Data processing method, device, computer equipment and storage medium

Abstract

The present application discloses a data processing method, apparatus, computer equipment, and storage medium. The method includes: obtaining target user data of multiple users, wherein the target user data of any user includes: indicator data of any user under each indicator; analyzing and quantifying the causal relationship between each indicator in M ​​indicators based on the indicator data of each user under each indicator; generating first metadata based on the analysis and quantification results; when it is necessary to display the indicator causal graph of the target object in the front-end application, converting the first metadata into second metadata; displaying the indicator causal graph based on the second metadata in the user interface displayed by the front-end application; the indicator causal graph includes at least M nodes and N directed edges, where one node records one indicator, and one directed edge is used to connect two nodes corresponding to a trusted indicator pair. By displaying the indicator causal graph, the present application can facilitate technical personnel to search for indicators and improve the efficiency and accuracy of indicator search.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, specifically to the field of computer technology, and in particular to a data processing method, a data processing device, a computer device, and a computer storage medium. Background Art

[0002] With the continuous development of Internet technology, more and more objects (such as smart devices and apps (clients)) are appearing in people's daily lives. Each object is usually configured with multiple indicators, so relevant technical personnel can improve the operating performance of each object by optimizing all or some of the indicators configured for each object, thereby increasing user stickiness for each object.

[0003] Currently, when optimizing the target indicators of a target object, technicians typically need to combine related indicators with those indicators to achieve optimization. Research has shown that technicians typically find related indicators by manually observing the relationships between the various indicators. This approach, on the one hand, consumes a significant amount of time for technicians to manually observe the relationships between the various indicators, resulting in low efficiency in indicator search. On the other hand, the technicians' subjective judgment can affect the accuracy of the observed relationships between the various indicators, thereby affecting the accuracy of indicator search. Therefore, how to facilitate indicator search and improve the efficiency and accuracy of indicator search has become a research hotspot. Summary of the Invention

[0004] The embodiments of the present application provide a data processing method, apparatus, computer equipment, and storage medium, which can facilitate technical personnel to search for indicators by displaying an indicator cause-and-effect diagram, thereby improving the efficiency and accuracy of indicator search.

[0005] In one aspect, an embodiment of the present application provides a data processing method, the method comprising:

[0006] Obtain target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user with respect to a target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes indicator data of the user under each indicator;

[0007] Analyze and quantify the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain analysis and quantification results;

[0008] generating first metadata based on the analysis and quantification results, the first metadata including attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0009] When the indicator cause-effect diagram of the target object needs to be displayed in a front-end application, the first metadata is converted into second metadata, where the data format of the second metadata is compatible with the front-end application;

[0010] In the user interface displayed by the front-end application, the indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0011] On the other hand, an embodiment of the present application provides a data processing device, the device comprising:

[0012] an acquisition unit, configured to acquire target user data of a plurality of users, wherein the target user data of any user is generated based on the usage behavior of any user with respect to a target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes indicator data of the user under each indicator;

[0013] a processing unit, configured to analyze and quantify the causal relationship between each of the M indicators based on the indicator data of each user under each indicator, and obtain an analysis and quantification result;

[0014] The processing unit is further configured to generate first metadata based on the analysis and quantification results, wherein the first metadata includes attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0015] The processing unit is further configured to convert the first metadata into second metadata when displaying the indicator cause-effect diagram of the target object in a front-end application, wherein the data format of the second metadata is compatible with the front-end application;

[0016] The processing unit is also used to display the indicator causal graph based on the second metadata in the user interface displayed by the front-end application; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0017] In another aspect, an embodiment of the present application provides a computer device, the computer device including an input interface and an output interface, and the computer device further including:

[0018] a processor adapted to implement one or more instructions; and

[0019] A computer storage medium storing one or more instructions adapted to be loaded by the processor and executed by the processor:

[0020] Obtain target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user with respect to a target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes indicator data of the user under each indicator;

[0021] Analyze and quantify the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain analysis and quantification results;

[0022] generating first metadata based on the analysis and quantification results, the first metadata including attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0023] When the indicator cause-effect diagram of the target object needs to be displayed in a front-end application, the first metadata is converted into second metadata, where the data format of the second metadata is compatible with the front-end application;

[0024] In the user interface displayed by the front-end application, the indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0025] In another aspect, an embodiment of the present application provides a computer storage medium storing one or more instructions, wherein the one or more instructions are suitable for being loaded by a processor and executing the following steps:

[0026] Obtain target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user with respect to a target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes indicator data of the user under each indicator;

[0027] Analyze and quantify the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain analysis and quantification results;

[0028] generating first metadata based on the analysis and quantification results, the first metadata including attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0029] When the indicator cause-effect diagram of the target object needs to be displayed in a front-end application, the first metadata is converted into second metadata, where the data format of the second metadata is compatible with the front-end application;

[0030] In the user interface displayed by the front-end application, the indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0031] In an embodiment of the present application, the target user data of multiple users using the target object can be obtained first. The target user data of any user includes the indicator data of any user under each indicator. Since the target user data of each user is generated based on the usage behavior of each user for the target object, the target user data of each user has a high guiding value for analyzing the relationship between the various indicators of the target object. In this way, when the causal relationship between the various indicators configured for the target object is automatically analyzed and quantified based on the indicator data of each user under each indicator, a more accurate analysis and quantification result can be obtained. Then, a first metadata can be generated based on the analysis and quantification result, and the first metadata can be converted into a second metadata with a data format that is compatible with the front-end application, so that the indicator causal graph of the target object can be displayed in the user interface based on the second metadata. The above process automatically discovers, quantifies and draws the causal relationship between the various indicators of the target object, and can realize the intuitive display of the causal relationship between the indicators for the technicians by displaying the indicator causal graph, which facilitates the technicians to search for indicators and improves the efficiency and accuracy of indicator search. Moreover, since the indicator causal graph uses directed edges to connect the two nodes corresponding to two indicators with a causal relationship, technical personnel can clearly understand which indicator is the "cause" and which indicator is the "effect" through the direction of the directed edge, which further facilitates technical personnel to quickly locate the indicators they need and further improve the efficiency of indicator search. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0033] Figure 1aThis is a structural diagram of a data processing system provided in an embodiment of the present application;

[0034] Figure 1b This is a system architecture diagram of a data processing system provided by an embodiment of the present application;

[0035] Figure 2 This is a flow chart of a data processing method provided in an embodiment of the present application;

[0036] Figure 3 is a flow chart of a data processing method provided by another embodiment of the present application;

[0037] Figure 4a This is a schematic diagram of a cause-effect diagram list page displayed in a user interface provided by an embodiment of the present application;

[0038] Figure 4b This is a schematic diagram of displaying an indicator cause-effect diagram in a user interface provided by an embodiment of the present application;

[0039] Figure 4c This is a schematic diagram of switching the display style of each node in an indicator causal diagram from a first style to a second style, provided by an embodiment of the present application;

[0040] Figure 4d This is a schematic diagram showing each node according to its hierarchy, provided by an embodiment of the present application;

[0041] Figure 4e This is a schematic diagram of switching the theme color of an indicator cause-effect diagram from a current theme color to a target theme color, provided by an embodiment of the present application;

[0042] Figure 4f This is a schematic diagram of screening and viewing an indicator cause-effect diagram according to target information provided by an embodiment of the present application;

[0043] Figure 4g This is another schematic diagram of screening and viewing the indicator cause-effect diagram according to target information provided by an embodiment of the present application;

[0044] Figure 5 is a structural diagram of a data processing device provided in an embodiment of the present application;

[0045] Figure 6 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0047] In the embodiment of the present application, the target object mentioned later can be any smart device, such as a smart phone, smart watch, tablet computer, smart TV, etc.; it can also be any APP, such as a social APP, browser APP, video playback APP, etc., and the embodiment of the present application is not limited to this. In addition, the target object can be configured with M indicators, and these M indicators can be configured by relevant technical personnel according to the actual business needs of the target object; where M is an integer greater than 1. Taking the target object as a browser APP as an example, the M indicators configured for the target object may include but are not limited to the following indicators: "Increase click volume", "DAU (Daily Active User, number of active users on the same day)", "play_num_small_video (number of times a small video is played)", "play_num_short_video (number of times a short video is played)", "Is_youzhi_dau (whether it is a high-quality active user on the same day)", "Is_consume (whether there is content consumption)", etc. The small videos mentioned here refer to videos whose memory usage (i.e., video size) is less than the memory usage threshold, such as a video with a size of 2 megabytes; short videos refer to videos whose video length is less than the length threshold, such as a video with a length of 10 seconds.

[0048] In order to facilitate the technical personnel of the target object to quickly find other indicators related to a certain indicator of the target object when it is necessary to optimize the indicator of the target object, so as to optimize the indicator in combination with the other found indicators; the embodiment of the present application provides a data processing system for automatically discovering and quantifying the relationship between indicators. The data processing system can automatically discover and quantify the causal relationship between indicators for the target object, and automatically draw a graph based on the analysis and quantification results to obtain an indicator causal graph of the target object; then, the indicator causal graph is displayed with a friendly user interface, so that the causal relationship between the indicators of the target object can be intuitively displayed to the technical personnel through the indicator causal graph, so that the technical personnel can search for indicators based on the indicator causal graph, thereby improving the efficiency and accuracy of indicator search.

[0049] Among them, the causal relationship mentioned above refers to: the functional relationship between one indicator (i.e., "cause") and another indicator (i.e., "result"); for two indicators with a causal relationship, the change in the indicator value of the former indicator may cause the change in the indicator value of the latter indicator; for the convenience of explanation, the embodiment of the present application refers to the former indicator as the cause indicator and the latter indicator as the result indicator. For example, for the two indicators of "increase in clicks" and "DAU", when the indicator value of the indicator "for increase in clicks" changes, it usually causes the change in the indicator value of the indicator "DAU". Therefore, it can be considered that there is a causal relationship between the two indicators "for increase in clicks" and "DAU", and the indicator "for increase in clicks" can be called the cause indicator, and the indicator "DAU" can be called the result indicator. For example, for the two indicators "Is_youzhi_dau" and "Is_consume", when the indicator value of the indicator "Is_consume" changes, it will usually cause the indicator value of the indicator "Is_youzhi_dau" to change. Therefore, it can be considered that there is a causal relationship between the two indicators "Is_consume" and "Is_youzhi_dau", and the indicator "Is_consume" can be called the cause indicator, and the indicator "Is_youzhi_dau" can be called the result indicator, and so on.

