Index establishment method, device and equipment and data management system

Abstract

The embodiment of the invention provides an index establishment method, device and equipment and a data management system. The index establishment method comprises the following steps: acquiring an index establishment request; based on the index establishment request, attribute information and an atomic index used for establishing a preset index are determined, the attribute information at least comprises a time range of the preset index, the atomic index corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used for determining an index result of the data granularity and an index result of the overall granularity; and performing an index establishment operation based on the attribute information and the atomic index to obtain a preset index corresponding to the index establishment request. According to the technical scheme provided by the embodiment of the invention, the preset index is established through the attribute information and the atomic index, and the atomic index corresponds to the multi-layer aggregation operator, so that the index results of different granularities corresponding to the application scene can be determined when the preset index is processed, the application range is effectively expanded, and the application efficiency is improved. And the user can conveniently manage the data.

Description

Technical Field

[0001] The present invention relates to the field of information processing, and in particular to an indicator establishment method, device, equipment and data management system. Background Art

[0002] At present, data assets refer to data resources that are owned or controlled by individuals or enterprises and can bring future economic benefits to enterprises and are recorded in physical or electronic form. In the prior art, when defining indicators using data asset management platforms, the defined indicator processing granularity is often single and fixed. At this time, the indicator is only applicable to a single application scenario, such as the application scenario of control experiments, or the application scenario of information search and query operations, etc. This greatly limits the scope of application of the data management platform. Summary of the invention

[0003] The embodiments of the present invention provide an indicator establishment method, device, equipment and data management system, which can obtain indicator results of different granularities, thereby effectively expanding the scope of application of the data management platform.

[0004] In a first aspect, an embodiment of the present invention provides an indicator establishment method, comprising:

[0005] Get indicator creation request;

[0006] Based on the indicator establishment request, determine attribute information and atomic indicators used to establish the preset indicator, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity;

[0007] An indicator establishment operation is performed based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0008] In a second aspect, an embodiment of the present invention provides an indicator establishment device, comprising:

[0009] A first acquisition module is used to obtain an indicator establishment request;

[0010] A first determination module is used to determine attribute information and atomic indicators used to establish a preset indicator based on the indicator establishment request, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity;

[0011] The first processing module is used to perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0012] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the indicator establishment method in the above-mentioned first aspect.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer storage medium for storing a computer program, wherein the computer program enables a computer to implement the indicator establishment method in the first aspect when executed.

[0014] In a fifth aspect, an embodiment of the present invention provides a computer program product, comprising: a computer-readable storage medium storing computer instructions, which, when executed by one or more processors, causes the one or more processors to execute the steps in the indicator establishment method shown in the first aspect above.

[0015] In a sixth aspect, an embodiment of the present invention provides a data management system, including:

[0016] An indicator definition unit is used to obtain an indicator establishment request; based on the indicator establishment request, determine attribute information and atomic indicators used to establish a preset indicator, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity; perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request;

[0017] A control experiment unit is communicatively connected with the indicator definition unit, and is used to obtain a promotion plan to be processed; determine associated indicators and traffic allocation information corresponding to the promotion plan, and the traffic allocation information is used to determine a first user group applicable to the promotion plan and a second user group not applicable to the promotion plan; based on the associated indicators and the traffic allocation information, determine the feasibility of the promotion plan.

[0018] The index establishment method, device, equipment and data management system provided in the present embodiment obtain an index establishment request; determine the attribute information and atomic indicators used to establish the preset index based on the index establishment request, and then perform the index establishment operation based on the attribute information and the atomic indicators, so that the preset index corresponding to the index establishment request can be stably obtained. It should be noted that since the preset index is obtained by establishing atomic indicators corresponding to multiple layers of aggregation operators, when performing analysis and processing based on the preset index, the indicator results of the data granularity and the indicator results of the overall granularity can be determined. The above-mentioned indicator results of the data granularity and the indicator results of the overall granularity can be applicable to different application scenarios. For example, the indicator results of the data granularity can be applicable to the application scenario of the control experiment, and the indicator results of the overall granularity can be applicable to the application scenario of the market analysis, thereby effectively expanding the scope of application corresponding to the method. This not only makes it convenient for users to manage data in different application scenarios, but also reduces the cost of data management, further improving the practicality of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of a scenario of an indicator establishment method provided by an embodiment of the present invention;

[0021] Figure 2 A flowchart of an indicator establishment method provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of generating an atomic indicator for establishing a preset indicator provided by an embodiment of the present invention;

[0023] Figure 4 A flowchart of another indicator establishment method provided by an embodiment of the present invention;

[0024] Figure 5 A flowchart of another indicator establishment method provided by an embodiment of the present invention;

[0025] Figure 6 A flowchart of another indicator establishment method provided by an embodiment of the present invention;

[0026] Figure 7 A flowchart of another indicator establishment method provided by an embodiment of the present invention;

[0027] Figure 8 A structural diagram of a data management platform provided by an application embodiment of the present invention;

[0028] Fig. 9 A schematic diagram of a flow chart of an indicator definition operation provided by an application embodiment of the present invention;

[0029] Fig.10 A schematic diagram of the control experiment operation process provided for the application example of the present invention;

[0030] Fig.11 A schematic diagram of the structure of an indicator establishment device provided by an embodiment of the present invention;

[0031] Fig.12 For Fig.11 A schematic diagram of the structure of an electronic device corresponding to the indicator establishment device provided in the illustrated embodiment;

[0032] Fig.13 A structural diagram of a data management system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "an", and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two, but does not exclude the situation of including at least one.

[0035] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0036] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0037] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.

[0038] In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0039] Definition of terms:

[0040] A / B experiment: also known as control experiment or A / B test, it randomly assigns users to different variants to obtain experimental group users and control group users. It verifies the effects of different versions of products by statistically analyzing the operating data of the experimental group users and the operating data of the control group users. It is often used in the Internet, medicine and other industries.

[0041] OneData methodology: is a methodology for big data construction, including data standard specifications, model specifications, data research and development specifications, data service specifications, etc., and also includes related tools for the implementation of specifications.

[0042] KeyType: is the data granularity used in the A / B experiment, and identifies the granularity of data from which the indicator value is aggregated, for example: buyer ID, device ID.

[0043] Anomaly analysis: It is a complete solution for anomaly detection and attribution of time series data, which is used to solve the detection and attribution of daily fluctuations of indicators.

[0044] Causal analysis: Based on the causal discovery algorithm, a causal network diagram of the scenario is generated through observation data. The causal diagram can be used to find the causal structure associated with the target, providing a reference direction for optimizing application effects.

[0045] In order to facilitate understanding of the specific implementation process and implementation effect of the indicator establishment method, device, equipment and data management system in this embodiment, the relevant technology is briefly described below:

[0046] Data assets refer to data resources that are owned or controlled by individuals or enterprises and can bring future economic benefits to the enterprise and are recorded in physical or electronic form. Due to the large number of data assets, users often manage data resources through data asset platforms. For data asset platforms, current data management platforms tend to combine insight analysis, experiments, and data asset management into a single product, such as: data asset platforms specifically used for data asset management operations and asset upstream and downstream management operations, data asset platforms specifically used for data visualization and analytical insight functions, data asset platforms specifically used for A / B experimental operations, etc.

[0047] However, the data management capabilities of the above-mentioned data asset platforms are relatively scattered, which is prone to the following problems: data assets cannot be shared, and the same indicator cannot be observed and analyzed in A / B experiments and dashboards at the same time, resulting in the inability to achieve a unified definition of caliber, data differences, and reduced credibility of analysis and experimental results. In addition, since the application field of data asset operations is relatively wide, in order to realize a data asset platform with comprehensive capabilities, when establishing a data asset platform, the rules corresponding to the entire data production line or data node are often required. The entire data production line needs to perform data processing operations according to unified indicators or rules, which greatly increases the difficulty of establishing a data asset platform. Therefore, data asset platforms rarely have data products that combine the three capabilities of data asset management, experimental platform, and insight analysis.

[0048] In addition, when users use the data asset platform, in order to implement asset management operations, users are often required to define indicators for application scenarios. However, since the defined indicators are defined for specific application scenarios, the corresponding indicator processing methods and data processing granularity are often only applicable to the above application scenarios, but not to other application scenarios, resulting in inconsistent indicator definition calibers in different application scenarios. This not only makes it inconvenient for users to manage data, but also increases the cost of data management.

[0049] In order to solve the above technical problems, this embodiment provides an indicator establishment method, device, equipment and data management system. Figure 1As shown, the execution subject of the indicator establishment method is the indicator establishment device 200, and the indicator establishment device 200 is connected to the client 100 in communication. It should be noted that the indicator establishment device 200 can be implemented as a personal computer, a tablet computer, a local server or a server in the cloud. At this time, when the indicator establishment device 200 is implemented as a server in the cloud, the indicator establishment method can be executed in the cloud, and several computing nodes (cloud servers) can be deployed in the cloud, and each computing node has computing, storage and other processing resources. In the cloud, multiple computing nodes can be organized to provide a certain service. Of course, a computing node can also provide one or more services. The way to provide the service in the cloud can be to provide a service interface to the outside, and the user calls the service interface to use the corresponding service. The service interface includes software development kits (Software Development Kit, referred to as SDK), application programming interfaces (Application Programming Interface, referred to as API) and other forms.

[0050] Among them, the client 100 is used for users to apply so as to realize the index establishment operation. The above-mentioned client can be any computing device with a certain data transmission capability. When it is specifically implemented, the client 100 can be a mobile phone, a personal computer PC, a tablet computer, a setting application, etc. In addition, the basic structure of the client 100 may include: at least one processor. The number of processors depends on the configuration and type of the client 100. The client 100 may also include a memory, which may be volatile, such as a random access memory (Random Access Memory, referred to as RAM), or non-volatile, such as a read-only memory (Read-Only Memory, referred to as ROM), flash memory, etc., or may also include both types. The memory usually stores an operating system (Operating System, referred to as OS), one or more application programs, and may also store program data, etc. In addition to the processing unit and the memory, the client 100 also includes some basic configurations, such as a network card chip, an IO bus, a display component, and some peripheral devices, etc. Optionally, some peripheral devices may include, for example, a keyboard, a mouse, an input pen, a printer, etc. Other peripheral devices are well known in the art and will not be described in detail here.

[0051] The indicator establishment device 200 refers to a device that can provide indicator establishment operations in a network virtual environment, and generally refers to a device that uses the network to perform information planning and indicator establishment operations. In physical implementation, the indicator establishment device 200 can be any device that can provide computing services, respond to indicator establishment requests, and can perform indicator establishment operations based on indicator establishment requests, for example: it can be a cluster server, a conventional server, a cloud server, a cloud host, a virtual center, etc. The composition of the indicator establishment device 200 mainly includes a processor, a hard disk, a memory, a system bus, etc., which is similar to a general computer architecture.

[0052] In the above embodiment, the client 100 is connected to the indicator establishment device 200 through a network, and the network connection can be a wireless or wired network connection. If the client 100 can be connected to the indicator establishment device 200 through communication, the network standard of the mobile network can be any one of 2G (GSM), 2.5G (GPRS), 3G (WCDMA, TD-SCDMA, CDMA2000, UTMS), 4G (LTE), 4G+ (LTE+), WiMax, 5G, 6G, etc.

[0053] The client 100, when the user has an indicator establishment demand, the client 100 can generate or obtain an indicator establishment request. In some instances, the indicator establishment request can be obtained based on a human-computer interaction operation. At this time, the client 100 can display a human-computer interaction interface, obtain the interaction operation input by the user in the human-computer interaction interface, and obtain and determine the indicator establishment request based on the interaction operation; or, the indicator establishment request can be obtained by transmitting through a preset interface or a third device. At this time, obtaining the indicator establishment request may include: obtaining a communication interface on the client 100; obtaining the indicator establishment request actively or passively transmitted by the third device based on the communication interface. After obtaining the indicator establishment request, in order to implement the indicator establishment operation, the indicator establishment request can be sent to the indicator establishment device 200, so that the indicator establishment device 200 obtains the indicator establishment request.

[0054] The indicator establishment device 200 is used to obtain an indicator establishment request, and determine the attribute information and atomic indicators used to establish a preset indicator based on the indicator establishment request, wherein the attribute information at least includes a time range for establishing the preset indicator; and the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator results of the data granularity (keytype) and the indicator results of the overall granularity; after obtaining the multi-layer aggregation operator, attribute information and atomic indicators, the indicator establishment operation can be performed based on the attribute information and atomic indicators to obtain the preset indicator corresponding to the indicator establishment request, thereby effectively realizing the indicator establishment operation.

[0055] The technical solution provided in this embodiment can stably obtain the preset indicators corresponding to the indicator establishment request. It should be noted that since the preset indicators are obtained by establishing atomic indicators corresponding to multi-layer aggregation operators, the indicator results of the data granularity (keytype) and the indicator results of the overall granularity can be determined when analyzing and processing based on the preset indicators. The indicator results of the above data granularity (keytype) and the indicator results of the overall granularity can be applicable to different application scenarios. For example, the indicator results of the data granularity (keytype) can be applicable to the application scenario of the control experiment, and the indicator results of the overall granularity can be applicable to the application scenario of the market analysis, thereby effectively expanding the scope of application corresponding to the method. This not only makes it convenient for users to manage data in different application scenarios, but also reduces the cost of data management, further improving the practicality of the method.

