of the funnel system to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different.

[0117] The third processing module 13 is used to determine the problematic funnel step and the cause of the problem based on the node data and conversion rate corresponding to each analysis dimension.

[0118] The intelligent funnel data processing device provided in this embodiment is used to execute the technical solution provided in any of the foregoing method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0119] Figure 7 This is a schematic diagram of the structure of a second embodiment of the intelligent funnel data processing device provided by the present invention; as shown below. Figure 7 As shown, the intelligent funnel data processing device 10 further includes: a fourth processing module 14, used for:

[0120] Obtain node data and / or conversion rate of the funnel system for the input flow data;

[0121] If the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required.

[0122] If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

[0123] Based on any of the above embodiments, the first processing module 11 is specifically used for:

[0124] Based on the preset dimension depth, the dimensions in the preset atomic dimension set are arranged and combined according to the dimension priority principle to generate multiple analysis dimensions, wherein each analysis dimension includes one or more dimensions in the dimension set.

[0125] Furthermore, the first processing module 11 is also used for:

[0126] In response to the user's operation, the identified historical problem dimensions are added to the multi-analysis dimensions to obtain new multi-analysis dimensions.

[0127] Optionally, the third processing module 13 is specifically used for:

[0128] Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data map is generated, in which problematic node data and conversion rates are identified;

[0129] Based on the data map, the problematic funnel step and the cause of the problem are determined, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

[0130] The intelligent funnel data processing device provided in any of the foregoing embodiments is used to execute the technical solution provided in any of the foregoing method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0131] Figure 8 This is a schematic diagram of the structure of an embodiment of the electronic device provided by the present invention; as shown below. Figure 8 As shown, the electronic device 100 includes:

[0132] Processor 111, memory 112 and interaction interface 113;

[0133] The memory 112 is used to store the executable instructions of the processor 111;

[0134] The processor 111 is configured to execute the intelligent funnel data processing method provided in any of the foregoing method embodiments by executing the executable instructions.

[0135] Optionally, the memory 112 can be either standalone or integrated with the processor 111.

[0136] Optionally, when the memory 112 is a device independent of the processor 111, the electronic device 100 may further include:

[0137] A bus is used to connect the aforementioned devices.

[0138] The electronic device is used to execute the technical solutions in any of the foregoing method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0139] This invention also provides a readable storage medium storing a computer program thereon, which, when executed by a processor, implements the intelligent funnel data processing method provided in any of the foregoing method embodiments.

[0140] This invention also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the intelligent funnel data processing method provided in any of the foregoing method embodiments.

[0141] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing intelligent funnel data, characterized in that, include: Based on the pre-configured set of atomic dimensions and dimension depth, multiple analysis dimensions are obtained using a dimension-first principle; The multi-dimensional data analysis is performed in each sub-funnel system of the funnel system to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different. Based on the node data and conversion rate corresponding to each analysis dimension, determine the problematic funnel step and the cause of the problem; The step of determining the problematic funnel step and the cause of the problem based on the node data and conversion rate corresponding to each analysis dimension includes: Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data map is generated, in which problematic node data and conversion rates are identified; Based on the data map, the problematic funnel step and the cause of the problem are determined, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

2. The method according to claim 1, characterized in that, The method further includes: Obtain node data and / or conversion rate of the funnel system for the input flow data; If the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required. If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

3. The method according to claim 1, characterized in that, The process of obtaining multiple analysis dimensions based on a pre-configured set of atomic dimensions and dimension depth, using a dimension-first principle, includes: Based on the preset dimension depth, the dimensions in the preset atomic dimension set are arranged and combined according to the dimension priority principle to generate multiple analysis dimensions, wherein each analysis dimension includes one or more dimensions in the dimension set.

4. The method according to claim 3, characterized in that, The method further includes: In response to the user's operation, the identified historical problem dimensions are added to the multi-analysis dimensions to obtain new multi-analysis dimensions.

5. A device for processing intelligent funnel data, characterized in that, include: The first processing module is used to obtain multiple analysis dimensions based on a pre-configured set of atomic dimensions and dimension depth, using a dimension-first principle. The second processing module is used to perform the multi-analysis dimension data analysis in each sub-funnel system of the funnel system to obtain the node data and conversion rate corresponding to each analysis dimension. Each sub-funnel system includes a sub-funnel generated according to the funnel system and consistent with the funnel system process. Each sub-funnel system performs data analysis of one analysis dimension, and the analysis dimensions performed by the sub-funnels in each sub-funnel system are different. The third processing module is used to determine the problematic funnel step and the cause of the problem based on the node data and conversion rate corresponding to each analysis dimension. The third processing module is specifically used for: Based on the pre-configured data indicators and conversion rate indicators for each node, and the node data and conversion rate output by each sub-funnel system in one analytical dimension, a data map is generated, in which problematic node data and conversion rates are identified; Based on the data map, the problematic funnel step and the cause of the problem are determined, wherein the problematic funnel step is closest to the node where the problem occurred, and the cause of the problem includes at least one sub-funnel system that includes the problematic funnel step.

6. The apparatus according to claim 5, characterized in that, The device further includes: a fourth processing module, used for: Obtain node data and / or conversion rate of the funnel system for the input flow data; If the node data is less than a preset node data threshold or the conversion rate is less than a preset conversion rate threshold, then it is determined that intelligent analysis of the funnel system is required. If the node data is greater than or equal to the node data threshold, and the conversion rate is greater than or equal to the conversion rate threshold, then it is determined that intelligent analysis of the funnel system is not required.

7. The apparatus according to claim 5, characterized in that, The first processing module is specifically used for: Based on the preset dimension depth, the dimensions in the preset atomic dimension set are arranged and combined according to the dimension priority principle to generate multiple analysis dimensions, wherein each analysis dimension includes one or more dimensions in the dimension set.

8. The apparatus according to claim 7, characterized in that, The first processing module is also used for: In response to the user's operation, the identified historical problem dimensions are added to the multi-analysis dimensions to obtain new multi-analysis dimensions.

9. An electronic device, characterized in that, include: Processor, memory, and interaction interface; The memory is used to store the executable instructions of the processor; The processor is configured to execute the intelligent funnel data processing method according to any one of claims 1 to 4 by executing the executable instructions.

10. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the intelligent funnel data processing method according to any one of claims 1 to 4.

11. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, is used to implement a method for processing intelligent funnel data as described in any one of claims 1 to 4.

Images