[0050] The following combination Figure 1a The system architecture diagram shown in FIG. 1 illustrates the specific architecture of the data processing system proposed in the embodiment of the present application; see Figure 1aAs shown, the data processing system may include, but is not limited to, the following modules: an external service module, a data reading module, a causal graph discovery and quantification module, a causal graph metadata conversion module, and a front-end application display module. The external service module can call the data reading module through one or more services; such services may include HTTP (Hypertext Transfer Protocol) services, TRPC (a high-performance, multi-language RPC (Remote Procedure Call) development framework), and so on. Data streams can be transmitted between the data reading module and the causal graph discovery and quantification module, between the causal graph discovery and quantification module and the causal graph metadata conversion module, and between the causal graph metadata conversion module and the front-end application display module. Furthermore, the data reading module can communicate with various external data sources; such external data sources may be referred to as data sources and may include, but are not limited to, text files, distributed files, MySQL (a relational database management system), ClickHouse (a column-based database for real-time data analysis, essentially a column-oriented database), Apache Hive (a data warehouse tool based on Hadoop (a text search library)), and so on.

[0051] Specifically, the data reading module is mainly used to support data reading from multiple data sources, and after encapsulating the data read from different data sources into a unified data format, transmits it to the causal graph discovery and quantification module; the causal graph discovery and quantification module is mainly used to automatically analyze and quantify the causal relationship between the various indicators of the target object based on the data transmitted by the data reading module, and generate metadata related to the target object; the causal graph data conversion module is mainly used to complete the two-way conversion between the metadata generated by the causal graph discovery and quantification module and the metadata required for front-end application display; the front-end application display module is mainly used to provide a UI interface (user interface), draw and display the indicator causal graph through the UI interface, and modify the indicator causal graph and other operations. Based on this, the various modules in the data processing system work together to realize the general principle of displaying the indicator causal graph of the target object as follows:

[0052] The external service module can call the data reading module through one or more services, causing the data reading module to read the original user data of multiple users from multiple data sources through target code. The original user data of each user is generated based on each user's usage behavior for the target object. After reading the original user data of multiple users, the data reading module can encapsulate the original user data read from different external data sources into a unified data format to obtain the target user data of multiple users, and use this target user data of multiple users as input to the causal graph discovery and quantification module. The target user data of each user may include: indicator data of each user under each indicator of the target object; accordingly, the causal graph discovery and quantification module can call the built-in causal graph analysis function and causal graph quantification function to automatically analyze and quantify the causal relationship between the various indicators of the target object based on the indicator data of each user under each indicator, and generate first metadata that can be recognized by the causal graph metadata conversion module based on the analysis and quantification results, and then transmit this first metadata to the causal graph metadata conversion module. After receiving the first metadata, the causal diagram metadata conversion module can call the built-in data conversion function to convert the first metadata into second metadata that can be recognized by the front-end application, and transmit the second metadata to the front-end application display module, so that the front-end application display module displays the indicator causal diagram of the target object based on the second metadata.

[0053] It should be noted that the target code mentioned above can be Python (a cross-platform language) code or other types of code (such as Java (a cross-platform language) code, etc.), and there is no limitation on this. When the target code is Python code and the data source includes Apache Hive (Hive for short), considering that Python is usually not able to directly read the data in Apache Hive conveniently; based on this, the data processing system may also include a data transfer module (HiveData Transfer), which can be used to be called by the external service module, and transfer the original user data in the Apache Hive data source to other data sources (such as MySQL) through the PrestoAPI Client (a data processing tool for data transfer) in the public service, so that the data reading module can read the original user data stored in Apache Hive from other data sources (such as MySQL) through the target code, thereby improving the convenience of data reading and the success rate of data reading.

[0054] It can be seen that the data processing system proposed in the embodiment of the present application can support various data storage systems including text files, distributed files, MySQL, ClickHouse, and Apache Hive as data sources for discovering causal relationships, and has high flexibility and applicability (or ease of use); through the data reading module, various data sources are shielded, a unified data format is generated, and automatic causal relationships are discovered, quantified, and mapped, and finally the indicator causal graph is displayed to the technical personnel in a friendly interface. By displaying the indicator causal graph, the causal relationship between indicators can be intuitively reflected, so that the technical personnel can quickly locate the direct traction indicators and indirect traction indicators of the North Star indicator of the target object. The so-called North Star indicator refers to the key indicator formulated from the M indicators of the target object according to business needs; the direct traction indicator of the North Star indicator refers to: the indicator that directly optimizes the North Star indicator by optimizing itself (that is, the cause indicator of the North Star indicator), and the indirect traction indicator refers to: the indicator that indirectly optimizes the North Star indicator by optimizing itself.

[0055] Regarding the data processing system described above, the following points need to be explained:

[0056] ①The data processing system mentioned above is only Figure 1a The system structure is exemplarily represented and does not limit the system structure of the data processing system; for example, Figure 1a The data handling module, data reading module and external service module in the data processing system shown are three independent modules, but in other embodiments, the data handling module, data reading module and external service module in the above data processing system can also be integrated into one module, and so on.

[0057] ② The various modules in the data processing system mentioned above can be integrated into a computer device, which can be a terminal or a server. Alternatively, the various modules in the data processing system mentioned above can be deployed in different computer devices, and one computer device can deploy one or more modules. For example, see Figure 1bAs shown, the external service module can be deployed in server 10, the data reading module can be deployed in server 11, the causal graph discovery and quantification module can be deployed in server 12, the causal graph metadata conversion module can be deployed in server 13, the front-end application display module can be deployed in terminal 14, the data handling module can be deployed in server 15, and so on. It should be noted that the terminal mentioned here can be: a smart phone, a tablet computer, a laptop computer, a desktop computer or a smart TV, etc.; the server mentioned here can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as basic cloud computing services such as big data and artificial intelligence platforms, etc.

[0058] ③ The various modules in the aforementioned data processing system may be located outside or within the blockchain network, without limitation. The so-called blockchain network is a network composed of a peer-to-peer network (P2P network) and a blockchain. Blockchain refers to a new application model for computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. It is essentially a decentralized database, a series of data blocks (or blocks) generated using cryptographic methods. Specifically, a blockchain may be composed of multiple blocks, and the block with the earliest creation time among these multiple blocks is called a genesis block; the genesis block includes a block header and a block body, the block header stores the input information characteristic value, version number, timestamp and difficulty value, and the block body stores the input information; the next block of the genesis block uses the genesis block as its parent block, and the next block also includes a block header and a block body, the block header stores the input information characteristic value of the current block, the block header characteristic value, version number, timestamp and difficulty value of the parent block, and so on, so that the block data stored in each block in the blockchain is associated with the block data stored in the parent block, thereby ensuring the security of the input information in the block.

[0059] Based on the above description, the embodiment of the present application proposes a data processing method, which can be executed in the data processing system mentioned above. In the embodiment of the present application, the target object can be configured with M indicators; see Figure 2 , the data processing method may include the following steps S201-S205:

[0060] S201, obtaining target user data of multiple users.

[0061] The target user data for any user is generated based on the user's usage behavior for the target object. Specifically, the target user data for any user includes the user's indicator data for each indicator. The indicator data for any user for the mth indicator can be used to indicate whether the user's usage behavior for the target object meets the indicator conditions for the mth indicator; alternatively, it can be used to indicate the indicator value generated by the user's usage behavior for the target object under the mth indicator, where m∈[1,M].

[0062] For example, assume that the M indicators configured for the target object include the following indicators: "Increase Clicks", "Is_youzhi_dau", "Is_consume (whether there is consumed content)", etc. If the indicator data for "Increase Clicks" for any user is 5, then this indicator data can be used to indicate that the increase in clicks generated by any user's usage behavior for the target object is 5; if the indicator data for "Is_youzhi_dau" for any user is 1, then this indicator data can be used to indicate that any user's usage behavior for the target object meets the indicator conditions for high-quality active users on that day; if the indicator data for "Is_consume" for any user is 0, then this indicator data can be used to indicate that any user's usage behavior for the target object does not meet the indicator conditions for having consumed content, and so on.

[0063] S202 , analyzing and quantifying the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator, and obtaining analysis and quantification results.

[0064] In a specific implementation, the specific implementation method of step S202 may be: according to the indicator data of each user under each indicator, the causal relationship between each indicator in the M indicators is analyzed to obtain causal relationship data. Specifically, a causal analysis algorithm may be used to analyze the causal relationship between each indicator in the M indicators according to the indicator data of each user under each indicator to obtain causal relationship data. Among them, the causal analysis algorithm may include but is not limited to: PC (PC Algorithm, a causal analysis algorithm) algorithm, or FCI (Fast Causal Inference, fast causal inference) algorithm, or other open source or self-developed causal relationship inference algorithms, etc. After obtaining the causal relationship data, the causal relationship data can be directly used as the analysis and quantification results; that is, under this implementation method, the analysis and quantification results include causal relationship data.

[0065] Among them, the causal relationship data can be represented in the form of a graph, or in the form of a table or text, and there is no limitation on this. The causal relationship data can be used to indicate N (N is a positive integer) credible indicator pairs, and any credible indicator pair includes a cause indicator and a result indicator; the so-called credible indicator pair refers to: an indicator pair consisting of two indicators with a causal relationship, and the probability that the two indicators in the credible indicator pair have a causal relationship is greater than a probability threshold. Optionally, the causal relationship data obtained by the causal analysis algorithm can also be used to indicate one or more suspicious indicator pairs; the so-called suspicious indicator pair refers to: an indicator pair consisting of two indicators that may have a causal relationship, and the probability that the two indicators in the suspicious indicator pair have a causal relationship is less than a probability threshold and greater than 0.