[0056] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the case where there is no conflict between the embodiments, the following embodiments and the features in the embodiments can be combined with each other. In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0057] Reference Figure 2 As shown, this embodiment provides an index establishment method, and the execution subject of the method is an index establishment device. It can be understood that the index establishment device can be implemented as software, or a combination of software and hardware. Specifically, when the index establishment device is implemented as hardware, it can be various electronic devices with index establishment operations, including but not limited to personal computers, servers, etc. When the index establishment device is implemented as software, it can be installed in the electronic devices listed above. Based on the above-mentioned index establishment device, the index establishment method may include:

[0058] Step S201: Obtain an indicator establishment request.

[0059] Step S202: Based on the indicator establishment request, determine the attribute information and atomic indicators used to establish the preset indicators, where the attribute information at least includes the time range used to establish the preset indicators, wherein the atomic indicators correspond to multiple layers of aggregation operators, and the multiple layers of aggregation operators are used to determine the indicator results of the data granularity (keytype) and the indicator results of the overall granularity.

[0060] Step S203: Perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0061] The specific implementation process and effects of each of the above steps are described in detail below:

[0062] Step S201: Obtain an indicator establishment request.

[0063] Among them, the indicator establishment device can realize the unified definition or unified establishment operation of indicators suitable for different application scenarios (for example: control experiment scenarios, market statistics scenarios, etc.), and after the indicators are uniformly defined or uniformly established, the established indicators can also be stored and managed.

[0064] Specifically, when a user uses an indicator establishment device, the corresponding data processing operation can be directly performed based on the indicator stored in the indicator establishment device. When there is no preset indicator in the indicator establishment device, the user can perform the establishment operation of the preset indicator through the indicator establishment device, so that the indicator establishment device can obtain the indicator establishment request. In some instances, the indicator establishment request can be obtained through human-computer interaction operations. At this time, obtaining the indicator establishment request may include: displaying a human-computer interaction interface; obtaining an execution operation input by the user in the human-computer interaction interface; obtaining an indicator establishment request based on the execution operation, and the indicator establishment request may include identification information of the preset indicator to be established.

[0065] In other instances, the indicator establishment request can be obtained not only through human-computer interaction operations, but also through a client or a third device. In this case, obtaining the indicator establishment request may include: obtaining a client or a third device that is communicatively connected to the indicator establishment device, wherein the indicator establishment request is generated or stored in the client or the third device; and then actively or passively obtaining the indicator establishment request through the client or the third device, thereby effectively ensuring the accuracy and reliability of obtaining the indicator establishment request.

[0066] Step S202: Based on the indicator establishment request, determine the attribute information and atomic indicators used to establish the preset indicators, where the attribute information includes at least the time range of the preset indicators, wherein the atomic indicators correspond to multiple layers of aggregation operators, and the multiple layers of aggregation operators are used to determine the indicator results of the data granularity (keytype) and the indicator results of the overall granularity.

[0067] Among them, the preset indicators are composed of at least a time range and atomic indicators. The above-mentioned atomic indicators are used to constitute the main content of the preset indicators. The time range can be one of the attribute information used to limit the preset indicators. For example, the atomic indicators may include at least one of the following: the number of buyers, the number of sellers, the transaction volume, the order volume, etc., and the preset indicators may include at least one of the following: the number of buyers in the last three days, the transaction volume in the last seven days, the order volume in the last week, the number of sellers in the last month, etc.

[0068] Based on the above content, it can be known that after obtaining the indicator establishment request, in order to realize the establishment operation of the preset indicator, it is necessary to first determine the attribute information and atomic indicators used to establish the preset indicator, wherein the attribute information at least includes the time range of the preset indicator, and the time range may include at least one of the following: three days, five days, seven days, one day, ten days, etc. It should be noted that the attribute information may not only include the time range of the preset indicator, but also include other attribute information of the preset indicator, for example, the field information of the preset indicator, the group information to which the preset indicator belongs, the level information of the preset indicator, etc. The technicians in this field can configure and adjust the attribute information used to establish the preset indicator according to the specific application scenario or application requirements, which will not be repeated here.

[0069] In some instances, attribute information and atomic indicators can be determined through human-computer interaction operations. At this time, based on the indicator establishment request, determining the attribute information and atomic indicators used to establish preset indicators may include: displaying a human-computer interaction interface based on the indicator establishment request; obtaining an execution operation input by the user in the human-computer interaction interface; and determining the attribute information and atomic indicators used to establish the preset indicators based on the execution operation.

[0070] In other instances, attribute information can not only be obtained through human-computer interaction operations, but also be obtained by selecting from multiple established alternative attribute information. In this case, based on the indicator establishment request, determining the attribute information used to establish the preset indicator may include: based on the indicator establishment request, determining multiple established alternative attribute information; obtaining the attribute selection operation input by the user from the multiple alternative attribute information; based on the attribute selection operation, selecting one or more alternative attribute information from the multiple alternative attribute information as the attribute information for establishing the preset indicator. For example, after obtaining the indicator establishment request, the following multiple alternative attribute information "country of use", "user level of use", "indicator group", "indicator level" and so on can be displayed in a menu bar in the display interface. When the user inputs a selection operation for the alternative attribute information "indicator level", the "indicator level" can be determined as the attribute information, thereby effectively ensuring the accuracy and reliability of the determination of the attribute information.

[0071] Similarly, atomic indicators can not only be obtained through human-computer interaction operations, but also be determined based on multiple established atomic indicators. At this time, based on the indicator establishment request, determining the atomic indicator used to establish a preset indicator may include: based on the indicator establishment request, obtaining multiple established atomic indicators; identifying whether there is a target atomic indicator corresponding to the preset indicator among the multiple established atomic indicators; if so, determining the target atomic indicator as the atomic indicator corresponding to the preset indicator; if not, generating an atomic indicator for establishing the preset indicator.

[0072] Specifically, after obtaining the indicator establishment request, multiple established atomic indicators can be obtained based on the indicator establishment request, and then it can be identified whether there is a target atomic indicator corresponding to the preset indicator among the multiple established atomic indicators. If the multiple established atomic indicators include the target atomic indicator corresponding to the preset indicator, the target atomic indicator can be determined as the atomic indicator corresponding to the preset indicator; if the multiple established atomic indicators do not include the target atomic indicator corresponding to the preset indicator, an atomic indicator generation operation can be performed to generate and obtain the atomic indicator used to establish the preset indicator, which effectively ensures the stable reliability of the determination of the atomic indicator used to establish the preset indicator.

[0073] For example, after obtaining the indicator establishment request, multiple established atomic indicators such as "number of buyers", "number of sellers", "transaction volume", "transaction amount", "number of orders", etc. can be determined based on the indicator establishment request, and the user determines whether the above-mentioned multiple established atomic indicators meet the atomic indicators of the preset requirements. If so, the above-mentioned atomic indicator "transaction volume" that meets the preset requirements can be used as the atomic indicator for establishing the preset indicator; if not, for example, the user wants the atomic indicator "logistics and transportation efficiency", and the above-mentioned atomic indicator does not exist in the above-mentioned established atomic indicators, then a new atomic indicator for establishing the preset indicator can be created, and then the indicator establishment operation can be performed based on the established atomic indicator.

[0074] Step S203: Perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0075] After acquiring the attribute information and atomic indicators, an indicator establishment operation can be performed based on the attribute information and the atomic indicators to obtain a preset indicator corresponding to the indicator establishment request. In some instances, the indicator establishment operation can be implemented through a pre-trained machine learning model or a neural network model. At this time, performing an indicator establishment operation based on the attribute information and the atomic indicators to obtain a preset indicator corresponding to the indicator establishment request can include: acquiring a pre-trained machine learning model or a neural network model, inputting the attribute information and the atomic indicators into the machine learning model or the neural network model, so that the neural network model can perform an indicator establishment operation based on the attribute information and the atomic indicators, thereby obtaining the preset indicator corresponding to the indicator establishment request output by the machine learning model or the neural network model, thereby effectively implementing the preset indicator establishment operation.

[0076] In some other instances, the indicator establishment operation can be implemented through a pre-trained machine learning model or a neural network model, and can also be combined with different scenario-limiting conditions to perform the indicator establishment operation. At this time, performing the indicator establishment operation based on attribute information and atomic indicators to obtain the preset indicator corresponding to the indicator establishment request may include: obtaining the scenario-limiting conditions corresponding to the preset indicator; performing the indicator establishment operation based on the attribute information, the scenario-limiting conditions and the atomic indicators to obtain the preset indicator corresponding to the indicator establishment request.

[0077] Specifically, since different preset indicators need to be established in different application scenarios, for example, application scenarios applicable to the application side, application scenarios applicable to the personal computer side, application scenarios applicable to the browser, application scenarios applicable to preset areas or preset countries, etc., in order to generate corresponding preset indicators in combination with different application scenarios, the scene limitation conditions corresponding to the preset indicators can be obtained, wherein the scene limitation conditions can be determined based on human-computer interaction operations or the current scene information.

[0078] After obtaining the scene limiting conditions corresponding to the preset indicators, the indicator establishment operation can be performed based on the attribute information, the scene limiting conditions and the atomic indicators, so as to obtain the preset indicators corresponding to the indicator establishment request. Among them, the specific implementation method and implementation effect of the indicator establishment operation are similar to the specific implementation method and implementation effect of the indicator establishment operation, and the specific contents can be referred to the above statements, which will not be repeated here.

[0079] In other instances, after obtaining the preset indicator corresponding to the indicator establishment request, in order to improve the practicality of the method, the method can also record and store any operation records performed by the user based on the preset indicator. At this time, the method in this embodiment can also include: obtaining the indicator processing operation corresponding to the preset indicator, the indicator processing information includes at least one of the following: indicator editing operation, indicator calling operation, indicator modification operation; processing the preset indicator based on the indicator processing operation to obtain the processing result corresponding to the indicator processing operation; and storing the processing result in association with the indicator processing operation.

[0080] Specifically, when a user has a processing requirement for a preset indicator, after obtaining the preset indicator corresponding to the indicator establishment request, the user can perform any processing operation on the established preset indicator according to the processing requirement. At this time, the indicator processing operation corresponding to the preset indicator can be obtained. The indicator processing operation may include at least one of the following: an indicator editing operation for implementing an editing processing operation on the preset indicator, an indicator calling operation for calling the preset indicator, an indicator modification operation for modifying the preset indicator, and the like. After obtaining the indicator processing operation, the preset indicator can be processed based on the indicator processing operation, so as to obtain a processing result corresponding to the indicator processing operation. In some instances, when the indicator processing operation is an indicator editing operation, the edited indicator and the editing operation record can be obtained; when the indicator processing operation is an indicator calling operation, the indicator calling record can be obtained; when the indicator processing operation is an indicator modification operation, the modified indicator and the indicator modification record can be obtained.

[0081] In order to facilitate users to view the processing operations and related information of preset indicators at any time, after obtaining the processing results, the processing results can be associated with the indicator processing operations for storage operations. For example: when the preset indicator has undergone multiple editing operations and there are multiple version indicators, the multiple version indicators corresponding to the preset indicator and the editing operation records corresponding to each version indicator can be viewed; when the preset indicator has undergone multiple call operations, the multiple call operations corresponding to the preset indicator and the call records corresponding to each call operation can be viewed, so that users can directly and quickly view historical version information, upstream and downstream information, and present the ins and outs of the preset indicators, further improving the practicality of this method.

[0082] In some other instances, after obtaining the preset indicator corresponding to the indicator establishment request, the method in this embodiment can also perform credibility authentication operations on the preset indicator. At this time, the method in this embodiment can also include: obtaining the credibility of the preset indicator; when the credibility is greater than or equal to a preset threshold, adding a trusted identifier to the preset indicator; and displaying the trusted identifier in association with the preset indicator.

[0083] Specifically, after obtaining the preset indicator corresponding to the indicator establishment request, the preset indicator can be analyzed and processed to obtain the credibility of the preset indicator. In some instances, the credibility of the preset indicator can be obtained by analyzing and processing the preset indicator through a pre-trained machine learning model or a neural network model, or the credibility of the preset indicator can be obtained by identifying and judging the preset indicator through a professional user or a preset professional rule. After obtaining the credibility of the preset indicator, the credibility can be analyzed and compared with the preset threshold. When the credibility is greater than or equal to the preset threshold, it means that the credibility of the preset indicator is high, and then the preset indicator can be added with a trusted identifier, and the trusted identifier and the preset indicator are displayed in association, so that the user can increase the degree of trust in the analysis and processing of the preset indicator when calling the preset indicator for analysis and processing. Correspondingly, when the credibility is less than the preset threshold, it means that the credibility of the preset indicator is low. At this time, no identification operation can be performed on the preset indicator, so that the user can quickly identify the credible indicator and the ordinary indicator, further improving the stability and reliability of the analysis and processing based on the indicator.

[0084] The indicator establishment method provided in this embodiment obtains an indicator establishment request; determines the attribute information and atomic indicators used to establish the preset indicator based on the indicator establishment request, and then performs the indicator establishment operation based on the attribute information and the atomic indicators, so that the preset indicator corresponding to the indicator establishment request can be stably obtained. It should be noted that since the preset indicator is obtained by establishing atomic indicators corresponding to multiple layers of aggregation operators, when performing analysis and processing based on the preset indicator, the indicator results of the data granularity (keytype) and the indicator results of the overall granularity can be determined. The above-mentioned indicator results of the data granularity (keytype) and the indicator results of the overall granularity can be applicable to different application scenarios. For example, the indicator results of the data granularity (keytype) can be applicable to the application scenario of the control experiment, and the indicator results of the overall granularity can be applicable to the application scenario of the market analysis, thereby effectively expanding the scope of application corresponding to the method. This not only makes it convenient for users to manage data in different application scenarios, but also reduces the cost of data management, further improves the practicality of the method, and is conducive to market promotion and application.

[0085] Based on the above embodiments, Figure 3 As shown, this embodiment provides an implementation method for generating an atomic indicator for establishing a preset indicator based on a calculation method of an atomic indicator and a source table. At this time, generating an atomic indicator for establishing a preset indicator in this embodiment may include:

[0086] Step S301: Obtain the calculation method of the atomic indicator and the source table corresponding to the atomic indicator.

[0087] For atomic indicators, the atomic indicators in different application scenarios may correspond to different calculation methods and source tables. For example, in application scenario 1, the atomic indicator may correspond to the active table a, and then the source table a may be processed in a counting manner to determine the atomic indicator; in application scenario 2, the atomic indicator may correspond to the active table b, and the source table b may be processed in an average manner to determine the atomic indicator; in application scenario 3, the atomic indicator may correspond to the active table c, and the source table c may be processed in a ratio manner to determine the atomic indicator, and so on. Therefore, in order to ensure the accuracy and reliability of the determination of atomic indicators, the calculation method of the atomic indicators and the source table corresponding to the atomic indicators can be obtained. In some instances, the calculation method and the source table can be obtained through manual selection operations. At this time, obtaining the calculation method of the atomic indicator and the source table corresponding to the atomic indicator can include: obtaining multiple alternative calculation methods and multiple alternative source tables; obtaining the selection operation input by the user for any one of the alternative calculation methods, and determining the calculation method of the atomic indicator among multiple alternative calculation methods; obtaining the selection operation input by the user for any one of the alternative source tables, and determining the source table corresponding to the atomic indicator among multiple alternative source tables.

[0088] In other instances, the calculation method and the source table can not only be obtained through manual selection operations, but also can be determined through preset mapping relationships. At this time, obtaining the calculation method of the atomic indicator and the source table corresponding to the atomic indicator may include: obtaining a calculation mapping relationship for defining the atomic indicator and the calculation method and a source table mapping relationship for defining the atomic indicator and the source table; determining the calculation method corresponding to the atomic indicator based on the calculation mapping relationship; determining the source table corresponding to the atomic indicator based on the source-table mapping relationship, thereby effectively ensuring the accuracy and reliability of the determination of the calculation method and the source table.

[0089] Step S302: configure a multi-layer aggregation operator corresponding to the atomic index, and the multi-layer aggregation operator is used to determine the index result of the data granularity (keytype) and the index result of the overall granularity.

[0090] For atomic indicators, different application scenarios have different indicator processing granularities, for example, in the application scenario of a control experiment, it is necessary to perform analysis and processing operations based on the indicator results of the data granularity (keytype); in the application scenario of a non-control experiment (for example, the application scenario of a large-scale market analysis), it is necessary to perform analysis and processing operations based on the indicator results of the overall granularity. In order to be applicable to multiple different application scenarios, after determining the atomic indicator used to establish the preset indicator, a multi-layer aggregation operator corresponding to the atomic indicator can be configured. In some instances, the configuration operation of the multi-layer aggregation operator can be implemented through human-computer interaction operations. At this time, configuring the multi-layer aggregation operator corresponding to the atomic indicator may include: displaying a human-computer interaction interface for configuring the multi-layer aggregation operator; obtaining the configuration operation input by the user in the human-computer interaction interface; configuring the multi-layer aggregation operator corresponding to the atomic indicator based on the configuration operation, thereby effectively ensuring the stability and reliability of configuring the multi-layer aggregation operator.

[0091] In addition, for multi-layer aggregation operators, the multi-layer aggregation operators may include multiple layers of aggregation operators for analyzing and processing atomic indicators to obtain indicator results of different granularities, and the above-mentioned indicator results of different granularities may include indicator results of data granularity (keytype) and indicator results of overall granularity. In some instances, the multi-layer aggregation operator can be implemented as a two-layer aggregation operator. In this case, the two-layer aggregation operator may include a first-layer aggregation operator for determining the indicator results of data granularity (keytype) and a second-layer aggregation operator for determining the indicator results of overall granularity. Alternatively, the multi-layer aggregation operator can also be implemented as a three-layer aggregation operator. In this case, the multi-layer aggregation operator may include a first-layer aggregation operator, a second-layer aggregation operator and a third-layer aggregation operator. The above-mentioned first-layer aggregation operator and the second-layer aggregation operator are used to determine the indicator results of data granularity (keytype), and the third-layer aggregation operator is used to determine the indicator results of overall granularity.

[0092] Step S303: Generate atomic indicators based on the calculation method, source table and multi-layer aggregation operators.

[0093] After obtaining the calculation method, source table and multi-layer aggregation operator, the indicator generation operation can be performed based on the calculation method, source table and multi-layer aggregation operator to generate and obtain atomic indicators. In some instances, the atomic indicator generation operation can be implemented by a pre-trained machine learning model or neural network model. At this time, based on the calculation method, source table and multi-layer aggregation operator, generating atomic indicators can include: obtaining a pre-trained machine learning model or neural network model; inputting the calculation method, source table and multi-layer aggregation operator into the machine learning model or neural network model to obtain the atomic indicators output by the machine learning model or neural network model.

[0094] In other instances, not only can atomic indicators be generated through pre-trained machine learning models or neural network models, but in order to improve the quality and efficiency of atomic indicator establishment, atomic indicators can also be generated after filtering the source table. At this time, based on the calculation method, source table and multi-layer aggregation operators, establishing atomic indicators may include: obtaining filtering conditions for filtering data in the source table; filtering the source table based on the filtering conditions to obtain the filtered source table; and performing indicator establishment operations based on the calculation method, the filtered source table and the multi-layer aggregation operator to obtain atomic indicators.

[0095] Among them, since the source table includes thousands of various types of data, for example: the source table may include the user's identity, the user's behavior information, various types of indicators and attribute information corresponding to the indicators, etc., when configuring the indicators in a specific application scenario, the source table often includes a lot of data that is irrelevant to the indicator configuration operation. Therefore, after obtaining the source table, the source table can be filtered first. Specifically, the filtering conditions for filtering the data in the source table can be obtained. The above filtering conditions can be obtained through human-computer interaction operations. For filtering conditions, the filtering conditions corresponding to different application scenarios can be different. For example, the filtering conditions can include at least one of the following: keyword filtering conditions, application field filtering conditions, application status filtering conditions, attribute filtering conditions, etc.

[0096] After obtaining the filtering conditions for filtering the data in the source table, the source table can be filtered based on the filtering conditions to obtain the filtered source table, and then the indicator establishment operation can be performed based on the calculation method, the filtered source table and the multi-layer aggregation operator to complete the indicator establishment operation and obtain the atomic indicator.

[0097] In this embodiment, by obtaining the calculation method of the atomic indicator and the source table corresponding to the atomic indicator, configuring the multi-layer aggregation operator corresponding to the atomic indicator, and then generating the atomic indicator based on the calculation method, the source table and the multi-layer aggregation operator, the generation and establishment operation of the atomic indicator is realized, and the stable reliability of the establishment of the atomic indicator is ensured. Then, it is convenient to perform the establishment operation of the preset indicator based on the atomic indicator, further ensuring the stable reliability of the operation of the indicator establishment method.

[0098] Based on any of the above embodiments, Figure 4 As shown, the method in this embodiment can not only realize the index establishment operation, but also realize the control experiment operation. Specifically, the method in this embodiment can also include:

[0099] Step S401: Obtain a control experiment request.

[0100] When the user has a demand for a control experiment, the indicator establishment device can obtain a control experiment request. The control experiment request refers to request information for implementing an AB experiment or an A / B experiment. In some instances, the indicator establishment request can be obtained through a human-computer interaction operation. At this time, obtaining the control experiment request may include: displaying a human-computer interaction interface; obtaining an execution operation input by the user in the human-computer interaction interface; and obtaining a control experiment request based on the execution operation. The control experiment request may include control experiment information to be executed.

[0101] In other instances, the control experiment request can be obtained not only through human-computer interaction operations, but also through a client or a third device. In this case, obtaining the control experiment request may include: obtaining a client or a third device that is in communication with the indicator establishment device, wherein the control experiment request is generated or stored in the client or the third device; and then actively or passively obtaining the control experiment request through the client or the third device, thereby effectively ensuring the accuracy and reliability of obtaining the control experiment request.

[0102] Step S402: Based on the control experiment request, determine the associated index and traffic allocation information, where the traffic allocation information is used to determine the experimental user group and the control user group.

[0103] Since the control experiment needs to compare the processing results of the control group users and the experimental group users for a certain indicator, in order to be able to perform the control experiment operation stably, after obtaining the control experiment request, the related indicators and traffic allocation information can be determined based on the control experiment request. The above-mentioned traffic allocation information is used to determine the experimental user group and the control user group, wherein both the experimental user group and the control user group include information of multiple users participating in the control experiment, and the number of experimental user groups can be one or more, and the number of control user groups can also be one or more.

[0104] Specifically, with respect to associated indicators and traffic allocation information, they may be obtained based on user interaction operations. At this time, based on a control experiment request, determining associated indicators and traffic allocation information may include: displaying a human-computer interaction page based on the control experiment request; obtaining an execution operation input by the user in the human-computer interaction page, and determining associated indicators and traffic allocation information based on the execution operation.

[0105] In other instances, the associated indicators and traffic allocation information may be pre-stored in the indicator establishment device or in a third device. Specifically, the indicator establishment device or the third device may store the associated indicators and traffic allocation information corresponding to different scenarios. Therefore, after obtaining the control experiment request, the corresponding scenario information may be determined based on the control experiment request; the associated indicators and traffic allocation information are obtained in the indicator establishment device or the third device based on the scenario information, thereby effectively ensuring the stability and reliability of the determination of the associated indicators and traffic allocation information.

[0106] It should be noted that the specific method for determining the associated indicators and traffic allocation information is not limited to the above two implementation methods. Technical personnel in this field can flexibly select the method for determining the associated indicators and traffic allocation information according to the specific application scenarios or application requirements. As long as the associated indicators and traffic allocation information can be accurately determined, they will not be elaborated here.

[0107] In addition, after determining the traffic distribution information, since the traffic distribution information may be reasonable or unreasonable, in order to ensure the stable progress of the control experiment operation, the method in this embodiment may also include a technical solution for identifying whether the traffic distribution information is reasonable. At this time, the method in this embodiment may also include: obtaining the first traffic information corresponding to the control user group and the second traffic information corresponding to the experimental user group; determining the traffic difference between the first traffic information and the second traffic information; based on the traffic difference, determining whether the traffic distribution information is reasonable.

[0108] Specifically, after obtaining the traffic distribution information, the control user group and the experimental user group can be determined based on the traffic distribution information, and then the first traffic information corresponding to the control user group and the second traffic information corresponding to the experimental user group can be determined, and then the first traffic information and the second traffic information can be analyzed and compared to determine the traffic difference between the first traffic information and the second traffic information. Since the traffic difference can reflect whether the traffic distribution information is reasonable, after obtaining the traffic difference, the traffic difference can be analyzed and processed to determine whether the traffic distribution information is reasonable.

[0109] In some instances, based on the traffic difference, determining whether the traffic distribution information is reasonable may include: analyzing and comparing the traffic difference with a preset threshold value. When the traffic difference is less than or equal to the preset threshold value, it means that the traffic deviation between the control user group and the experimental user group is small, and thus the traffic distribution information can be determined to be reasonable; when the traffic difference is greater than the preset threshold value, it means that the traffic deviation between the control user group and the experimental user group is large, and thus the traffic distribution information can be determined to be unreasonable.

[0110] Furthermore, when it is determined that the flow distribution information is unreasonable, the method in this embodiment may also include: prohibiting the determination of the control experiment result corresponding to the control experiment request based on the associated index and the flow distribution information. In addition, in order to ensure the normal operation of the control experiment, prompt information corresponding to the flow distribution information may be generated to prompt the user to adjust the flow distribution information so as to obtain reasonable flow distribution information, and perform the control experiment operation based on the reasonable flow distribution information, thereby effectively ensuring the reasonable reliability of the determination of the flow distribution information.

[0111] Step S403: Determine the control experiment result corresponding to the control experiment request based on the correlation index and the traffic distribution information.

[0112] After obtaining the associated indicators and traffic allocation information, a control experiment operation can be performed based on the associated indicators and traffic allocation information, so that the control experiment result corresponding to the control experiment request can be determined. In some instances, the control experiment operation can be implemented by a pre-trained machine learning model or a neural network model. At this time, based on the associated indicators and traffic allocation information, determining the control experiment result corresponding to the control experiment request can include: obtaining a pre-trained machine learning model or a neural network model, inputting the associated indicators and traffic allocation information into the machine learning model or the neural network model, and obtaining the control experiment result output by the machine learning model or the neural network model.

[0113] In other instances, the control experiment can be implemented not only through a machine learning model or a neural network model, but also directly using an aggregation operator. In this case, based on the associated indicators and traffic distribution information, determining the control experiment results corresponding to the control experiment request may include: obtaining the first behavior log of each user in the control user group and the second behavior log of each user in the experimental user group; determining a multi-layer aggregation operator for analyzing and processing the first behavior log and the second behavior log; using the multi-layer aggregation operator to analyze and process the first behavior log and the second behavior log respectively, and determine the control group results and the experimental group results corresponding to the associated indicators; based on the control group results and the experimental group results, determining the control experiment results.