[0066] In another specific implementation, step S202 may be implemented as follows: first, based on the indicator data of each user under each indicator, the causal relationship between each of the M indicators may be analyzed to obtain causal relationship data. Next, the causal relationship data and the target user data of multiple users may be used to quantify the effect gain of each trustworthy indicator pair. Specifically, the causal relationship data and the target user data of multiple users may be input into a causal inference toolkit, which then quantifies the effect gain of each trustworthy indicator pair based on the input causal relationship data and the target user data of multiple users. The causal inference toolkit mentioned herein may be configured based on empirical values, such as the DoWhy toolkit, which is an open source Python causal inference library. After obtaining the effect gain of each trustworthy indicator pair, the effect gain of each trustworthy indicator pair may be added to the analysis and quantification results. In other words, in this embodiment, the analysis and quantification results may include the effect gain of each trustworthy indicator pair. Optionally, the causal relationship data may also be added to the analysis and quantification results. In this case, the analysis and quantification results may include the causal relationship data and the effect gain of each trustworthy indicator pair.

[0067] The effect gain of any credible indicator pair is used to measure: when the cause indicator in any credible indicator pair is optimized to a baseline degree, the degree to which the result indicator is optimized based on the optimization of the cause indicator. The meaning of "the indicator is optimized" is: the indicator value of the indicator is improved; for example, "the cause indicator is optimized to a baseline degree" means: the indicator value of the cause indicator is improved by a baseline degree. The baseline degree mentioned here can be set according to an empirical value. For example, if the baseline degree is set to 1%, if the effect gain of any credible indicator pair is 0.5%, then the effect gain can be used to measure: when the cause indicator in any credible indicator pair is optimized by 1%, the degree to which the result indicator is optimized due to the optimization of the cause indicator is 0.5%; that is, the effect gain can be used to measure: when the indicator value of the cause indicator in any credible indicator pair is improved by 1%, the indicator value of the result indicator can also be improved by 0.5%. Optionally, if the causal relationship data is also used to indicate one or more suspicious indicator pairs, the analysis and quantification results can also include the effect gain of each suspicious indicator pair.

[0068] S203 : Generate first metadata based on the analysis and quantification results. The first metadata includes attribute information of the N trustworthy indicator pairs. That is, the first metadata can be understood as data for describing the attribute information of the N trustworthy indicator pairs.

[0069] In a specific implementation, a data table can be constructed first. The data table can include multiple storage locations; each storage location includes at least the following fields: a cause indicator field, a result indicator field, and an effect gain field. Secondly, the attribute information of the nth trusted indicator pair can be obtained based on the analysis and quantification results, n∈[1, N]; the attribute information includes the following sub-information: the indicator identifier of the cause indicator in the nth trusted indicator pair, the indicator identifier of the result indicator in the nth trusted indicator pair, and the effect gain of the nth trusted indicator pair. Then, each sub-information in the attribute information of the nth trusted indicator pair can be filled into each field of the nth storage location in the data table respectively; the data table filled with the attribute information of N trusted indicator pairs is determined as the first metadata.

[0070] It should be understood that, in addition to the three fields mentioned above, each storage location may also include other fields, and the embodiments of the present application do not limit this. For example, each storage location may also include a view identification field, which is used to store the causal graph identification of the corresponding indicator causal graph; then, in this case, the attribute information of the nth trusted indicator pair may also include: the causal graph identification of the indicator causal graph corresponding to the nth trusted indicator pair. For another example, each storage location may also include a type identification field, which is used to store the type identification of the indicator pair; when the type identification of any indicator pair is the first identification (such as the value 1), it indicates that any indicator pair is a trusted indicator pair, and when the type identification of any indicator pair is the second identification (such as the value 0), it indicates that any indicator pair is a suspicious indicator pair. Then in this case, the attribute information of the nth trusted indicator pair may also include: the first identification.

[0071] Optionally, if the causal relationship data can also be used to indicate one or more suspicious indicator pairs, and the analysis and quantification results also include the effect gain of each suspicious indicator pair, then the attribute information of each suspicious indicator pair can also be obtained based on the analysis and quantification results, and the attribute information of each suspicious indicator pair can be respectively filled into each field of the corresponding storage location in the data table; the specific implementation method is similar to the method of filling the attribute information of the trusted indicator pair into the data table, and will not be repeated here. After the attribute information of each trusted indicator pair and the attribute information of each suspicious indicator pair are filled into the data table, the first metadata can be obtained; that is, the first metadata in this case can include not only the attribute information of N trusted indicator pairs, but also the attribute information of one or more suspicious indicator pairs.

[0072] For example, assume that the target object is configured with M indicators as follows:

[0073] is_youzhi_dau (whether it is a high-quality active user on the day), is_top_dau (whether it is the best active user), is_top (whether it is the best), is_consume (whether there is consumer content), active_days_7d (number of active days of the user within 7 days), active_days_30d (number of active days of the user within 30 days), age (age), gender (gender)…interest_tag_num (the number of interests the user has)…back_exp_total (total number of background exposures)…play_duration_small_video (total duration of small video playback), play_duration_short_video (total duration of short video playback), play_num_small_video (number of small video playbacks), play_num_short_video (total duration of short video playback), etc.

[0074] Assume that target user data of 5 users is obtained, and the obtained target user data is as follows:

[0075] The target user data of user 1 includes the following indicator data: 1, 1, 1, 1, 7, 30, 28, 1…12…3…0, 615, 0, 9.

[0076] The target user data of user 2 includes the following indicator data: 1, 1, 1, 1, 7, 26, 25, 2…888…253…3504, 0, 170, 0;

[0077] The target user data of user 3 includes the following indicator data: 0, 0, 0, 1, 2, 8, 9, 2…1…33…0, 0, 0, 0;

[0078] The target user data of user 4 includes the indicator data of each indicator as follows: 0, 1, 1, 0, 2, 2, 22, 2…0…118…0, 0, 0, 0;

[0079] The target user data of user 5 includes the following indicator data: 1, 1, 1, 1, 7, 21, 16, 1…675…119…2831, 0, 188, 0;

[0080] Based on the description of steps S202-S203, by analyzing and quantifying the causal relationship between the M indicators for the indicator data of each user under each indicator mentioned in the above example, the first metadata generated based on the analysis and quantification results can be shown in Table 1:

[0081] Table 1

[0082] uuid source target zengliang view_id 1 interest_tag_num is_ss 0.1355％ 2021_737c7a 1 total_duration interest_tag_num 0.1715％ 2021_737c7a 1 is_consume back_exp_total_short_video 0.3746％ 2021_737c7a … … … …

[0083] Among them, uuid represents the type identification field, source represents the cause indicator field, target represents the result indicator field, zengliang represents the effect gain field, and view_id represents the view identification field. It should be noted that the embodiment of the present application is described by taking the first metadata in the form of a table as an example, but does not limit the form of the first metadata; for example, in other embodiments, the first metadata can be in the form of a text file, etc. In addition, after the first metadata is generated, the first metadata can also be saved in a background database, which can be MySQL, or ClickHouse, Oracle (a relational database management system suitable for large, medium and micro computers), Presto (data query engine), MongoDB (a database based on distributed file storage), Elasticsearch (a search server for a full-text search engine that provides distributed multi-user capabilities), Apache Druid (a distributed column storage engine) and other databases, without limitation.

[0084] S204, when the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, the first metadata is converted into second metadata; the data format of the second metadata is adapted to the front-end application, and the second metadata can be understood as data used to describe the indicator cause-effect diagram.

[0085] In a specific implementation, when it is necessary to display the indicator cause-effect diagram of the target object in the front-end application, the indicator identifiers of M indicators and the data format adapted to the front-end application can be used to generate node data of M nodes, with one node being used to record one indicator. Specifically, M nodes for recording M indicators can be determined, and the indicator identifier of each indicator can be determined as the node identifier of the corresponding node, and the display coordinates of each node can be determined. Secondly, the level of each node can be determined based on the attribute information of N trusted indicator pairs; wherein the level of the node corresponding to the cause indicator in any trusted indicator pair is lower than the level of the node corresponding to the result indicator. Then, the data format adapted to the front-end application can be used to generate node data of M nodes based on the node identifiers, corresponding display coordinates and corresponding levels of the M nodes. It should be understood that in other embodiments, the level of each node can also be ignored; that is, the data format adapted to the front-end application can be directly used to generate node data of M nodes based on the node identifiers and corresponding display coordinates of the M nodes.