[0114] Specifically, after obtaining the traffic distribution information, the control user group and the experimental user group can be determined based on the traffic distribution information, and then the first behavior log of each user in the control user group and the second behavior log of each user in the experimental user group can be obtained, wherein the first behavior log includes the behavior information of the users in the control user group, and the second behavior log includes the behavior information of the users in the experimental user group. In order to implement the control experiment operation, after obtaining the first behavior log and the second behavior log, a multi-layer aggregation operator for analyzing and processing the first behavior log and the second behavior log can be determined, and the multi-layer aggregation operator can be implemented as a two-layer aggregation operator or a three-layer aggregation operator, etc., and then the first behavior log and the second behavior log can be analyzed and processed respectively by using the multi-layer aggregation, so as to obtain the control group result and the implementation group result corresponding to the associated indicators, wherein the control group result is obtained by analyzing and processing the associated indicators in the first behavior log by the multi-layer aggregation operator, and the experimental group result is obtained by analyzing and processing the associated indicators in the second behavior log by the multi-layer aggregation operator.

[0115] In some instances, when performing a control experiment, since the amount of data in the behavior log is large and the behavior log often includes some other data that is not related to the control experiment, in order to improve the quality and efficiency of the control experiment, the behavior log can be cleaned during the control experiment. Specifically, in this embodiment, the first behavior log and the second behavior log are analyzed and processed respectively using a multi-layer aggregation operator to determine the control group results and the experimental group results corresponding to the associated indicators, which may include: cleaning the first behavior log and the second behavior log respectively to obtain a first cleaned log and a second cleaned log; and then the first cleaned log and the second cleaned log can be processed respectively using a multi-layer aggregation operator to determine the control group results and the experimental group results corresponding to the associated indicators.

[0116] Specifically, after obtaining the first behavior log and the second behavior log, the first behavior log and the second behavior log can be cleaned respectively, and the first cleaned log and the second cleaned log can be obtained. Then, the first cleaned log and the second cleaned log can be processed respectively using a multi-layer aggregation operator, so that the control group results and the experimental group results corresponding to the associated indicators can be stably determined.

[0117] In addition, in the process of obtaining the control group results and the experimental group results, although the efficiency of obtaining the control group results and the experimental group results can be improved by cleaning the behavior logs, since the number of users in the control user group and the number of users in the experimental user group are often large, it is necessary to perform analysis and processing operations on the behavior logs of each user based on the associated indicators, and the quality and efficiency of the processing operations are low. Therefore, in order to further improve the quality and efficiency of obtaining the control group results and the experimental group results, the first cleaned log and the second cleaned log are processed respectively by multi-layer aggregation operators in this embodiment, and determining the control group results and the experimental group results corresponding to the associated indicators may include: performing covariance processing on the indicator data corresponding to the associated indicators in all the first cleaned logs and the indicator data corresponding to the associated indicators in all the second cleaned logs, respectively, to obtain the first indicator covariance corresponding to all the first cleaned logs and the second indicator covariance corresponding to all the second cleaned logs; using multi-layer aggregation operators to process the first indicator covariance and the second indicator covariance respectively, to determine the control group results and the experimental group results corresponding to the associated indicators.

[0118] Specifically, since there are a large number of first cleaned logs, the number of indicator data corresponding to the associated indicators in the first cleaned logs is also large. In order to improve the quality and efficiency of determining the control group results, after all first cleaned logs are obtained, the indicator data corresponding to the associated indicators in all first cleaned logs can be processed by covariance processing, so that the first indicator covariance corresponding to all first cleaned logs can be obtained. Similarly, after all second cleaned logs are obtained, the indicator data corresponding to the associated indicators in all second cleaned logs can be processed by covariance processing, so that the second indicator covariance corresponding to all second cleaned logs can be obtained.

[0119] After obtaining the first indicator covariance and the second indicator covariance, the first indicator covariance and the second indicator covariance can be processed respectively using multi-layer aggregation operators, so that the control group results and the experimental group results corresponding to the associated indicators can be determined quickly and stably, further improving the quality and efficiency of determining the control group results and the experimental group results.

[0120] After obtaining the control group results and the experimental group results, the control group results and the experimental group results can be analyzed and processed to determine the control experiment results. In some examples, the control experiment results can include the control group results and the experimental group results. In this case, the control group results and the experimental group results can be directly determined as the control experiment results. Alternatively, in other examples, the control experiment results can include the comparison results between the control group results and the experimental group results. In this case, after obtaining the control group results and the experimental group results, the control group results and the experimental group results can be analyzed and compared, and the comparison results can be determined as the control experiment results.

[0121] In other instances, after determining the control experiment results corresponding to the control experiment request, in order to improve the practicality of the method, the feasibility of the promotion plan can be judged based on the control experiment results. At this time, the method in this embodiment may also include: obtaining the promotion plan corresponding to the control experiment request; determining the feasibility of the promotion plan based on the control experiment results; when the feasibility is greater than or equal to a preset threshold, executing the promotion plan; when the feasibility is less than the preset threshold, generating prompt information for adjusting the promotion plan.

[0122] Specifically, a control experiment request is often generated based on a promotion plan to be analyzed. Therefore, after determining the control experiment result corresponding to the control experiment request, the promotion plan corresponding to the control experiment request can be obtained, and then the control experiment result can be analyzed and processed to determine the feasibility of the promotion plan. In some instances, the feasibility of the promotion plan can be obtained by analyzing and processing the control experiment results using a pre-trained machine learning model or a neural network model, or the feasibility of the promotion plan can be obtained by analyzing and processing the control experiment results using preset rules or preset algorithms.

[0123] After obtaining the feasibility of the promotion plan, the feasibility can be analyzed and compared with the preset threshold (the specific value can be flexibly configured or adjusted according to the application scenario). When the feasibility is greater than or equal to the preset threshold, it means that the feasibility of the promotion plan is high, and the experimental promotion effect of the promotion plan is good, and then the promotion plan can be executed. When the feasibility is less than the preset threshold, it means that the feasibility of the promotion plan is low, and the experimental promotion effect of the promotion plan is poor, and then the promotion operation of the promotion plan can be omitted, and prompt information for adjusting the promotion plan can be generated, so that the user can adjust the promotion plan in a timely and rapid manner based on the prompt information, further improving the practicality of the method.

[0124] In this embodiment, by obtaining a control experiment request, and then determining associated indicators and traffic allocation information based on the control experiment request, and determining the control experiment results corresponding to the control experiment request based on the associated indicators and traffic allocation information, the accuracy and reliability of the control experiment results are effectively guaranteed, thereby further improving the practicality of the method.

[0125] Based on the above embodiments, Figure 5 As shown, when performing a control experiment operation, since the flow distribution information is often obtained by system professionals or personnel with preset permissions through an indicator establishment device, not all users can obtain the flow distribution information conveniently and flexibly. At this time, in order to realize the control experiment operation, for users who cannot quickly and directly obtain the flow distribution information, the control experiment operation can be performed through a preset data table. Specifically, after obtaining the control experiment request, the method in this embodiment may also include:

[0126] Step S501: Detect whether the flow distribution information can be determined based on the control experiment request.

[0127] During the control experiment, if the user can obtain the traffic distribution information, the corresponding control experiment can be performed directly based on the traffic distribution information. For details, please refer to Figure 4 Description content in the illustrated embodiment. If the user cannot obtain the flow distribution information, it is necessary to perform the corresponding control experiment operation through the data table. Since whether the flow distribution information can be obtained directly determines the specific implementation method of the control experiment operation, after obtaining the control experiment request, it can be detected whether the flow distribution information can be determined based on the control experiment request. Specifically, detecting whether the flow distribution information can be determined based on the control experiment request can include: obtaining a plurality of pre-stored backup flow distribution information; detecting whether there is target flow distribution information that is relatively biased against the control experiment request in the plurality of backup flow distribution information, if so, it is determined that the flow distribution information can be determined based on the control experiment request; if not, it is determined that the flow distribution information cannot be determined based on the control experiment request, thereby effectively realizing the accurate detection operation of whether the flow distribution information can be determined based on the control experiment request.

[0128] Step S502: when the flow distribution information cannot be determined based on the control experiment request, a data table corresponding to the control experiment request is obtained.

[0129] When the test result shows that the flow distribution information can be determined based on the control experiment request, the control experiment operation can be performed directly based on the flow distribution information. For details, please refer to the above Figure 4The description of the illustrated embodiment is not repeated here. When the detection result is that the flow distribution information cannot be determined based on the control experiment request, the control experiment operation can be performed through the data table. At this time, the data table corresponding to the control experiment request can be obtained. In some instances, the data table can be stored in a preset area or a preset device. At this time, after obtaining the control experiment request, the preset area or preset device can be accessed based on the control experiment request to obtain the data table corresponding to the control experiment request. Alternatively, the data table can be obtained by transmitting through a preset interface. At this time, obtaining the data table corresponding to the control experiment request may include: displaying a preset communication interface for transmitting the data table; obtaining the data table corresponding to the control experiment request uploaded by the user through the preset communication interface (specifically, the storage address or network link address of the data table can be entered in the preset communication interface, etc.), thereby effectively ensuring the accuracy and reliability of obtaining the data table.

[0130] Step S503: Based on the data table, determine the experimental user group and the control user group.

[0131] After obtaining the data table, the experimental user group and the control user group can be determined based on the data table. In some instances, the experimental user group and the control user group can be determined by user tags. At this time, based on the data table, determining the experimental user group and the control user group may include: obtaining user tags for identifying users in the experimental user group or users in the control user group; distinguishing all users corresponding to the data table based on the user tags to obtain the experimental user group and the control user group, wherein the users in the experimental user group are suitable for the experimental data to be analyzed, and the users in the control user group are suitable for the original data.

[0132] For example, all the corresponding users in the data table may include user a, user b, user c, user d, and user e. The above user a, user c, and user d have corresponding labels "11", and the above user b and user e have corresponding labels "10". Then, the user labels used to identify the users in the experimental user group or the control user group can be determined. For example, when the user label used to identify the users in the experimental user group is "11", the above user a, user c, and user d can be determined as users in the experimental user group based on the above user label used to identify the users in the experimental user group, and then the other users (user b, user e) can be determined as users in the control user group. Alternatively, when the user label used to identify the users in the control user group is "10", the above user b and user e can be determined as users in the control user group based on the above user label used to identify the users in the control user group, and then the other users (user a, user c, user d) can be determined as users in the experimental user group, thereby effectively ensuring the accuracy and reliability of the determination of the experimental user group and the control user group.

[0133] Step S504: Determine a control experiment result corresponding to the control experiment request based on the experimental user group and the control user group.

[0134] After obtaining the experimental user group and the control user group, a control experiment operation can be performed based on the experimental user group and the control user group, so that the control experiment result corresponding to the control experiment request can be determined. In some instances, the control experiment result corresponding to the control experiment request can be obtained by analyzing and processing the experimental user group and the control user group through a pre-trained machine learning model or a neural network model. Or, in other instances, the control experiment operation can be realized by analyzing and processing the first data table corresponding to the control user group and the second data table corresponding to the experimental user group, respectively. At this time, based on the experimental user group and the control user group, determining the control experiment result corresponding to the control experiment request may include: obtaining a first data table corresponding to the users in the control user group and a second data table corresponding to the users in the experimental user group; determining the control group indicator result based on the first data table and the associated indicator; determining the experimental group indicator result based on the second data table and the associated indicator; determining the control experiment result based on the control group indicator result and the experimental group indicator result.

[0135] Specifically, after obtaining the experimental user group and the control user group, the first data table corresponding to the users in the control user group and the second data table corresponding to the user lines of the experimental user group can be obtained, wherein the number of users in the control user group is multiple, and the number of first data tables corresponding to the users is also multiple; the number of users in the experimental user group is multiple, and the number of second data tables corresponding to the users is also multiple. After obtaining the first data table and the second data table, the first data table and the associated indicators can be analyzed and processed. Specifically, all the first data tables and associated indicators can be analyzed and processed through a preset algorithm or a pre-trained machine learning model, so that the indicator results of the control group can be obtained. Similarly, after obtaining the second data table and the associated indicators, the second data table and the associated indicators can be analyzed and processed, so that the indicator results of the experimental group can be determined.

[0136] After obtaining the control group indicator results and the experimental group indicator results, the control group indicator results and the experimental group indicator results can be analyzed and processed to determine the control experiment results. In some instances, the control group indicator results and the experimental group indicator results can be directly determined as the control experiment results; or, the control group indicator results and the experimental group indicator results can be analyzed and compared, and the results of the analysis and comparison can be determined as the control experiment results, which effectively ensures the stability and reliability of the analysis and processing of the control experiment results.

[0137] In this embodiment, after obtaining the control experiment request, it is detected whether the traffic allocation information can be determined based on the control experiment request. When the traffic allocation information cannot be determined based on the control experiment request, a data table corresponding to the control experiment request is obtained, and then the experimental user group and the control user group are determined based on the data table, and the control experiment results corresponding to the control experiment request are determined based on the experimental user group and the control user group. This effectively achieves that when the user cannot determine the traffic allocation based on the control experiment request, the control experiment operation can be implemented through the data table, thereby ensuring the stability and reliability of the control experiment operation, and further improving the flexibility and reliability of the use of this method.

[0138] Based on any of the above embodiments, Figure 6 As shown, the indicator establishment device in this embodiment can not only realize the indicator definition operation, but also realize the indicator dashboard function. At this time, the method in this embodiment can also include:

[0139] Step S601: Get an indicator viewing request.