[0086] The data format adapted to the front-end application mentioned above can be set based on empirical values. For example, assuming the data format adapted to the front-end application is JSON (a lightweight data exchange format); then taking the indicator "is_youzhi_dau" as an example, the node assigned to this indicator is called node a. The node data of node a can be obtained through the above steps as follows:

[0087]

[0088] In addition, the attribute information of the N trusted indicator pairs in the first metadata and the data format adapted to the front-end application can be used to generate edge data of N directed edges; the edge data of any directed edge includes the weight of any directed edge, which is equal to the effect gain of the trusted indicator pair corresponding to any directed edge. Among them, a directed edge refers to an edge with a direction, and a directed edge has an arrow for indicating the direction. Optionally, the edge data of any directed edge may also include: the edge identifier of any directed edge, the node identifiers of the two nodes connected by any directed edge (i.e., the node identifier of the starting node (the node for recording the cause indicator) and the node identifier of the end node (the node for recording the result indicator)), indication information for indicating whether the edge has an arrow, the display coordinates of any directed edge, and so on. Assume that the data format adapted to the front-end application is still in JSON format, and take the trusted indicator pair "is_youzhi_dau" and "back_exp_total_live" as an example, where "is_youzhi_dau" is the cause indicator and "back_exp_total_live" is the result indicator. The edge data of the directed edge corresponding to this trusted indicator pair can be as follows:

[0089]

[0090]

[0091] Then, the node data of M nodes and the edge data of N directed edges can be integrated to obtain the second metadata; specifically, the node data of M nodes and the edge data of N directed edges can be directly integrated to obtain the second metadata. Optionally, if the first metadata also includes the attribute information of one or more suspicious indicator pairs, the attribute information of one or more suspicious indicator pairs in the first metadata and the data format adapted to the front-end application can also be used to generate the edge data of one or more undirected edges; the edge data of any undirected edge includes the weight of any undirected edge (that is, the effect gain of the suspicious indicator pair corresponding to any undirected edge). Among them, an undirected edge refers to an edge without direction, and an undirected edge does not have an arrow for indicating the direction. In this case, the node data of M nodes, the edge data of N directed edges and the edge data of each undirected edge can be integrated to obtain the second metadata. For example, taking the first metadata shown in Table 1 as an example, after the first metadata is converted into the second metadata, part of the data in the second metadata can be seen as shown in the following json code:

[0092]

[0093]

[0094] Optionally, in other embodiments, after the first metadata is generated or the second metadata is converted, the first metadata and the second metadata may be stored in the blockchain of the blockchain network to prevent the first metadata and the second metadata from being maliciously tampered with, thereby improving data security.

[0095] S205 : Displaying an indicator cause-and-effect diagram based on the second metadata in a user interface displayed by the front-end application.

[0096] In one specific implementation, an indicator causal graph can be drawn based on the second metadata first, and then the drawn indicator causal graph is sent to the front-end application, which directly displays the indicator causal graph in the user interface. In another implementation, the second metadata can be sent directly to the front-end application, which draws the indicator causal graph based on the second metadata and displays the indicator causal graph in the user interface. The indicator causal graph includes M nodes and N directed edges, where one node records one indicator, and one directed edge is used to connect two nodes corresponding to a trusted indicator pair; optionally, the indicator causal graph can also include one or more undirected edges, and one undirected edge is used to connect two nodes corresponding to a suspicious indicator pair.

[0097] After the indicator cause and effect diagram is displayed, the technician can also be supported to perform a move operation (or drag operation) on any element (any node or the weight of any edge) in the indicator cause and effect diagram to adjust the display position of each element in the indicator cause and effect diagram. Accordingly, in the process of displaying the indicator cause and effect diagram, if there is a move operation for any element in the indicator cause and effect diagram, any element can be moved according to the move operation to update the indicator cause and effect diagram. Then, new second metadata can be generated based on the updated indicator cause and effect diagram, and the new second metadata can be converted into new first metadata; and the new first metadata can be stored so that the next time the indicator cause and effect diagram of the target object needs to be displayed, the new first metadata can be converted into new second metadata, and the updated indicator cause and effect diagram can be displayed based on the new second metadata.

[0098] The specific method for generating new second metadata based on the updated indicator causal graph is as follows: using the display coordinates and other information of each node in the updated causal graph to generate new node data for each node, and using the display coordinates and other information of each edge in the updated causal graph to generate new edge data for each edge; and integrating the new node data for each node and the new edge data for each edge to obtain new second metadata. For example, the new second metadata may be as follows:

[0099] Part of the data in the second metadata can be seen in the following json code:

[0100]

[0101]

[0102] In an embodiment of the present application, the target user data of multiple users using the target object can be obtained first. The target user data of any user includes the indicator data of any user under each indicator. Since the target user data of each user is generated based on the usage behavior of each user for the target object, the target user data of each user has a high guiding value for analyzing the relationship between the various indicators of the target object. In this way, when the causal relationship between the various indicators configured for the target object is automatically analyzed and quantified based on the indicator data of each user under each indicator, a more accurate analysis and quantification result can be obtained. Then, a first metadata can be generated based on the analysis and quantification result, and the first metadata can be converted into a second metadata with a data format that is compatible with the front-end application, so that the indicator causal graph of the target object can be displayed in the user interface based on the second metadata. The above process automatically discovers, quantifies and draws the causal relationship between the various indicators of the target object, and can realize the intuitive display of the causal relationship between the indicators for the technicians by displaying the indicator causal graph, which facilitates the technicians to search for indicators and improves the efficiency and accuracy of indicator search. Moreover, since the indicator causal graph uses directed edges to connect the two nodes corresponding to two indicators with a causal relationship, technical personnel can clearly understand which indicator is the "cause" and which indicator is the "effect" through the direction of the directed edge, which further facilitates technical personnel to quickly locate the indicators they need and further improve the efficiency of indicator search.

[0103] See Figure 3 , is a flow chart of another data processing method provided in an embodiment of the present application. This data processing method can be executed in the data processing system mentioned above. Figure 3 , the data processing method may include the following steps S301-S308:

[0104] S301: Determine Q data sources, where Q is an integer greater than 1.

[0105] In an embodiment of the present application, the Q data sources may include but are not limited to: text files, distributed files, MySQL, ClickHouse, Apache Hive, and the like. Any data source stores the original user data of at least one user, and the original user data of each user is generated based on each user's usage behavior for the target object. The Q data sources may be pre-set according to business needs; or, the Q data sources may be specified by a technician through a user interface, such as displaying data source identifiers of multiple data sources (such as data source names, network addresses of data sources, etc.) in the user interface, and the data sources indicated by the selected Q data source identifiers are determined as the Q data sources according to the identifier selection operation.

[0106] S302: Determine a target data processing tool that is compatible with the qth data source according to the adaption relationship between the data source and the data processing tool.

[0107] Where q∈[1, Q]; the adaptation relationship between data sources and data processing tools can be pre-set based on empirical values, that is, the data processing tool adapted to each data source is pre-set. For example, for a text file data source, the adapted data processing tool can be the Python Pandas file reading interface, Pandas is a data analysis package in Python; for a MySQL data source, the adapted data processing tool can be the Python PyMySQL package; for a ClickHouse data source, the adapted data processing tool can be the Python clickhouse_driver package (a data reading package); for a Apache Hive data source, the adapted data processing tool can be the Presto API Client, and so on.

[0108] S303 , calling the target code to read the original user data of K users in the q th data source through the target data processing tool, where K is a positive integer.

[0109] During implementation, the data storage format of the qth data source can be determined first. If the data storage format is the first storage format, the target code can be directly invoked to read the raw user data of K users from the qth data source through the target data processing tool. The first storage format refers to a storage format compatible with the read and encapsulation capabilities of the target code. For example, if the target code is Python code, the first storage format can be a text file format, MySQL format, ClickHouse format, etc., which are compatible with the read and encapsulation capabilities of Python code.

[0110] If the data storage format is the second storage format, the target code can be called first to use the target data processing tool to move the original user data of K users from the qth data source to the designated data source, so as to store the original user data of K users in the data storage format of the designated data source; and the target code can be called to read the original user data of K users from the designated data source through the data processing tool adapted to the designated data source. The second storage format refers to a storage format that is incompatible with the reading and packaging capabilities of the target code; for example, if the target code is Python code, the second storage format can be an Apache Hive format that is incompatible with the reading and packaging capabilities of the Python code. The designated data source refers to a data source of the first storage format; for example, the designated data source can be MySQL, etc.

[0111] Specifically, for a text file data source, the target code can be directly called to read at least one user's raw user data from the text file using the Python Pandas file reading interface. For a MySQL data source, the target code can be directly called to read at least one user's raw user data from MySQL using the Python PyMySQL package. For a ClickHouse data source, the target code can be directly called to read at least one user's raw user data from ClickHouse using the Python clickhouse_driver package. For an Apache Hive data source, the target code can first be called to move each user's raw user data stored in Apache Hive to the specified data source using the Presto API Client. Then, the target code can be called to read the moved raw user data from the specified data source using the data reading tool adapted for the indicator data source. Furthermore, if the specified data source is MySQL, given that MySQL cannot store large amounts of data, when moving the raw user data from Apache Hive to MySQL, the amount of raw user data to be moved can be limited to or less than a target number (e.g., 1 million) to ensure MySQL's performance.

[0112] S304 , after reading the original user data of the K users, encapsulate the original user data of the K users in a preset format to obtain the target user data of the K users.

[0113] The preset format mentioned here can be set based on empirical data. For example, when the target code is Python code, the preset format can be a Pandas dataframe, a tabular data structure. By encapsulating the raw user data from each data source into the preset format, uniformly formatted target user data can be obtained, facilitating subsequent analysis and quantitative processing. The target user data for any user is generated based on the user's usage behavior for the target object; specifically, the target user data for any user includes the user's indicator data for each indicator.

[0114] It should be noted that the above steps S301-S304 describe Figure 2 The specific implementation of step S201 in the method embodiment shown in FIG. 1 ; and the specific implementation of the following steps S305-S306 can be found in the above Figure 2 The description of steps S202 - S203 in the illustrated method embodiment will not be repeated here.

[0115] S305 , analyzing and quantifying the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator, and obtaining an analysis and quantification result.

[0116] S306: Generate first metadata based on the analysis and quantification results. The first metadata includes attribute information of N trustworthy indicator pairs, where N is a positive integer.

[0117] S307 , when the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, the first metadata is converted into second metadata, and the data format of the second metadata is adapted to the front-end application.

[0118] In a specific implementation, a cause-effect diagram list page may be displayed in the user interface of the front-end application; see Figure 4a As shown, the causal diagram list page may include at least one or more causal diagram options, and among the one or more causal diagram options, there is at least a target causal diagram option 41 for viewing the indicator causal diagram of the target object. When the target causal diagram option is triggered, it can be determined that the indicator causal diagram of the target object needs to be displayed in the front-end application. In another specific implementation, after the first metadata is successfully generated through step S306, it can be directly determined that the indicator causal diagram of the target object needs to be displayed in the front-end application; or, when a user voice is received and the user voice is used to instruct to view the indicator causal diagram of the target object, it can be determined that the indicator causal diagram of the target object needs to be displayed in the front-end application. It should be understood that the embodiment of the present application only exemplifies several situations in which the indicator causal diagram of the target object needs to be displayed in the front-end application, and is not exhaustive. After determining that the indicator causal diagram of the target object needs to be displayed in the front-end application, the first metadata can be converted into the second metadata. The specific conversion method can be found in the above. Figure 2 The description of step S204 in the illustrated method embodiment will not be repeated here.