[0140] When a user has a need to view indicators, an indicator viewing request can be obtained, wherein the indicator viewing request includes name information or identification information of the indicator to be viewed, so that the user can perform corresponding viewing operations on the data corresponding to the indicator to be viewed based on the indicator viewing request.

[0141] In some instances, the indicator viewing request can be obtained through a human-computer interaction operation. In this case, obtaining the indicator viewing request may include: displaying a human-computer interaction interface, in which a plurality of candidate indicators that can be viewed are displayed; obtaining an execution operation input by a user for any candidate indicator in the human-computer interaction interface, and then determining the indicator viewing request based on the execution operation. For example, the human-computer interaction interface displays candidate indicators a, candidate indicator b, and candidate indicator c, and obtaining a selection operation input by a user for candidate indicator c; and generating an indicator viewing request corresponding to candidate indicator c based on the selection operation.

[0142] Step S602: Determine a data table corresponding to the indicator to be viewed based on the indicator viewing request.

[0143] After obtaining the indicator viewing request, in order to implement the indicator viewing operation and enable the user to view the relevant data of the indicator to be viewed, the data table corresponding to the indicator to be viewed can be determined based on the indicator viewing request. The data table includes various types of data such as attribute information corresponding to the indicator to be viewed, related indicators, indicator operation results, indicator call status, etc.

[0144] Step S603: Based on the data table, determine a data set corresponding to the indicator to be checked, where the data set includes the indicator type of the indicator to be checked and the indicator result corresponding to the indicator to be checked.

[0145] Since the data table includes multiple data such as attribute information and indicator data corresponding to the indicator to be viewed, in order to facilitate users to view the indicator data corresponding to the indicator to be viewed, after obtaining the data table, the data table can be analyzed and processed. Specifically, information extraction operations can be performed on the data table, so as to determine the data set corresponding to the indicator to be viewed, which can include the indicator type of the indicator to be viewed and the indicator result corresponding to the indicator to be viewed.

[0146] Step S604: Generate chart information corresponding to the indicator viewing request based on the data set.

[0147] Since the data set includes the indicator type of the indicator to be viewed and the indicator result corresponding to the indicator to be viewed, after obtaining the data set, chart information corresponding to the indicator viewing request can be generated based on the data set, wherein the chart information can be obtained by analyzing and processing the data set through a preset algorithm or a pre-trained machine learning model or neural network model, and the chart information can be implemented as at least one of the following: a table, a trend chart, a bar chart, etc. This not only facilitates the service provider to establish and generate chart information, but also enables users to intuitively display the indicator results corresponding to the indicator to be viewed, effectively improving the practicality of the method.

[0148] In addition, after generating chart information corresponding to the indicator viewing request based on the data set, the method in this embodiment can also perform an exception analysis operation on the data in the chart information. At this time, the method in this embodiment can also include: obtaining rule information for analyzing and processing the chart information; identifying whether there are any abnormalities in the chart information based on the rule information; when there are abnormalities in the chart information, determining the cause information that caused the abnormality; and displaying the chart information and the cause information in association.

[0149] Specifically, after obtaining the chart information, in order to perform an abnormal analysis operation on the data in the chart information, the rule information used for analyzing and processing the chart information can be obtained first, wherein the rule information can be a rule for identifying when the data is in an abnormal state; or the rule information can be a rule for identifying when the data is in a normal state. For the rule information, in some instances, it can be obtained through human-computer interaction operations, or the rule information can be information pre-stored in a preset area or preset device. After obtaining the chart information, the rule information used for analyzing and processing the chart information can be obtained by accessing the preset area or preset device. After obtaining the rule information used for analyzing and processing the chart information, it can be identified based on the rule information whether the chart information has an abnormal situation, that is, the rule information is used to analyze and process each data in the chart information, so that it can be identified whether the chart information has an abnormal situation.

[0150] When it is determined that there is no abnormality in the chart information, the chart information can be displayed normally. When there is an abnormality in the chart information, the cause information causing the abnormality can be determined. Specifically, the chart information can be analyzed and processed by a preset algorithm or a pre-trained machine learning model to determine the cause information causing the abnormality. After obtaining the cause information causing the abnormality, the chart information and the cause information can be displayed in association. This not only allows the user to quickly and timely understand the abnormality in the chart information, but also allows the user to quickly locate the abnormal position through the displayed cause information, which is convenient for the user to adjust the viewing indicators based on the cause information, further improving the practicality of the method.

[0151] In other instances, when the chart information and cause information are displayed in association, not only can the chart information and cause information be displayed directly in association, but they can also be displayed in combination with the associated indicators and associated events corresponding to the cause information. At this time, the associated display of the chart information and cause information may include: determining the associated indicators and associated events corresponding to the cause information; and associated displaying the chart information, cause information, associated indicators, and associated events.

[0152] Specifically, after determining the cause information causing the abnormal situation, since the indicator data causing the abnormal situation not only includes the indicator data itself, but may also include associated indicators and / or associated events that have a correlation with the indicator data, after obtaining the cause information, it is possible to obtain whether there are associated indicators and associated events corresponding to the cause information. When there are associated indicators and associated events corresponding to the cause information, the associated indicators and associated events corresponding to the cause information can be determined, and then the chart information, cause information, associated indicators and associated events can be displayed in association, so that the user can quickly view a more comprehensive range of causes that lead to the abnormal situation, further improving the practicality of the method.

[0153] In this embodiment, by obtaining an indicator viewing request, determining a data table corresponding to the indicator to be viewed based on the indicator viewing request, and then determining a data set corresponding to the indicator to be viewed based on the data table, and generating chart information corresponding to the indicator viewing request based on the data set, the generated chart information can help the user to quickly view the chart information, and can analyze and process the chart information, identify whether there are any abnormal changes in the icon information, and when there are abnormal changes, can perform attribution operations based on a variety of indicator abnormal fluctuation attribution methods such as result inference and automatic pruning of decision trees, thereby realizing the abnormal change attribution operation, and can provide the user with some adjustment directions or optimization directions based on the abnormal change attribution operation, further improving the practicality of the method.

[0154] Based on any of the above embodiments, Figure 7 As shown, the indicator establishment device in this embodiment can not only realize the indicator definition operation, but also realize the configuration operation of the indicator relationship diagram. Specifically, the method in this embodiment can also include:

[0155] Step S701: Obtain a configuration request for an indicator relationship graph.

[0156] Among them, for multiple indicators that have an associated relationship, the relationship between the multiple indicators can be reflected or identified through an indicator relationship diagram. Specifically, when the user has a configuration requirement for the indicator relationship diagram, the indicator establishment device can obtain a configuration request for the indicator relationship diagram. In some instances, the configuration request can be obtained through a human-computer interaction operation. At this time, obtaining the configuration request for the indicator relationship diagram may include: displaying a human-computer interaction interface; obtaining an execution operation input by the user in the human-computer interaction interface; and obtaining the configuration request for the indicator relationship diagram based on the execution operation, and the configuration request is used to implement the configuration operation of the indicator relationship diagram.

[0157] In other instances, the configuration request can be obtained not only through human-computer interaction operations, but also through a client or a third device. In this case, obtaining the configuration request for the indicator relationship graph may include: obtaining a client or a third device that is in communication with the indicator establishment device, wherein the configuration request for the indicator relationship graph is generated or stored in the client or the third device; and then actively or passively obtaining the configuration request for the indicator relationship graph through the client or the third device, thereby effectively ensuring the accuracy and reliability of obtaining the configuration request for the indicator relationship graph.

[0158] Step S702: Determine the independent variable index and the dependent variable index corresponding to the independent variable index based on the configuration request.

[0159] After obtaining the configuration request, the independent variable indicator and the dependent variable indicator corresponding to the independent variable indicator can be determined based on the configuration request, wherein the independent variable indicator can refer to the variable indicator that needs to be controlled or adjusted, and the dependent variable indicator refers to the variable indicator affected by the independent variable. For example, the independent variable indicator can be the number of user views, the number of user clicks, etc., and the dependent variable can be the transaction volume, order volume, etc. affected by the independent variable. In some instances, the independent variable indicator can be obtained through an indicator selection operation. At this time, determining the independent variable indicator based on the configuration request may include: displaying multiple pre-configured alternative indicators based on the configuration request; obtaining the selection operation input by the user for at least one indicator among the multiple alternative indicators; and determining at least one alternative indicator corresponding to the selection operation as the independent variable indicator. After obtaining the independent variable indicator, the dependent variable indicator corresponding to the independent variable indicator can be determined based on a preset algorithm or a preset mapping relationship.

[0160] In other instances, the independent variable indicators and the dependent variable indicators can be obtained not only through human-computer interaction operations, but also by analyzing and processing the configuration request through a pre-trained machine learning model or a neural network model. At this time, determining the independent variable indicators and the dependent variable indicators corresponding to the independent variable indicators based on the configuration request may include: obtaining a pre-trained machine learning model or a neural network model, inputting the configuration request into the machine learning model or the neural network model, and obtaining the independent variable indicators output by the machine learning model or the neural network model and the dependent variable indicators corresponding to the independent variable indicators.

[0161] Step S703: Based on the independent variable indicators and the dependent variable indicators, a causal relationship diagram for identifying the relationship between the independent variable indicators and the dependent variable indicators is generated.

[0162] After the independent variable indicators and the dependent variable indicators are obtained, the independent variable indicators and the dependent variable indicators can be analyzed and processed to generate a causal relationship diagram for identifying the relationship between the independent variable indicators and the dependent variable indicators. In some examples, since there is an association relationship between the independent variable indicators and the dependent variable indicators, after the independent variable indicators and the dependent variable indicators are obtained, the causal relationship diagram can be determined based on the association relationship between the independent variable indicators and the dependent variable indicators.

[0163] In other instances, not only can the causal relationship diagram be determined by the correlation between the independent variable indicators and the dependent variable indicators, but the causal relationship diagram can also be generated by a pre-trained machine learning model or a neural network model. At this time, based on the independent variable indicators and the dependent variable indicators, generating a causal relationship diagram for identifying the relationship between the independent variable indicators and the dependent variable indicators may include: obtaining a pre-trained machine learning model or a neural network model, inputting the independent variable indicators and the dependent variable indicators into the machine learning model or the neural network model, and obtaining the causal relationship diagram output by the machine learning model or the neural network model, thereby effectively ensuring the accuracy and reliability of the generation of the causal relationship diagram.

[0164] After generating a causal relationship diagram for identifying the relationship between the independent variable indicator and the dependent variable indicator, in order to improve the practicality of the method, the method in this embodiment may also include the ability to perform indicator adjustment operations based on the causal relationship diagram. Specifically, the method in this embodiment may also include: obtaining the indicator to be adjusted; determining the causal relationship diagram corresponding to the indicator to be adjusted; based on the causal relationship diagram, determining the dependent variable indicator corresponding to the indicator to be adjusted; adjusting the indicator to be adjusted and the dependent variable indicator based on the indicator adjustment request to obtain the adjusted indicator to be adjusted and the dependent variable indicator.

[0165] Specifically, when a user has an adjustment demand for a certain indicator to be adjusted, in order to stably implement the indicator adjustment operation, the indicator to be adjusted can be obtained, wherein the indicator to be adjusted can be obtained through human-computer interaction or a client storing the indicator to be adjusted. After obtaining the indicator to be adjusted, a causal relationship graph corresponding to the indicator to be adjusted can be determined. In some instances, the causal relationship graph can be determined by a preset mapping relationship. Since the causal relationship graph may include a dependent variable indicator that has an associated relationship with the indicator to be adjusted, in order to stably perform the indicator adjustment operation, after obtaining the causal relationship graph and the indicator to be adjusted, the dependent variable indicator corresponding to the indicator to be adjusted can be determined based on the causal relationship graph, and then the indicator to be adjusted and the dependent variable indicator can be adjusted based on the indicator adjustment request, and then the adjusted indicator to be adjusted and the dependent variable indicator can be obtained, thereby effectively implementing the indicator adjustment operation.

[0166] In this embodiment, a configuration request for an indicator relationship diagram is obtained, and then based on the configuration request, an independent variable indicator and a dependent variable indicator corresponding to the independent variable indicator are determined, and a causal relationship diagram for identifying the relationship between the independent variable indicator and the dependent variable indicator is generated based on the independent variable indicator and the dependent variable indicator. Then, the causal relationship diagram can be used to help users adjust and manage indicators, and when indicator adjustment or optimization operations are required, the user can be provided with adjustment directions or optimization directions, thereby facilitating users to perform stable indicator adjustments or indicator optimization operations, further improving the practicality of the method.

[0167] For specific applications, refer to the attached Figure 8 As shown, taking the three-layer aggregation operator as the multi-layer aggregation operator as an example, this application embodiment provides a data management platform that can realize functions such as data asset definition and management, control experiments (A / B experiments), and insight analysis, wherein data asset definition and management can be provided to data producers for use, and data producers can include at least one of the following: data R&D engineers, algorithm engineers, data scientists, business intelligence (Business intelligence, referred to as BI), etc., and control experiments and insight analysis can be provided to data consumers for use, and data consumers can include at least one of the following: product designers (Product Designer, referred to as PD), R&D engineers, data scientists, BI, etc. The specific implementation principle and process of the data management platform are explained below in conjunction with the accompanying drawings:

[0168] 1. Data asset definition and management unit

[0169] Among them, the data asset definition and management unit may include:

[0170] Associated asset module: used to establish an association relationship between the defined indicators / metrics and data tables (or source tables). The above data tables store user transaction data, user behavior data (which may include click behavior, transaction behavior, logistics information, etc.), buyer data, merchant data, traffic data, interaction & content data, etc.

[0171] Downstream consumption analysis module: used to record the calling information of indicators. For example, when the preset indicator is used in a control experiment, a calling record of the preset indicator can be generated; when the preset indicator is used by the preset user, a usage record of the preset indicator can be generated, and then the above-mentioned calling record and usage record can be displayed.

[0172] Asset authentication module: used to identify whether the defined indicator data or the stored indicator data is credible. If the indicator data is credible, a credible mark can be added to the indicator data; if the indicator data is not credible, no processing operation can be performed on the indicator data.

[0173] Caliber / dependency / version module: used to determine the caliber information of indicator data (used to identify the calculation method of the indicator, such as: average processing, sum processing, median processing, etc.), and each time the indicator data is called or modified, it can store and record the call record, modification record and modification version information of the indicator data.

[0174] Storage module: used to store the metadata of indicator data (owner information, indicator description information, category information).

[0175] Data exploration module: used to view the corresponding indicator data from different indicator dimensions to implement data exploration operations.

[0176] Asset retrieval & sorting module: used to implement indicator retrieval operations in multiple indicator data to display and obtain corresponding preset indicators; and can sort multiple indicator data according to preset dimensions, and then display the sorted multiple indicator data.

[0177] It should be noted that the data asset definition and management unit may not only include the modules listed above, but also include related modules for implementing other functions. Those skilled in the art may adjust the data asset definition and management unit according to specific application requirements or application scenarios, which will not be repeated here.

[0178] In order to realize the definition and management of data assets, the definition and management of the above data assets can be realized based on the "OneData methodology and KeyType model". Specifically, the OneData methodology and KeyType model can be used to standardize the definition of indicators and support the unified definition of experimental indicators and market indicators. For details, refer to the attached Fig. 9 As shown, the indicator definition operation may include the following steps:

[0179] Step 1: Start defining your metrics.

[0180] When a user has an indicator definition requirement, an indicator definition request may be obtained, and then the indicator definition operation may be started based on the indicator definition request.

[0181] Step 2: Select basic indicator information such as application field, indicator group, and indicator level.

[0182] Among them, in order to facilitate the indicator management operation, the basic information such as the application field, indicator group, indicator level and the like corresponding to the indicator to be established can be determined first, and the above-mentioned indicator level is used to identify the management degree, security degree or credibility degree, etc. of the indicator to be established. In addition, when the user cannot determine the application field, indicator group and indicator level information corresponding to the indicator to be established, the default application field, default indicator group and default indicator level information corresponding to the indicator to be established can be obtained, and then the default application field, default indicator group and default indicator level information are determined as the application field, indicator group and indicator level information corresponding to the indicator to be established.

[0183] Step 3: Determine the time frame corresponding to the indicator to be established.

[0184] Specifically, multiple time range information can be displayed for users to select, for example: 3 days, 1 day, 7 days, 15 days, 30 days, 60 days, etc. In order to accurately define the indicator, the selection operation input by the user for any time range information can be obtained, and then the time limit corresponding to the indicator to be established is determined based on the selection operation.

[0185] Step 4: Detect whether there is a measurement corresponding to the indicator to be established (ie, corresponding to the atomic indicator in the above embodiment), and the above measurement corresponds to the OneData methodology.

[0186] Specifically, a plurality of pre-configured candidate metrics may be obtained, and it may be detected whether there is a metric matching the indicator to be established among the plurality of candidate metrics. In some examples, the metric detection operation may be implemented by keyword matching.

[0187] Step 5: When there is no metric corresponding to the indicator to be established, a new metric creation operation can be performed. Specifically, basic information such as the input caliber is determined, and whether there is a source table corresponding to the metric is detected. If there is no source table corresponding to the metric, a source table corresponding to the metric is established based on the preset flow table creation process; if there is a source table corresponding to the metric, the filtering conditions corresponding to the source table can be selected, and the key Type type and the three-layer aggregation operator for implementing the indicator creation operation can be set, wherein the three-layer aggregation operator includes a first-layer aggregation operator and a second-layer aggregation operator for generating indicator results of the keytype granularity and a third-layer aggregation operator for generating indicator results of the overall granularity, and then the metric creation operation is performed based on the basic information such as the input caliber, the source table, the filtering conditions, and the three-layer aggregation operator to obtain the metric corresponding to the indicator to be established, and then the metric is reviewed. After the review is passed, the metric can be stored.

[0188] Step 6: When there is a measurement corresponding to the indicator to be established or the measurement corresponding to the indicator to be established has been established, it can be determined whether there is scenario-limited information corresponding to the indicator to be established (which may include: application scenarios on the application side, application scenarios on the personal computer side, application scenarios in multinational countries, etc.); if there is no scenario-limited information, then new scenario-limited information corresponding to the indicator to be established can be created based on the preset scenario-limited information flow; if there is scenario-limited information, then the indicator establishment operation can be performed based on the scenario-limited information, time limitation, measurement and indicator basic information to obtain the new indicator, and the established new indicator can be previewed and reviewed. When the established new indicator passes the review, the established new indicator can be stored.

[0189] In order to enable users to accurately identify the difference between the metric and the newly created indicator, the following example can be used to illustrate. For example, when the metric is the number of paying buyers and the time limit is three days, the newly created indicator can be the number of paying buyers within three days; when the metric is the number of completed orders and the time limit is seven days, the newly created indicator can be the number of completed orders within seven days. It should be noted that after the new indicator is created, in order to facilitate the indicator management operation, the label corresponding to the newly created indicator can be obtained. The label can include the user's attribute information, keyword information, etc. For example, the label can include the user's place of origin, height, geographical range, time range, etc.

[0190] The above-mentioned data asset definition and management unit can flexibly, clearly and quickly perform indicator definition and management operations. Specifically, when establishing or defining an indicator, the calculation method corresponding to the indicator can be determined. The calculation method may include at least one of the following: counting method, mean method, ratio and other definition methods, so as to meet the indicator definition operation in various application scenarios, and further improve the stability and reliability of the establishment of the indicator. In addition, after the indicator is established, the user can call or view the indicator according to the needs. The data asset definition and management unit can record and store the relevant operation records of the indicator and the upstream and downstream association information of the indicator, and can store and control the various versions of the indicator. In addition, the data asset definition and management unit can also support indicator retrieval and multi-dimensional screening, which further improves the flexibility and reliability of the use of this method.

[0191] 2. Controlled Experimental Unit

[0192] Among them, the control experiment unit may include an authority control module for managing and controlling indicator information according to preset authorities, a cleaning processing Trigger module for implementing data cleaning operations, a structured experimental report module for generating report data based on indicator results, a global incremental module, a multidimensional analysis module, an experimental dashboard module, a statistical engine module, a life cycle management module for managing the life cycle of newly created indicators, an experimental data calculation module for implementing control calculation operations, a diversion & tracking module, a traffic distribution module for implementing diversion operations, and the like.

[0193] Specifically, the control experiment unit can be applied to consumer application scenarios and can provide efficient and scientific end-to-end A / B experiment capabilities, thereby helping users to efficiently and scientifically measure the value of preset products. For details, please refer to the attached Fig.10 As shown, the control experiment operation may include the following steps:

[0194] Step 11: When the user has a need for a control experiment, the control experiment request can be obtained through the data management platform.

[0195] Specifically, the control experiment request may include data such as experiment configuration information, indicator configuration information, and experiment report to implement the control operation.

[0196] Step 12: Based on the control experiment request, determine the associated indicators and traffic allocation information, and the traffic allocation information is used to determine the experimental user group and the control user group.

[0197] Specifically, after obtaining the control experiment request, the associated indicators and traffic allocation information can be determined based on the control experiment request. The traffic allocation information is used to determine the experimental user group and the control user group, wherein the ratio between the users in the experimental user group and the users in the control user group needs to meet the preset requirements, for example: the number of users in the experimental user group can be 300 million, and the number of users in the control user group can be 250 million; or, the number of users in the experimental user group can be 100 million, and the number of users in the control user group can be 100 million; or, the number of users in the experimental user group can be 120 million, and the number of users in the control user group can be 100 million, that is, the ratio between the user traffic in the experimental user group and the user traffic in the control user group can be 1:1, 1.2:1, 1.1:1, etc., to achieve reasonable traffic allocation operations. In addition, the users in the above-mentioned experimental user group are applicable to the new promotion technical solution, and the users in the control user group are applicable to the original promotion technical solution.

[0198] It should be noted that when the traffic allocation operation achieved through the traffic allocation information is unreasonable, a prompt message corresponding to the traffic allocation information can be generated, and the control experiment operation based on the traffic allocation information can be prohibited; when the traffic allocation operation achieved through the traffic allocation information is reasonable, the control experiment operation based on the traffic allocation information can be allowed, thereby ensuring the stability and reliability of the control experiment operation.

[0199] Step 13: Based on the correlation indicator and the traffic distribution information, determine the control experiment result corresponding to the control experiment request.

[0200] After obtaining the associated indicators and traffic allocation information, the control experiment results corresponding to the control experiment request can be determined based on the associated indicators and traffic allocation information. Specifically, after obtaining the associated indicators and traffic allocation information, the above-mentioned associated indicators and traffic allocation information can be configured and distributed, that is, the associated indicators and traffic allocation information are sent to the software development tool kit SDK module for implementing the AB experiment, so that the SDK module can perform diversion operations based on the associated indicators and traffic allocation information, and further perform control experiment operations on users after the diversion operation.

[0201] Specifically, performing a control experiment operation based on associated indicators and traffic distribution information may include: obtaining a first behavior log (diversion log) of each user in a control user group and a second behavior log (user behavior log) of each user in an experimental user group. In order to further improve the quality and efficiency of data processing, the first behavior log and the second behavior log may be cleaned to obtain a first cleaned log and a second cleaned log; then a three-layer aggregation operator may be determined for analyzing and processing the first behavior log and the second behavior log; and the first cleaned log and the second cleaned log may be analyzed and processed using the three-layer aggregation operator, respectively, so as to determine the control group results and the experimental group results corresponding to the associated indicators; and the control experiment results corresponding to the control experiment request may be determined using the control group results and the experimental group results, wherein the control experiment results may be determined by a machine learning model and a causal inference model, and the above-mentioned machine learning model or causal extension model may be constructed through observational data.

[0202] In some instances, using a three-layer aggregation operator to analyze and process the first cleaned log and the second cleaned log respectively, and determining the control group results and the experimental group results corresponding to the associated indicators may include: performing covariance processing on the indicator data corresponding to the associated indicators in all the first cleaned logs and the indicator data corresponding to the associated indicators in all the second cleaned logs, respectively, to obtain the first indicator covariance corresponding to all the first cleaned logs and the second indicator covariance corresponding to all the second cleaned logs; using a three-layer aggregation operator to process the first indicator covariance and the second indicator covariance, respectively, to determine the control group results and the experimental group results corresponding to the associated indicators.

[0203] Step 14: Obtain a promotion plan corresponding to the control experiment request, determine the feasibility of the promotion plan based on the control experiment result, and control whether to execute the promotion plan based on the feasibility.

[0204] Among them, after obtaining the feasibility, the feasibility can be analyzed and compared with the preset threshold. When the feasibility is greater than or equal to the preset threshold, the promotion plan is executed; when the feasibility is less than the preset threshold, prompt information for adjusting the promotion plan is generated.

[0205] Step 12 ′: After obtaining the control experiment request, if the user cannot determine the flow distribution information based on the control experiment request, a data table corresponding to the control experiment request may be obtained.

[0206] Step 13': determining an experimental user group and a control user group based on the data table; and then determining a control experiment result corresponding to the control experiment request based on the experimental user group and the control user group.

[0207] When the user is unable to determine the traffic distribution information based on the control experiment request, the control experiment operation can be implemented through the data table. After obtaining the data table, the experimental user group and the control user group can be identified based on the user tags in the data table. Specifically, the user tags in the experimental user group are different from the user tags in the control user group; then the control experiment operation can be performed based on the experimental user group and the control user group, which effectively ensures the flexibility and reliability of the control experiment operation, and effectively implements the control experiment operation through the double difference technology, thereby effectively ensuring the quality and efficiency of the control experiment operation.

[0208] In some other instances, in order to ensure the stability and reliability of the control experiment operation, since there are a large number of users in the experimental user group and the control user group, there may be abnormal situations among the users in some scenarios. In order to ensure the accuracy and reliability of the control experiment, the behavior data in the experimental user group and the behavior data in the control user group can be subjected to outlier detection operations. When there are outliers in the behavior data in the experimental user group and / or the behavior data in the control user group, the abnormal behavior data in the experimental user group and / or the abnormal behavior data in the control user group are deleted, so that the behavior data in a normal state can be obtained. In this way, when the control experiment operation is performed based on the behavior data, the accuracy and reliability of the control experiment are effectively guaranteed.

[0209] In the process of conducting control experiments, the log data is processed by the pre-event variance reduction (CUPED) technology. Specifically, the minimum sample size required for the control experiment can be reduced based on the covariance corresponding to the log data, which effectively improves the quality and efficiency of the control experiment operation. In this way, the iteration efficiency of the control experiment can be improved by an average of 20% (up to 90%) on the preset indicators; in addition, the log information is cleaned by the double difference (PSM&DID) technology, and then the control experiment operation is carried out based on the cleaned log information. The control experiment results can be obtained by constructing a potential causal model using the observed data through machine learning and causal inference methods, and then the feasibility analysis of the promotion plan can be carried out based on the control experiment results. In addition, the model algorithm parameters are adaptively adjusted through data skewness to detect outliers and accurately eliminate noise; the quality and efficiency of the control experiment operation are further improved, so that the value of more products can be accurately evaluated, further improving the practicality of this method.