[0119] S308, in the user interface displayed by the front-end application, an indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect two nodes corresponding to a trusted indicator pair.

[0120] As can be seen from the foregoing, the second metadata may include node data for M nodes and edge data for N directed edges; the node data for any node includes at least: the node identifier and corresponding display coordinates of any node; and the edge data for any directed edge includes at least: the weight of any directed edge and the node identifiers of the two nodes connected by any directed edge. Therefore, the specific implementation of step S308 may be as follows: obtaining the display coordinates of each node from the node data for the M nodes; and drawing each node using a first style based on the display coordinates of each node; wherein, when drawing each node using the first style, the node identifiers of each node are hidden. Drawing the nth edge between the two nodes indicated by the node identifiers in the nth edge data, and drawing the weights in the nth edge data at a target location; the target location is associated with the drawing location of the nth edge, n∈[1,N]. After all N edges and the weights of each edge are drawn, an indicator causal graph is obtained; and the indicator causal graph is displayed in the user interface displayed by the front-end application.

[0121] Displaying each node in the first style can make the indicator cause-effect diagram more concise, making it easier for technical personnel to understand the topological structure of the overall cause-effect relationship. It should be noted that in other embodiments, when the first style is used to display each node, the weights of each edge can also be hidden in the indicator cause-effect diagram to further simplify the indicator cause-effect diagram, thereby highlighting the topological structure of the cause-effect relationship between indicators. That is to say, in the specific implementation process of step S308, after drawing the nth edge, the step of drawing the weight in the nth edge data at the target position can also be omitted. For example, taking the first style as a small dot style and hiding the weights of each edge as an example, the displayed indicator cause-effect diagram can be Figure 4b shown.

[0122] In an optional embodiment, the embodiment of the present application may support style switching of nodes in the indicator cause-effect diagram; accordingly, in the process of displaying the indicator cause-effect diagram, if a node style switching operation for the indicator cause-effect diagram is detected, the display style of each node in the indicator cause-effect diagram may be switched from the first style to the second style; wherein, when each node is drawn in the second style, the node identifier of each node is displayed. Optionally, when a node style switching operation is detected, if the indicator cause-effect diagram hides the weights of each edge, then when the display style of each node is switched from the first style to the second style, the weights of each edge may be displayed together. The node style switching operation mentioned here may include any of the following: a triggering operation for a style switching button in the user interface, an operation for inputting a style switching gesture, a click operation / pressing operation for a blank area of ​​the user interface, and the like. For example, if the user interface includes a style switching button 42, and the node style switching operation is a triggering operation for the style switching button 42, then a schematic diagram of switching the display style of each node in the indicator cause-effect diagram from the first style to the second style and displaying the weights of each edge can be found in Figure 4c It should be understood that Figure 4c The lower figure only exemplifies the weights of some edges. In actual applications, the weight of each edge can be displayed in the indicator cause-effect diagram. It should also be understood that in other embodiments, when the indicator cause-effect diagram is first displayed, the nodes of the indicator cause-effect diagram can also be displayed in the second style, and the display style of each node in the indicator cause-effect diagram can be switched from the second style to the first style.

[0123] Furthermore, if the node data of any node also includes: the level of any node; then the embodiment of the present application can use the corresponding display color to display each node according to the level of each node, regardless of whether the first style or the second style is used to display each node. That is, in this case, the display color of each node in the indicator cause-effect diagram is determined according to the level of each node; different display colors are presented between nodes of different levels. For example, if the display color of the node with level 0 is black and the display color of the node with level 1 is gray, then the schematic diagram of displaying each node with the corresponding display color according to the level of the node can be seen in Figure 4d As shown. Figure 4d As shown, by using different display colors to display nodes at different levels, the causal relationship between each node can be more intuitively highlighted.

[0124] In another optional embodiment, if the indicator cause and effect diagram has a theme color, the embodiment of the present application may support switching the theme color of the indicator cause and effect diagram. The so-called theme color refers to the color used to indicate the main hue of the indicator cause and effect diagram. Accordingly, during the display of the indicator cause and effect diagram, if a theme color switching operation for the indicator cause and effect diagram is detected, the theme color of the indicator cause and effect diagram can be switched from the current theme color to the target theme color. The current theme color refers to the theme color presented by the indicator cause and effect diagram when the theme color switching operation is detected. The theme color switching operation may include but is not limited to: triggering the theme color switching button in the user interface, inputting the theme color switching gesture, selecting one or more candidate theme colors displayed in the user interface, and so on. When the theme color switching operation includes a selection operation for one or more candidate theme colors displayed in the user interface, the target theme color may be the selected candidate theme color; when the theme color switching operation includes a triggering operation for the theme color switching button in the user interface, inputting the theme color switching gesture, and so on, the target theme color may be a pre-set default theme color. For example, assuming that the current theme color is black and the target theme color is white, the diagram of switching the theme color of the indicator cause-effect diagram from the current theme color to the target theme color can be seen in Figure 4e shown.

[0125] In another optional implementation, considering that the arrangement between automatically generated nodes may be a bit too much, the weights on the edges between nodes will overlap when the number of nodes is large; based on this, in order to facilitate technicians to view the part of the indicator cause-effect diagram that they are interested in, the embodiment of the present application can also support technicians to filter and view the indicator cause-effect diagram through filtering indicators such as node identification prefix or edge weight. Accordingly, when displaying the indicator cause-effect diagram, an information input area 43 can be displayed in the user interface; if target information is obtained in the information input area, the target nodes and target edges that match the target information are filtered out from the indicator cause-effect diagram; then, in the indicator cause-effect diagram displayed in the user interface, the target nodes and target edges can be kept displayed, and the nodes other than the target nodes and the edges other than the target edges can be canceled. Among them, the target information can be a target node identification prefix, a target weight, and so on.

[0126] If the target information is a target node identifier prefix, the specific method of filtering out target nodes and target edges that match the target information from the indicator causal graph may be: according to the node identifiers of each node in the indicator causal graph and the target node identifier prefix, determine the first node from the indicator causal graph, and determine the first node as the target node; the first node here refers to: the node corresponding to the node identifier including the target node identifier prefix. Optionally, the second node connected to the first node may also be determined as the target node. Then, the edges between the determined target nodes may be used as target edges. For example, taking the target node identifier prefix as "is_", and selecting the first node and the second node as the target nodes as an example, the schematic diagram for keeping the target nodes and target edges displayed can be seen in Figure 4f shown.

[0127] If the target information is a target weight, the specific method of filtering out target nodes and target edges that match the target information from the indicator causal graph can be: based on the weight of each edge in the indicator causal graph, determine the edge with a weight greater than or equal to the target weight as the target edge; and determine the node connected by each target edge as the target node. For example, if the target weight is 0.0001, the schematic diagram of keeping the target nodes and target edges displayed can be seen in Figure 4g shown.

[0128] Optionally, in order to facilitate the technicians to flexibly select the screening indicators, a screening indicator selection area (such as Figure 4f and Figure 4g When the technician wants to filter and view the indicator cause-effect diagram by using the node identifier prefix as a filter indicator, he can select the first filter indicator option in the filter indicator selection area (such as Figure 4f Then, enter the target node identifier prefix in the information input area 43 to filter and view the indicator cause-effect diagram. If the technician wants to filter and view the indicator cause-effect diagram by the weight as a filtering indicator, he can select the second filtering indicator option in the filtering indicator selection area (such as Figure 4g Then, enter the target weight in the information input area 43 to filter and view the indicator cause-effect diagram.

[0129] In an embodiment of the present application, the target user data of multiple users using the target object can be obtained first. The target user data of any user includes the indicator data of any user under each indicator. Since the target user data of each user is generated based on the usage behavior of each user for the target object, the target user data of each user has a high guiding value for analyzing the relationship between the various indicators of the target object. In this way, when the causal relationship between the various indicators configured for the target object is automatically analyzed and quantified based on the indicator data of each user under each indicator, a more accurate analysis and quantification result can be obtained. Then, a first metadata can be generated based on the analysis and quantification result, and the first metadata can be converted into a second metadata with a data format that is compatible with the front-end application, so that the indicator causal graph of the target object can be displayed in the user interface based on the second metadata. The above process automatically discovers, quantifies and draws the causal relationship between the various indicators of the target object, and can realize the intuitive display of the causal relationship between the indicators for the technicians by displaying the indicator causal graph, which facilitates the technicians to search for indicators and improves the efficiency and accuracy of indicator search. Moreover, since the indicator causal graph uses directed edges to connect the two nodes corresponding to two indicators with a causal relationship, technical personnel can clearly understand which indicator is the "cause" and which indicator is the "effect" through the direction of the directed edge, which further facilitates technical personnel to quickly locate the indicators they need and further improve the efficiency of indicator search.

[0130] Based on the description of the above data processing method embodiment, the embodiment of the present application further discloses a data processing device; the data processing device can be a computer program (including program code) running on a computer device, and the data processing device can execute Figure 2 or Figure 3 See the method shown in Figure 5 The data processing device can run the following units: an acquisition unit 501 and a processing unit 502; wherein the acquisition unit 501 may include the data reading module, data handling module, and external service module mentioned in the above data processing system, etc.; the processing unit 502 may include the causal graph analysis and quantification module, causal graph metadata conversion module, front-end application display module, etc. mentioned in the above data processing system. Specifically:

[0131] The acquisition unit 501 may be configured to acquire target user data of multiple users. The target user data of any user is generated based on the user's usage behavior for a target object. The target object is configured with M indicators, where M is an integer greater than 1. The target user data of any user includes indicator data of the user under each indicator.

[0132] The processing unit 502 may be configured to analyze and quantify the causal relationship between each of the M indicators based on the indicator data of each user under each indicator to obtain an analysis and quantification result;

[0133] The processing unit 502 is further configured to generate first metadata based on the analysis and quantification results, wherein the first metadata includes attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0134] The processing unit 502 is further configured to convert the first metadata into second metadata when displaying the indicator cause-effect diagram of the target object in a front-end application, wherein the data format of the second metadata is compatible with the front-end application;

[0135] The processing unit 502 is also used to display the indicator causal graph based on the second metadata in the user interface displayed by the front-end application; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0136] In one embodiment, when the processing unit 502 is configured to analyze and quantify the causal relationship between each of the M indicators based on the indicator data of each user under each indicator to obtain the analysis and quantification results, it can be specifically configured to:

[0137] Analyze the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain causal relationship data; the causal relationship data is used to indicate N credible indicator pairs, each credible indicator pair includes a cause indicator and a result indicator;

[0138] The causal relationship data and the target user data of the multiple users are used to quantify the effect gain of each trustworthy indicator pair; the effect gain of any trustworthy indicator pair is used to measure: when the cause indicator in any trustworthy indicator pair is optimized to a baseline degree, the degree to which the result indicator is optimized based on the optimization of the cause indicator;

[0139] The effect gain of each pair of credible indicators is added to the analysis quantification result.

[0140] In another embodiment, when the processing unit 502 is used to generate the first metadata in a data format based on the analysis and quantification result, it can be specifically used to:

[0141] Constructing a data table, the data table including a plurality of storage locations; each storage location including at least the following fields: a cause indicator field, a result indicator field, and an effect gain field;

[0142] Acquire attribute information of an nth credible indicator pair based on the analysis and quantification results, where n∈[1,N]; the attribute information includes the following sub-information: an indicator identifier of a cause indicator in the nth credible indicator pair, an indicator identifier of a result indicator in the nth credible indicator pair, and an effect gain of the nth credible indicator pair;

[0143] Fill each sub-information in the attribute information of the nth trust indicator pair into each field of the nth storage position in the data table;

[0144] The data table filled with the attribute information of the N trustworthy indicator pairs is determined as the first metadata.

[0145] In another embodiment, when the indicator cause-effect diagram of the target object is to be displayed in a front-end application, the processing unit 502, when converting the first metadata into the second metadata, may be specifically configured to:

[0146] When the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, the indicator identifiers of the M indicators and the data format adapted to the front-end application are used to generate node data of M nodes, where one node is used to record one indicator;

[0147] Generate edge data for N directed edges using attribute information of the N trusted indicator pairs in the first metadata and a data format compatible with the front-end application; the edge data for any directed edge includes a weight for the directed edge, where the weight is equal to an effect gain of the trusted indicator pair corresponding to the directed edge;

[0148] The node data of the M nodes and the edge data of the N directed edges are integrated to obtain second metadata.

[0149] In another embodiment, when the processing unit 502 is configured to generate node data of the M nodes using the indicator identifiers of the M indicators and a data format adapted to the front-end application, it may be specifically configured to:

[0150] Determining M nodes for recording the M indicators, determining the indicator identifier of each indicator as the node identifier of the corresponding node, and determining the display coordinates of each node;

[0151] Determine the level of each node based on the attribute information of the N trusted indicator pairs; wherein the level of the node corresponding to the cause indicator in any trusted indicator pair is lower than the level of the node corresponding to the result indicator;

[0152] The node data of the M nodes are generated based on the node identifiers, corresponding display coordinates and corresponding levels of the M nodes using a data format adapted to the front-end application.

[0153] In another embodiment, when the acquisition unit 501 is used to acquire target user data of multiple users, it can be specifically used to:

[0154] Determine Q data sources, where Q is an integer greater than 1; each data source stores original user data of at least one user, where the original user data of each user is generated based on the usage behavior of each user with respect to the target object;

[0155] According to the adaptation relationship between the data source and the data processing tool, determine the target data processing tool adapted to the qth data source; where q∈[1,Q];

[0156] Calling the target code to read the original user data of K users in the qth data source through the target data processing tool, where K is a positive integer;

[0157] After reading the original user data of the K users, the original user data of the K users are encapsulated using a preset format to obtain the target user data of the K users.

[0158] In another embodiment, when the acquisition unit 501 is used to call the target code to read the original user data of K users in the qth data source through the target data processing tool, it can be specifically used to:

[0159] Determine the data storage format of the qth data source;

[0160] If the data storage format is the first storage format, calling the target code to read the original user data of K users from the qth data source through the target data processing tool;

[0161] If the data storage format is the second storage format, calling the target code to move the original user data of the K users from the qth data source to the designated data source through the target data processing tool, so as to store the original user data of the K users in the data storage format of the designated data source; and calling the target code to read the original user data of the K users from the designated data source through the data processing tool adapted to the designated data source;

[0162] Among them, the first storage form refers to: a storage form that is compatible with the reading and packaging capability of the target code; the second storage form refers to: a storage form that is not compatible with the reading and packaging capability of the target code; the specified data source refers to: the data source of the first storage form.

[0163] In another embodiment, the second metadata includes node data of M nodes and edge data of N directed edges; the node data of any node includes at least: a node identifier of the node and corresponding display coordinates; the edge data of any directed edge includes at least: a weight of the directed edge and node identifiers of two nodes connected by the directed edge;

[0164] Accordingly, when displaying the indicator cause-effect diagram based on the second metadata in a user interface displayed by the front-end application, the processing unit 502 may be specifically configured to:

[0165] Obtaining display coordinates of each node from the node data of the M nodes; and drawing each node using a first style based on the display coordinates of each node; wherein, when drawing each node using the first style, a node identifier of each node is hidden;

[0166] Draw an nth edge between two nodes indicated by node identifiers in an nth edge data, and draw the weight in the nth edge data at a target position, the target position being associated with a drawing position of the nth edge; where n∈[1,N];

[0167] After N edges and the weight of each edge are drawn, the indicator cause-effect diagram is obtained; and the indicator cause-effect diagram is displayed in the user interface displayed by the front-end application.

[0168] In another embodiment, the processing unit 502 may be further configured to:

[0169] During the process of displaying the indicator cause-effect diagram, if a node style switching operation for the indicator cause-effect diagram is detected, the display style of each node in the indicator cause-effect diagram is switched from the first style to the second style;

[0170] When the second style is used to draw each node, the node identifier of each node is displayed.

[0171] In another embodiment, the node data of any node further includes: the level of any node;

[0172] The display color of each node in the indicator cause-effect diagram is determined according to the level of each node; nodes at different levels have different display colors.

[0173] In another embodiment, the indicator cause-effect diagram has a theme color; accordingly, the processing unit 502 may also be configured to:

[0174] During the display of the indicator cause-effect diagram, if a theme color switching operation for the indicator cause-effect diagram is detected, the theme color of the indicator cause-effect diagram is switched from a current theme color to a target theme color;

[0175] The current theme color refers to the theme color presented by the indicator cause-and-effect diagram when the theme color switching operation is detected.

[0176] In another embodiment, the processing unit 502 may be further configured to:

[0177] When displaying the indicator cause-effect diagram, displaying an information input area in the user interface;

[0178] If target information is obtained in the information input area, target nodes and target edges matching the target information are screened out from the indicator causal graph;

[0179] In the indicator causal graph displayed on the user interface, the target node and the target edge are kept displayed, and nodes other than the target node and edges other than the target edge are canceled from display.

[0180] In another embodiment, the processing unit 502 may be further configured to:

[0181] During the display of the indicator cause-effect diagram, if there is a move operation for any element in the indicator cause-effect diagram, the element is moved according to the move operation to update the indicator cause-effect diagram; the element includes: the weight of any node or any edge;

[0182] generating new second metadata according to the updated indicator causal graph, and converting the new second metadata into new first metadata;

[0183] The new first metadata is stored so that when the indicator cause-effect diagram of the target object needs to be displayed next time, the new first metadata is converted into the new second metadata, and the updated indicator cause-effect diagram is displayed based on the new second metadata.

[0184] According to one embodiment of the present application, Figure 2 or Figure 3 Each step involved in the method shown can be performed by Figure 5 The data processing apparatus shown in FIG. 1 is executed by each unit. For example, Figure 2 The step S201 shown in FIG. Figure 5 The acquisition unit 501 shown in FIG is executed, and steps S202-S205 can be performed by Figure 5 , the processing unit 502 shown in ; for example, Figure 3The steps S301-S304 shown can be performed by Figure 5 The acquisition unit 501 shown in FIG is executed, and steps S305-S308 can be performed by Figure 5 , and so on.

[0185] According to another embodiment of the present application, Figure 5 The various units in the data processing apparatus shown can be separately or all merged into one or several other units to constitute, or a certain unit (or units) therein can also be split into multiple smaller units in function to constitute, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above-mentioned units are divided based on logical functions. In practical applications, the function of a unit can also be realized by multiple units, or the function of multiple units can be realized by one unit. In other embodiments of the present application, other units can also be included based on the data processing apparatus. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented by the collaboration of multiple units.

[0186] According to another embodiment of the present application, the program can be executed by running on a general computing device such as a computer including a central processing unit (CPU), a random access memory (RAM), a read-only memory (ROM) and other processing elements and storage elements. Figure 2 or Figure 3 A computer program (including program code) for each step involved in the corresponding method shown in Figure 5 The data processing device shown in the embodiment of the present application is used to implement the data processing method of the embodiment of the present application. The computer program can be recorded on a computer-readable recording medium, for example, and loaded into the above-mentioned computing device through the computer-readable recording medium and run therein.