[0210] 3. Insight Analysis Unit

[0211] The insight analysis unit can realize rich data viewing and data analysis capabilities such as dashboard viewing, change analysis, cause-and-effect analysis, multi-dimensional analysis, and data exploration, helping users realize rich data viewing and data analysis capabilities. Specifically, the dashboard viewing function can include the following steps:

[0212] Step 21: Get the indicator viewing request.

[0213] Step 22: Determine the data table corresponding to the indicator to be viewed based on the indicator viewing request.

[0214] Step 23: Based on the data table, determine a data set corresponding to the indicator to be checked, where the data set includes the indicator type of the indicator to be checked and the indicator result corresponding to the indicator to be checked.

[0215] Step 24: Generate chart information corresponding to the indicator viewing request based on the data set.

[0216] After generating chart information corresponding to the indicator viewing request based on the data set, this embodiment may perform an abnormal analysis operation on the chart information. Specifically, the method in this embodiment may further include:

[0217] Step 25: Obtain rule information for analyzing and processing the chart information.

[0218] Step 26: Identify whether there is any abnormality in the chart information based on the rule information.

[0219] Step 27: When there is an abnormality in the chart information, determine the cause information causing the abnormality.

[0220] Step 28: Display the chart information and the cause information in association with each other.

[0221] Through the above operations, a data set can be quickly generated based on the indicators of the indicator center, helping users to efficiently build dashboards and helping data developers to improve development efficiency and quickly realize chart information. The chart information can include trend charts or tables. After obtaining the chart information, the chart information can be displayed so that users can quickly and intuitively view the corresponding indicator data through the chart information. In addition, after obtaining the chart information, the chart information displayed in the dashboard can be analyzed for abnormal changes. Specifically, the one-stop abnormal change attribution capability can be achieved through dimension drilling, related indicators, and related events, thereby effectively improving the quality and efficiency of abnormal change attribution operations. It can be specifically applicable to application scenarios such as household growth and advertising promotion, and can reduce the cost of abnormal change attribution operations, further improving the practicality of this method.

[0222] In some other examples, the data management platform in this embodiment can not only realize the function of viewing the dashboard and the function of analyzing changes, but also realize the function of exploring data. In this case, the method in this embodiment can include:

[0223] Step 31: Get the configuration request of the indicator relationship graph.

[0224] Step 32: Determine the independent variable indicator and the dependent variable indicator corresponding to the independent variable indicator based on the configuration request.

[0225] Step 33: Based on the independent variable indicators and the dependent variable indicators, a causal relationship diagram is generated for identifying the relationship between the independent variable indicators and the dependent variable indicators.

[0226] After generating a causal relationship diagram for identifying the relationship between the independent variable indicator and the dependent variable indicator, the method in this embodiment may also include: obtaining the indicator to be adjusted; determining the causal relationship diagram corresponding to the indicator to be adjusted; based on the causal relationship diagram, determining the dependent variable indicator corresponding to the indicator to be adjusted; adjusting the indicator to be adjusted and the dependent variable indicator based on the indicator adjustment request, and obtaining the adjusted indicator to be adjusted and the dependent variable indicator.

[0227] The above process realizes the configuration operation of the causal relationship diagram, that is, the user can flexibly configure the relationship between the independent variable and the dependent variable according to the needs. After the causal relationship diagram is generated, the causal relationship diagram can be used to help the user adjust the indicators. Specifically, when the user has the need to adjust or optimize a certain indicator data, it can help the user find the causal structure associated with the target, and provide a reference direction for the indicator optimization or adjustment operation.

[0228] The data management platform provided by this application embodiment integrates insight analysis functions, control experiment functions and data asset management functions, which can avoid the duplication of data asset construction, inconsistency in the caliber of the market and latency indicators, difficulty in finding assets, fear of using them after finding them, data differences, and reduced credibility of analysis and experimental results. It can provide a unified and reliable indicator system, and when performing data processing operations based on indicators, it can avoid problems such as manual running of experimental reports, lack of scientific assurance of experimental processes, and high insight analysis costs, effectively ensuring the quality and efficiency of AB experimental operations and insight analysis. Specifically, the data management platform has the following advantages:

[0229] (1) The data asset management unit can perform indicator management operations with preset indicators as the center, realize the configuration of indicators and the three-layer aggregation computing framework, and is compatible with indicator application scenarios such as Kanban experiments. It can improve the efficiency of indicator development and help lower the threshold for using data for all product functions.

[0230] (2) The control experiment unit provides efficient and scientific end-to-end A / B experiment capabilities to help users measure project value efficiently and scientifically. Specifically, it can obtain users in the control group and the experimental group. Users in the control group or the experimental group will generate behavioral data, and then calculate the experimental indicator data and the control indicator data through an automated data link. After the experiment has run for a period of time, the indicator data can be viewed based on the experimental report to implement the control experiment operation, so as to take the corresponding promotion plan offline or conduct promotion testing operations.

[0231] (3) The insight analysis unit provides insight analysis products such as dashboards / data sets, anomaly analysis, and causal analysis to help users achieve rich data insight and analysis capabilities. Among them, the technical innovation is the proposed method of attributing abnormal fluctuations of multiple indicators based on time series causal inference and automatic pruning of decision trees.

[0232] In addition, the data management platform in this embodiment may have the following features: a unified application programming interface (API) for different application scenarios, a unified data link for different application scenarios (which may include a data link for realizing indicator calculation, a data link for realizing data synchronization operation, a data link for realizing data refresh operation, etc.), and the data links corresponding to different application scenarios may be unified and reused, which can avoid the emergence of a chimney-type network architecture; in addition, the data management platform in this embodiment can be communicated with the control center. When the control center is able to implement a technology for a preset function, the data management platform will not repeat the construction of the technology already existing in the control center. It can reuse the preset technology that can be implemented in the control center. This can not only meet the complex and sophisticated consumer needs in different application scenarios, but also improve the flexibility and reliability of the use of the data management platform, further improve the practicality of the method, and is conducive to market promotion and application.

[0233] Reference Fig.11 As shown, this embodiment provides an indicator establishment device, which is used to perform the above Figure 2 The indicator establishment method shown, specifically, the indicator establishment device may include:

[0234] The first acquisition module 11 is used to obtain an indicator establishment request;

[0235] A first determination module 12 is used to determine attribute information and atomic indicators used to establish a preset indicator based on an indicator establishment request, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity;

[0236] The first processing module 13 is used to perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0237] In some instances, when the first determination module 12 determines an atomic indicator for establishing a preset indicator based on an indicator establishment request, the first determination module 12 is used to execute: based on the indicator establishment request, obtaining multiple established atomic indicators; identifying whether there is a target atomic indicator corresponding to the preset indicator among the multiple established atomic indicators; if so, determining the target atomic indicator as the atomic indicator corresponding to the preset indicator; if not, generating an atomic indicator for establishing the preset indicator.

[0238] In some instances, when the first determination module 12 generates an atomic indicator for establishing a preset indicator, the first determination module 12 is used to execute: obtaining a calculation method for the atomic indicator and a source table corresponding to the atomic indicator; configuring a multi-layer aggregation operator corresponding to the atomic indicator, the multi-layer aggregation operator being used to determine the indicator results of the data granularity and the indicator results of the overall granularity; generating an atomic indicator based on the calculation method, the source table and the multi-layer aggregation operator.

[0239] In some instances, the multi-layer aggregation operator includes a first-layer aggregation operator, a second-layer aggregation operator, and a third-layer aggregation operator. The first-layer aggregation operator and the second-layer aggregation operator are used to determine the indicator results of the data granularity, and the third-layer aggregation operator is used to determine the indicator results of the overall granularity.

[0240] In some instances, when the first determination module 12 establishes an atomic indicator based on the calculation method, the source table, and the multi-layer aggregation operator, the first determination module 12 is used to execute: obtaining a filtering condition for filtering data in the source table; filtering the source table based on the filtering condition to obtain a filtered source table; performing an indicator establishment operation based on the calculation method, the filtered source table, and the multi-layer aggregation operator to obtain an atomic indicator.

[0241] In some instances, when the first processing module 13 performs an indicator establishment operation based on attribute information and atomic indicators to obtain a preset indicator corresponding to the indicator establishment request, the first processing module 13 is used to execute: obtaining a scenario limitation condition corresponding to the preset indicator; performing an indicator establishment operation based on the attribute information, the scenario limitation condition and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

[0242] In some examples, after obtaining the preset indicator corresponding to the indicator establishment request, the first acquisition module 11 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0243] The first acquisition module 11 is used to acquire an indicator processing operation corresponding to a preset indicator, and the indicator processing information includes at least one of the following: an indicator editing operation, an indicator calling operation, and an indicator modifying operation;

[0244] The first processing module 13 is used to process the preset indicator based on the indicator processing operation to obtain a processing result corresponding to the indicator processing operation; and associate the processing result with the indicator processing operation for storage.

[0245] In some examples, after obtaining the preset indicator corresponding to the indicator establishment request, the first acquisition module 11 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0246] A first acquisition module 11 is used to acquire the credibility of a preset indicator;

[0247] The first processing module 13 is used to add a credible mark to the preset indicator when the credibility is greater than or equal to the preset threshold; and associate the credible mark with the preset indicator for display.

[0248] In some examples, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0249] A first acquisition module 11 is used to acquire a control experiment request;

[0250] A first determination module 12 is used to determine the associated index and traffic allocation information based on the control experiment request, and the traffic allocation information is used to determine the experimental user group and the control user group;

[0251] The first processing module 13 is used to determine the control experiment result corresponding to the control experiment request based on the correlation index and the traffic distribution information.

[0252] In some instances, when the first processing module 13 determines the control experiment results corresponding to the control experiment request based on the associated indicators and the traffic distribution information, the first processing module 13 is used to execute: obtaining the first behavior log of each user in the control user group and the second behavior log of each user in the experimental user group; determining a multi-layer aggregation operator for analyzing and processing the first behavior log and the second behavior log; using the multi-layer aggregation operator to analyze and process the first behavior log and the second behavior log respectively to determine the control group results and the experimental group results corresponding to the associated indicators; and determining the control experiment results based on the control group results and the experimental group results.

[0253] In some instances, when the first processing module 13 uses a multi-layer aggregation operator to analyze and process the first behavior log and the second behavior log respectively to determine the control group results and the experimental group results corresponding to the associated indicators, the first processing module 13 is used to execute: cleaning the first behavior log and the second behavior log respectively to obtain a first cleaned log and a second cleaned log; using a multi-layer aggregation operator to process the first cleaned log and the second cleaned log respectively to determine the control group results and the experimental group results corresponding to the associated indicators.

[0254] In some instances, when the first processing module 13 uses a multi-layer aggregation operator to process the first cleaned log and the second cleaned log respectively to determine the control group results and the experimental group results corresponding to the associated indicators, the first processing module 13 is used to execute: performing covariance processing on the indicator data corresponding to the associated indicators in all the first cleaned logs and the indicator data corresponding to the associated indicators in all the second cleaned logs, respectively, to obtain the first indicator covariance corresponding to all the first cleaned logs and the second indicator covariance corresponding to all the second cleaned logs; using a multi-layer aggregation operator to process the first indicator covariance and the second indicator covariance respectively, to determine the control group results and the experimental group results corresponding to the associated indicators.

[0255] In some examples, after determining the control experiment result corresponding to the control experiment request, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are used to perform the following steps:

[0256] A first acquisition module 11 is used to acquire a promotion plan corresponding to the control experiment request;

[0257] A first acquisition module 12 is used to determine the feasibility of the promotion plan based on the control experiment results;

[0258] The first processing module 13 is used to execute the promotion plan when the feasibility is greater than or equal to the preset threshold; when the feasibility is less than the preset threshold, generate prompt information for adjusting the promotion plan.

[0259] In some instances, after obtaining a control experiment request, the first processing module 13 in this embodiment is used to perform the following steps: detecting whether the traffic allocation information can be determined based on the control experiment request; when the traffic allocation information cannot be determined based on the control experiment request, obtaining a data table corresponding to the control experiment request; based on the data table, determining the experimental user group and the control user group; based on the experimental user group and the control user group, determining the control experiment results corresponding to the control experiment request.

[0260] In some instances, when the first processing module 13 determines a control experiment result corresponding to a control experiment request based on the experimental user group and the control user group, the first processing module 13 is used to execute: obtaining a first data table corresponding to users in the control user group and a second data table corresponding to users in the experimental user group; determining a control group indicator result based on the first data table and associated indicators; determining an experimental group indicator result based on the second data table and associated indicators; and determining a control experiment result based on the control group indicator result and the experimental group indicator result.

[0261] In some examples, after determining the flow distribution information, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are used to perform the following steps:

[0262] A first acquisition module 11 is used to acquire first flow information corresponding to a control user group and second flow information corresponding to an experimental user group;

[0263] A first determining module 12, configured to determine a flow difference between the first flow information and the second flow information;

[0264] The first processing module 13 is used to determine whether the flow distribution information is reasonable based on the flow difference.

[0265] In some examples, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0266] The first acquisition module 11 is used to obtain an indicator viewing request;

[0267] A first determining module 12, configured to determine a data table corresponding to the indicator to be viewed based on the indicator viewing request;

[0268] The first processing module 13 is used to determine a data set corresponding to the indicator to be viewed based on the data table, wherein the data set includes the indicator type of the indicator to be viewed and the indicator result corresponding to the indicator to be viewed; and generate chart information corresponding to the indicator viewing request based on the data set.