[0187] In an embodiment of the present application, the target user data of multiple users using the target object can be obtained first. The target user data of any user includes the indicator data of any user under each indicator. Since the target user data of each user is generated based on the usage behavior of each user for the target object, the target user data of each user has a high guiding value for analyzing the relationship between the various indicators of the target object. In this way, when the causal relationship between the various indicators configured for the target object is automatically analyzed and quantified based on the indicator data of each user under each indicator, a more accurate analysis and quantification result can be obtained. Then, a first metadata can be generated based on the analysis and quantification result, and the first metadata can be converted into a second metadata with a data format that is compatible with the front-end application, so that the indicator causal graph of the target object can be displayed in the user interface based on the second metadata. The above process automatically discovers, quantifies and draws the causal relationship between the various indicators of the target object, and can realize the intuitive display of the causal relationship between the indicators for the technicians by displaying the indicator causal graph, which facilitates the technicians to search for indicators and improves the efficiency and accuracy of indicator search. Moreover, since the indicator causal graph uses directed edges to connect the two nodes corresponding to two indicators with a causal relationship, technical personnel can clearly understand which indicator is the "cause" and which indicator is the "effect" through the direction of the directed edge, which further facilitates technical personnel to quickly locate the indicators they need and further improve the efficiency of indicator search.

[0188] Based on the description of the above method embodiment and apparatus embodiment, the present application embodiment also provides a computer device. Figure 6 , the computer device at least includes a processor 601, an input interface 602, an output interface 603 and a computer storage medium 604. Among them, the processor 601, input interface 602, output interface 603 and computer storage medium 604 in the computer device can be connected via a bus or other means. The computer storage medium 604 can be stored in the memory of the computer device, and the computer storage medium 604 is used to store a computer program, and the computer program includes program instructions. The processor 601 is used to execute the program instructions stored in the computer storage medium 604. The processor 601 (or CPU (Central Processing Unit)) is the computing core and control core of the computer device, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to realize the corresponding method flow or corresponding function.

[0189] In one embodiment, the processor 601 described in the embodiment of the present application can be used to perform a series of data processing, specifically including: obtaining target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user for the target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes: indicator data of any user under each indicator; based on the indicator data of each user under each indicator, analyzing and quantifying the causal relationship between each indicator in the M indicators to obtain analysis and quantification results; generating first metadata based on the analysis and quantification results, where the first metadata includes attribute information of N trusted indicator pairs, where N is a positive integer; a trusted indicator pair refers to an indicator pair consisting of two indicators with the causal relationship; when the indicator causal graph of the target object needs to be displayed in a front-end application, converting the first metadata into second metadata, where the data format of the second metadata is compatible with the front-end application; displaying the indicator causal graph based on the second metadata in the user interface displayed by the front-end application; the indicator causal graph includes M nodes and N directed edges, where one node records one indicator, and one directed edge is used to connect two nodes corresponding to a trusted indicator pair, and so on.

[0190] The embodiment of the present application also provides a computer storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer storage medium here can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer storage medium provides a storage space that stores the operating system of the computer device. In addition, one or more instructions suitable for being loaded and executed by the processor 601 are also stored in the storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage; optionally, it can also be at least one computer storage medium located away from the aforementioned processor.

[0191] In one embodiment, the processor 601 may load and execute one or more instructions stored in the computer storage medium to implement the above-mentioned Figure 2 or Figure 3 The corresponding steps of the method in the data processing method embodiment shown in FIG. 1 are as follows; in a specific implementation, one or more instructions in the computer storage medium are loaded by the processor 601 and executed as follows:

[0192] Obtain target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user with respect to a target object; the target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes indicator data of the user under each indicator;

[0193] Analyze and quantify the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain analysis and quantification results;

[0194] generating first metadata based on the analysis and quantification results, the first metadata including attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship;

[0195] When the indicator cause-effect diagram of the target object needs to be displayed in a front-end application, the first metadata is converted into second metadata, where the data format of the second metadata is compatible with the front-end application;

[0196] In the user interface displayed by the front-end application, the indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

[0197] In one embodiment, when analyzing and quantifying the causal relationship between each of the M indicators based on the indicator data of each user under each indicator to obtain the analysis and quantification results, the one or more instructions may be loaded and specifically executed by the processor 601:

[0198] Analyze the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain causal relationship data; the causal relationship data is used to indicate N credible indicator pairs, each credible indicator pair includes a cause indicator and a result indicator;

[0199] The causal relationship data and the target user data of the multiple users are used to quantify the effect gain of each trustworthy indicator pair; the effect gain of any trustworthy indicator pair is used to measure: when the cause indicator in any trustworthy indicator pair is optimized to a baseline degree, the degree to which the result indicator is optimized based on the optimization of the cause indicator;

[0200] The effect gain of each pair of credible indicators is added to the analysis quantification result.

[0201] In another embodiment, when generating the first metadata in the data format based on the analysis and quantification results, the one or more instructions may be loaded and specifically executed by the processor 601:

[0202] Constructing a data table, the data table including a plurality of storage locations; each storage location including at least the following fields: a cause indicator field, a result indicator field, and an effect gain field;

[0203] Acquire attribute information of an nth credible indicator pair based on the analysis and quantification results, where n∈[1,N]; the attribute information includes the following sub-information: an indicator identifier of a cause indicator in the nth credible indicator pair, an indicator identifier of a result indicator in the nth credible indicator pair, and an effect gain of the nth credible indicator pair;

[0204] Fill each sub-information in the attribute information of the nth trust indicator pair into each field of the nth storage position in the data table;

[0205] The data table filled with the attribute information of the N trustworthy indicator pairs is determined as the first metadata.

[0206] In another embodiment, when the indicator cause-effect graph of the target object needs to be displayed in a front-end application, when the first metadata is converted into the second metadata, the one or more instructions may be loaded and specifically executed by the processor 601:

[0207] When the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, the indicator identifiers of the M indicators and the data format adapted to the front-end application are used to generate node data of M nodes, where one node is used to record one indicator;

[0208] Generate edge data for N directed edges using attribute information of the N trusted indicator pairs in the first metadata and a data format compatible with the front-end application; the edge data for any directed edge includes a weight for the directed edge, where the weight is equal to an effect gain of the trusted indicator pair corresponding to the directed edge;

[0209] The node data of the M nodes and the edge data of the N directed edges are integrated to obtain second metadata.

[0210] In another embodiment, when the indicator identifiers of the M indicators and the data format adapted to the front-end application are used to generate the node data of the M nodes, the one or more instructions may be loaded and specifically executed by the processor 601:

[0211] Determining M nodes for recording the M indicators, determining the indicator identifier of each indicator as the node identifier of the corresponding node, and determining the display coordinates of each node;

[0212] Determine the level of each node based on the attribute information of the N trusted indicator pairs; wherein the level of the node corresponding to the cause indicator in any trusted indicator pair is lower than the level of the node corresponding to the result indicator;

[0213] The node data of the M nodes are generated based on the node identifiers, corresponding display coordinates and corresponding levels of the M nodes using a data format adapted to the front-end application.

[0214] In another embodiment, when acquiring target user data of multiple users, the one or more instructions may be loaded and specifically executed by the processor 601:

[0215] Determine Q data sources, where Q is an integer greater than 1; each data source stores original user data of at least one user, where the original user data of each user is generated based on the usage behavior of each user with respect to the target object;

[0216] According to the adaptation relationship between the data source and the data processing tool, determine the target data processing tool adapted to the qth data source; where q∈[1,Q];

[0217] Calling the target code to read the original user data of K users in the qth data source through the target data processing tool, where K is a positive integer;

[0218] After reading the original user data of the K users, the original user data of the K users are encapsulated using a preset format to obtain the target user data of the K users.

[0219] In another embodiment, when calling the target code to read the original user data of K users in the qth data source through the target data processing tool, the one or more instructions may be loaded and specifically executed by the processor 601:

[0220] Determine the data storage format of the qth data source;

[0221] If the data storage format is the first storage format, calling the target code to read the original user data of K users from the qth data source through the target data processing tool;

[0222] If the data storage format is the second storage format, calling the target code to move the original user data of the K users from the qth data source to the designated data source through the target data processing tool, so as to store the original user data of the K users in the data storage format of the designated data source; and calling the target code to read the original user data of the K users from the designated data source through the data processing tool adapted to the designated data source;

[0223] Among them, the first storage form refers to: a storage form that is compatible with the reading and packaging capability of the target code; the second storage form refers to: a storage form that is not compatible with the reading and packaging capability of the target code; the specified data source refers to: the data source of the first storage form.

[0224] In another embodiment, the second metadata includes node data of M nodes and edge data of N directed edges; the node data of any node includes at least: a node identifier of the node and corresponding display coordinates; the edge data of any directed edge includes at least: a weight of the directed edge and node identifiers of two nodes connected by the directed edge;

[0225] Accordingly, in the user interface displayed by the front-end application, when the indicator cause-effect diagram is displayed based on the second metadata, the one or more instructions may be loaded and specifically executed by the processor 601:

[0226] Obtaining display coordinates of each node from the node data of the M nodes; and drawing each node using a first style based on the display coordinates of each node; wherein, when drawing each node using the first style, a node identifier of each node is hidden;

[0227] Draw an nth edge between two nodes indicated by node identifiers in an nth edge data, and draw the weight in the nth edge data at a target position, the target position being associated with a drawing position of the nth edge; where n∈[1,N];

[0228] After N edges and the weight of each edge are drawn, the indicator cause-effect diagram is obtained; and the indicator cause-effect diagram is displayed in the user interface displayed by the front-end application.

[0229] In another implementation, the one or more instructions may also be loaded and specifically executed by the processor 601:

[0230] During the process of displaying the indicator cause-effect diagram, if a node style switching operation for the indicator cause-effect diagram is detected, the display style of each node in the indicator cause-effect diagram is switched from the first style to the second style;

[0231] When the second style is used to draw each node, the node identifier of each node is displayed.