[0269] In some examples, after generating chart information corresponding to the indicator viewing request based on the data set, the first acquisition module 11 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0270] A first acquisition module 11 is used to acquire rule information for analyzing and processing chart information;

[0271] The first processing module 13 is used to identify whether there is an abnormality in the chart information based on the rule information; when there is an abnormality in the chart information, determine the cause information causing the abnormality; and associate the chart information with the cause information for display.

[0272] In some examples, when the first processing module 13 displays the chart information and the cause information in an associated manner, the first processing module 13 is used to execute: determining associated indicators and associated events corresponding to the cause information; and displaying the chart information, the cause information, the associated indicators, and the associated events in an associated manner.

[0273] In some examples, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0274] A first acquisition module 11 is used to obtain a configuration request of an indicator relationship graph;

[0275] A first determination module 12, configured to determine an independent variable indicator and a dependent variable indicator corresponding to the independent variable indicator based on a configuration request;

[0276] The first processing module 13 is used to generate a causal relationship diagram for identifying the relationship between the independent variable indicators and the dependent variable indicators based on the independent variable indicators and the dependent variable indicators.

[0277] In some examples, after generating a causal relationship diagram for identifying the relationship between the independent variable indicator and the dependent variable indicator, the first acquisition module 11, the first determination module 12 and the first processing module 13 in this embodiment are respectively used to perform the following steps:

[0278] A first acquisition module 11 is used to acquire the indicator to be adjusted;

[0279] The first determination module 12 is used to determine a cause-effect relationship diagram corresponding to the indicator to be adjusted; based on the cause-effect relationship diagram, determine a dependent variable indicator corresponding to the indicator to be adjusted;

[0280] The first processing module 13 is further used to adjust the indicator to be adjusted and the dependent variable indicator based on the indicator adjustment request to obtain the adjusted indicator to be adjusted and the dependent variable indicator.

[0281] Fig.11 The device shown can perform Figure 1-Figure 10 For the method of the embodiment shown in the figure, the part not described in detail in this embodiment can be referred to Figure 1-Figure 10 The implementation process and technical effects of this technical solution refer to Figure 1-Figure 10 The description in the illustrated embodiment will not be repeated here.

[0282] In one possible design, Fig.11The structure of the indicator establishment device shown can be implemented as an electronic device, which can be a controller, a personal computer, a server, or other devices. Fig.12 As shown, the electronic device may include: a first processor 21 and a first memory 22. The first memory 22 is used to store the corresponding electronic device to execute the above Figure 1-Figure 10 The program of the indicator establishment method provided in the illustrated embodiment, the first processor 21 is configured to execute the program stored in the first memory 22 .

[0283] The program includes one or more computer instructions, wherein the one or more computer instructions can implement the following steps when executed by the first processor 21: obtaining an indicator establishment request; based on the indicator establishment request, determining attribute information and atomic indicators used to establish preset indicators, wherein the attribute information includes at least the time range of the preset indicators, wherein the atomic indicators correspond to multiple layers of aggregation operators, and the multiple layers of aggregation operators are used to determine the indicator results of the data granularity and the indicator results of the overall granularity; performing indicator establishment operations based on the attribute information and the atomic indicators to obtain the preset indicators corresponding to the indicator establishment request.

[0284] Furthermore, the first processor 21 is also used to execute the aforementioned Figure 1-Figure 10 All or part of the steps in the illustrated embodiment. The structure of the electronic device may further include a first communication interface 23, which is used for the electronic device to communicate with other devices or a communication network.

[0285] In addition, an embodiment of the present invention provides a computer storage medium for storing computer software instructions used by electronic devices, which includes instructions for executing the above Figure 1-Figure 10 The procedures involved in the indicator establishment method in the illustrated embodiment.

[0286] In addition, an embodiment of the present invention provides a computer program product, including: a computer-readable storage medium storing computer instructions, when the computer instructions are executed by one or more processors, causing the one or more processors to execute the above Figure 1-Figure 10 The steps in the indicator establishment method in the method embodiment are shown.

[0287] Reference Fig.13 As shown, this embodiment provides a data management system, which may include:

[0288] The indicator definition unit 31 is used to obtain an indicator establishment request; based on the indicator establishment request, determine the attribute information and atomic indicators used to establish the preset indicator, the attribute information at least includes the time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator results of the data granularity and the indicator results of the overall granularity; perform the indicator establishment operation based on the attribute information and the atomic indicator to obtain the preset indicator corresponding to the indicator establishment request.

[0289] In other examples, the data management system in this embodiment can not only implement the indicator definition operation through the indicator definition unit 31, but also implement the control experiment operation. In this case, the data management system in this embodiment can also include:

[0290] The control experiment unit 32 is connected in communication with the indicator definition unit 31, and is used to obtain the promotion plan to be processed; determine the associated indicators and traffic allocation information corresponding to the promotion plan, and the traffic allocation information is used to determine the first user group applicable to the promotion plan and the second user group not applicable to the promotion plan; based on the associated indicators and the traffic allocation information, determine the feasibility of the promotion plan.

[0291] In some other examples, the data management system in this embodiment can not only implement the indicator definition operation through the indicator definition unit 31, implement the control experiment operation through the control experiment unit 32, but also implement the insight analysis operation. In this case, the data management system in this embodiment can also include:

[0292] The insight analysis unit 33 is communicated with the indicator definition unit 31 and the control experiment unit 32, and is used to obtain an indicator viewing request; determine a data table corresponding to the indicator to be viewed based on the indicator viewing request; determine a data set corresponding to the indicator to be viewed based on the data table, the data set including the indicator type of the indicator to be viewed and the indicator result corresponding to the indicator to be viewed; generate chart information corresponding to the indicator viewing request based on the data set.

[0293] It should be noted that the indicator definition unit 31, the control experiment unit 32 and the insight analysis unit 33 in the data management system of this embodiment can also perform the above Figure 1-Figure 10 The other method steps shown are Fig.13 The system shown can perform Figure 1-Figure 10 For the method of the embodiment shown in the figure, the part not described in detail in this embodiment can be referred to Figure 1-Figure 10 The implementation process and technical effects of this technical solution refer to Figure 1-Figure 10 The description in the illustrated embodiment will not be repeated here.

[0294] 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, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0295] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by adding a necessary general hardware platform, and of course can also be implemented by combining hardware and software. Based on such an understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a computer product, and the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0296] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable device to produce a machine, so that the instructions executed by the processor of the computer or other programmable device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0297] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable device so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide the functions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1The steps of the functions specified in the blocks or multiple blocks. In a typical configuration, the computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0298] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0299] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for establishing an indicator, characterized in that: include: Get indicator creation request; Based on the indicator establishment request, determine attribute information and atomic indicators used to establish the preset indicator, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity; An indicator establishment operation is performed based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

2. The method according to claim 1, characterized in that Based on the indicator establishment request, determining an atomic indicator for establishing a preset indicator includes: Based on the indicator establishment request, obtain multiple established atomic indicators; Identify whether there is a target atomic indicator corresponding to the preset indicator among the multiple established atomic indicators; If so, determining the target atomic indicator as the atomic indicator corresponding to the preset indicator; If it does not exist, an atomic pointer is generated to establish the preset pointer.

3. The method according to claim 2, characterized in that Generates atomic indicators used to build preset indicators, including: Obtaining a calculation method for the atomic indicator and a source table corresponding to the atomic indicator; Configuring a multi-layer aggregation operator corresponding to the atomic indicator, wherein the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity; The atomic indicator is generated based on the calculation method, the source table and the multi-layer aggregation operator.

4. The method according to claim 3, characterized in that Based on the calculation method, the source table and the multi-layer aggregation operator, the atomic indicator is established, including: Obtaining a filtering condition for filtering the data in the source table; Filter the source table based on the filtering condition to obtain a filtered source table; An indicator establishment operation is performed based on the calculation method, the filtered source table and the multi-layer aggregation operator to obtain the atomic indicator.

5. The method according to any one of claims 1 to 4, characterized in that: Performing an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request includes: Acquire the scene limiting condition corresponding to the preset indicator; An indicator establishment operation is performed based on the attribute information, the scenario limiting condition and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

6. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Get control experiment request; Based on the control experiment request, determining related indicators and traffic allocation information, wherein the traffic allocation information is used to determine an experimental user group and a control user group; Based on the correlation indicator and the traffic distribution information, a control experiment result corresponding to the control experiment request is determined.

7. The method according to claim 6, characterized in that Determining a control experiment result corresponding to the control experiment request based on the correlation indicator and the traffic distribution information includes: Obtaining a first behavior log of each user in the control user group and a second behavior log of each user in the experimental user group; Determine a multi-layer aggregation operator for analyzing and processing the first behavior log and the second behavior log; Using the multi-layer aggregation operator to analyze and process the first behavior log and the second behavior log respectively, and determine the control group result and the experimental group result corresponding to the correlation index; The control experiment results are determined based on the control group results and the experimental group results.

8. The method according to claim 7, characterized in that The first behavior log and the second behavior log are analyzed and processed respectively by using the multi-layer aggregation operator to determine the control group result and the experimental group result corresponding to the correlation index, including: Cleaning the first behavior log and the second behavior log respectively to obtain a first cleaned log and a second cleaned log; The first cleaned log and the second cleaned log are processed respectively by using the multi-layer aggregation operator to determine the control group result and the experimental group result corresponding to the associated index.

9. The method according to claim 8, characterized in that Using the multi-layer aggregation operator to process the first cleaned log and the second cleaned log respectively to determine the control group result and the experimental group result corresponding to the correlation index, including: Respectively perform covariance processing on the indicator data corresponding to the associated indicator in all first cleaned logs and the indicator data corresponding to the associated indicator in all second cleaned logs to obtain the first indicator covariance corresponding to all first cleaned logs and the second indicator covariance corresponding to all second cleaned logs; The first indicator covariance and the second indicator covariance are processed respectively by using the multi-layer aggregation operator to determine the control group result and the experimental group result corresponding to the associated indicator.

10. The method according to claim 6, characterized in that After determining the control experiment result corresponding to the control experiment request, the method further includes: Obtaining a promotion plan corresponding to the control experiment request; Based on the control experiment results, determine the feasibility of the promotion plan; When the feasibility is greater than or equal to a preset threshold, the promotion plan is executed; When the feasibility is less than a preset threshold, prompt information for adjusting the promotion plan is generated.

11. The method according to claim 6, characterized in that After obtaining the control experiment request, the method further includes: detecting whether flow distribution information can be determined based on the control experiment request; When the flow distribution information cannot be determined based on the control experiment request, obtaining a data table corresponding to the control experiment request; Based on the data table, determining an experimental user group and a control user group; Based on the experimental user group and the control user group, a control experiment result corresponding to the control experiment request is determined.

12. The method according to claim 11, characterized in that Determining a control experiment result corresponding to the control experiment request based on the experimental user group and the control user group includes: Acquire a first data table corresponding to users in the control user group and a second data table corresponding to users in the experimental user group; Determine the index result of the control group based on the first data table and the associated index; Determine the indicator result of the experimental group based on the second data table and the associated indicator; The control experiment results are determined based on the indicator results of the control group and the indicator results of the experimental group.

13. The method according to claim 6, characterized in that After determining the flow distribution information, the method further includes: Acquire first flow information corresponding to the control user group and second flow information corresponding to the experimental user group; determining a flow rate difference between the first flow rate information and the second flow rate information; Based on the flow difference, it is determined whether the flow distribution information is reasonable.

14. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Get indicator viewing request; Determine a data table corresponding to the indicator to be viewed based on the indicator viewing request; Based on the data table, determining a data set corresponding to the indicator to be checked, wherein the data set includes an indicator type of the indicator to be checked and an indicator result corresponding to the indicator to be checked; Generate chart information corresponding to the indicator viewing request based on the data set.

15. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Get the configuration request of the indicator relationship graph; Determine an independent variable indicator and a dependent variable indicator corresponding to the independent variable indicator based on the configuration request; Based on the independent variable indicators and the dependent variable indicators, a causal relationship diagram for identifying the relationship between the independent variable indicators and the dependent variable indicators is generated.

16. An indicator establishment device, characterized in that: include: A first acquisition module is used to obtain an indicator establishment request; A first determination module is used to determine attribute information and atomic indicators used to establish a preset indicator based on the indicator establishment request, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine the indicator result of the data granularity and the indicator result of the overall granularity; The first processing module is used to perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request.

17. A data management system, characterized in that: include: An indicator definition unit, used to obtain an indicator establishment request; Based on the indicator establishment request, determine attribute information and atomic indicators for establishing a preset indicator, wherein the attribute information at least includes a time range of the preset indicator, wherein the atomic indicator corresponds to a multi-layer aggregation operator, and the multi-layer aggregation operator is used to determine an indicator result of a data granularity and an indicator result of an overall granularity; perform an indicator establishment operation based on the attribute information and the atomic indicator to obtain a preset indicator corresponding to the indicator establishment request; A control experiment unit is communicatively connected with the indicator definition unit, and is used to obtain a promotion plan to be processed; determine associated indicators and traffic allocation information corresponding to the promotion plan, and the traffic allocation information is used to determine a first user group applicable to the promotion plan and a second user group not applicable to the promotion plan; based on the associated indicators and the traffic allocation information, determine the feasibility of the promotion plan.

18. An electronic device, characterized in that: include: A memory, a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the method as described in any one of claims 1-16.

Images