[0232] In another embodiment, the node data of any node further includes: the level of any node;

[0233] The display color of each node in the indicator cause-effect diagram is determined according to the level of each node; nodes at different levels have different display colors.

[0234] In another embodiment, the indicator cause-effect diagram has a theme color; accordingly, the one or more instructions may also be loaded and specifically executed by the processor 601:

[0235] During the display of the indicator cause-effect diagram, if a theme color switching operation for the indicator cause-effect diagram is detected, the theme color of the indicator cause-effect diagram is switched from a current theme color to a target theme color;

[0236] The current theme color refers to the theme color presented by the indicator cause-and-effect diagram when the theme color switching operation is detected.

[0237] In another implementation, the one or more instructions may also be loaded and specifically executed by the processor 601:

[0238] When displaying the indicator cause-effect diagram, displaying an information input area in the user interface;

[0239] If target information is obtained in the information input area, target nodes and target edges matching the target information are screened out from the indicator causal graph;

[0240] In the indicator causal graph displayed on the user interface, the target node and the target edge are kept displayed, and nodes other than the target node and edges other than the target edge are canceled from display.

[0241] In another implementation, the one or more instructions may also be loaded and specifically executed by the processor 601:

[0242] During the display of the indicator cause-effect diagram, if there is a move operation for any element in the indicator cause-effect diagram, the element is moved according to the move operation to update the indicator cause-effect diagram; the element includes: the weight of any node or any edge;

[0243] generating new second metadata according to the updated indicator causal graph, and converting the new second metadata into new first metadata;

[0244] The new first metadata is stored so that when the indicator cause-effect diagram of the target object needs to be displayed next time, the new first metadata is converted into the new second metadata, and the updated indicator cause-effect diagram is displayed based on the new second metadata.

[0245] In an embodiment of the present application, the target user data of multiple users using the target object can be obtained first. The target user data of any user includes the indicator data of any user under each indicator. Since the target user data of each user is generated based on the usage behavior of each user for the target object, the target user data of each user has a high guiding value for analyzing the relationship between the various indicators of the target object. In this way, when the causal relationship between the various indicators configured for the target object is automatically analyzed and quantified based on the indicator data of each user under each indicator, a more accurate analysis and quantification result can be obtained. Then, a first metadata can be generated based on the analysis and quantification result, and the first metadata can be converted into a second metadata with a data format that is compatible with the front-end application, so that the indicator causal graph of the target object can be displayed in the user interface based on the second metadata. The above process automatically discovers, quantifies and draws the causal relationship between the various indicators of the target object, and can realize the intuitive display of the causal relationship between the indicators for the technicians by displaying the indicator causal graph, which facilitates the technicians to search for indicators and improves the efficiency and accuracy of indicator search. Moreover, since the indicator causal graph uses directed edges to connect the two nodes corresponding to two indicators with a causal relationship, technical personnel can clearly understand which indicator is the "cause" and which indicator is the "effect" through the direction of the directed edge, which further facilitates technical personnel to quickly locate the indicators they need and further improve the efficiency of indicator search.

[0246] It should be noted that, according to one aspect of the present application, a computer program product or computer program is also provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the above-mentioned Figure 2 or Figure 3 The data processing method shown in the embodiment is provided in various optional aspects.

[0247] Furthermore, it should be understood that what is disclosed above is only a preferred embodiment of the present application, and certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A data processing method, characterized in that: include: Obtain target user data of multiple users, where the target user data of any user is generated based on the usage behavior of any user for the target object; The target object is configured with M indicators, where M is an integer greater than 1; The target user data of any user includes: the indicator data of any user under each indicator; Analyze and quantify the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain analysis and quantification results; generating first metadata based on the analysis and quantification results, the first metadata including attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship; When the indicator cause-effect diagram of the target object needs to be displayed in a front-end application, the first metadata is converted into second metadata, where the data format of the second metadata is compatible with the front-end application; In the user interface displayed by the front-end application, the indicator causal graph is displayed based on the second metadata; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

2. The method according to claim 1, wherein The causal relationship between each indicator in the M indicators is analyzed and quantified based on the indicator data of each user under each indicator to obtain the analysis and quantification results, including: Analyze the causal relationship between each indicator in the M indicators based on the indicator data of each user under each indicator to obtain causal relationship data; the causal relationship data is used to indicate N credible indicator pairs, each credible indicator pair includes a cause indicator and a result indicator; The causal relationship data and the target user data of the multiple users are used to quantify the effect gain of each trustworthy indicator pair; the effect gain of any trustworthy indicator pair is used to measure: when the cause indicator in any trustworthy indicator pair is optimized to a baseline degree, the degree to which the result indicator is optimized based on the optimization of the cause indicator; The effect gain of each pair of credible indicators is added to the analysis quantification result.

3. The method according to claim 2, wherein The step of generating first metadata in a data format based on the analysis and quantification results includes: Constructing a data table, the data table including a plurality of storage locations; each storage location including at least the following fields: a cause indicator field, a result indicator field, and an effect gain field; Acquire attribute information of an nth credible indicator pair based on the analysis and quantification results, where n∈[1,N]; the attribute information includes the following sub-information: an indicator identifier of a cause indicator in the nth credible indicator pair, an indicator identifier of a result indicator in the nth credible indicator pair, and an effect gain of the nth credible indicator pair; Fill each sub-information in the attribute information of the nth trust indicator pair into each field of the nth storage position in the data table; The data table filled with the attribute information of the N trustworthy indicator pairs is determined as the first metadata.

4. The method according to claim 1, wherein When the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, converting the first metadata into second metadata includes: When the indicator cause-effect diagram of the target object needs to be displayed in the front-end application, the indicator identifiers of the M indicators and the data format adapted to the front-end application are used to generate node data of M nodes, where one node is used to record one indicator; Generate edge data for N directed edges using attribute information of the N trusted indicator pairs in the first metadata and a data format compatible with the front-end application; the edge data for any directed edge includes a weight for the directed edge, where the weight is equal to an effect gain of the trusted indicator pair corresponding to the directed edge; The node data of the M nodes and the edge data of the N directed edges are integrated to obtain second metadata.

5. The method according to claim 4, wherein The step of using the indicator identifiers of the M indicators and a data format adapted to the front-end application to generate node data of the M nodes includes: Determining M nodes for recording the M indicators, determining the indicator identifier of each indicator as the node identifier of the corresponding node, and determining the display coordinates of each node; Determine the level of each node based on the attribute information of the N trusted indicator pairs; wherein the level of the node corresponding to the cause indicator in any trusted indicator pair is lower than the level of the node corresponding to the result indicator; The node data of the M nodes are generated based on the node identifiers, corresponding display coordinates and corresponding levels of the M nodes using a data format adapted to the front-end application.

6. The method according to claim 1, wherein The acquiring target user data of multiple users includes: Determine Q data sources, where Q is an integer greater than 1; each data source stores original user data of at least one user, where the original user data of each user is generated based on the usage behavior of each user with respect to the target object; According to the adaptation relationship between the data source and the data processing tool, determine the target data processing tool adapted to the qth data source; where q∈[1,Q]; Calling the target code to read the original user data of K users in the qth data source through the target data processing tool, where K is a positive integer; After reading the original user data of the K users, the original user data of the K users are encapsulated using a preset format to obtain the target user data of the K users.

7. The method according to claim 6, wherein The calling target code reads the original user data of K users in the qth data source through the target data processing tool, including: Determine the data storage format of the qth data source; If the data storage format is the first storage format, calling the target code to read the original user data of K users from the qth data source through the target data processing tool; If the data storage format is the second storage format, calling the target code to move the original user data of the K users from the qth data source to the designated data source through the target data processing tool, so as to store the original user data of the K users in the data storage format of the designated data source; and calling the target code to read the original user data of the K users from the designated data source through the data processing tool adapted to the designated data source; Among them, the first storage form refers to: a storage form that is compatible with the reading and packaging capability of the target code; the second storage form refers to: a storage form that is not compatible with the reading and packaging capability of the target code; the specified data source refers to: the data source of the first storage form.

8. The method according to claim 1, wherein The second metadata includes node data of M nodes and edge data of N directed edges; The node data of any node includes at least: the node identifier of the node and the corresponding display coordinates; the edge data of any directed edge includes at least: the weight of the directed edge and the node identifiers of the two nodes connected by the directed edge; Displaying the indicator cause-effect diagram based on the second metadata in a user interface displayed by the front-end application includes: Obtaining display coordinates of each node from the node data of the M nodes; and drawing each node using a first style based on the display coordinates of each node; wherein, when drawing each node using the first style, a node identifier of each node is hidden; Draw an nth edge between two nodes indicated by node identifiers in an nth edge data, and draw the weight in the nth edge data at a target position, the target position being associated with a drawing position of the nth edge; where n∈[1,N]; After N edges and the weight of each edge are drawn, the indicator cause-effect diagram is obtained; and the indicator cause-effect diagram is displayed in the user interface displayed by the front-end application.

9. The method according to claim 8, wherein The method further comprises: During the process of displaying the indicator cause-effect diagram, if a node style switching operation for the indicator cause-effect diagram is detected, the display style of each node in the indicator cause-effect diagram is switched from the first style to the second style; When the second style is used to draw each node, the node identifier of each node is displayed.

10. The method according to claim 8 or 9, characterized in that The node data of any node further includes: the level of any node; The display color of each node in the indicator cause-effect diagram is determined according to the level of each node; nodes at different levels have different display colors.

11. The method according to claim 1, wherein The indicator cause-effect diagram has a theme color; and the method further includes: During the display of the indicator cause-effect diagram, if a theme color switching operation for the indicator cause-effect diagram is detected, the theme color of the indicator cause-effect diagram is switched from a current theme color to a target theme color; The current theme color refers to the theme color presented by the indicator cause-and-effect diagram when the theme color switching operation is detected.

12. The method according to claim 1, wherein The method further comprises: When displaying the indicator cause-effect diagram, displaying an information input area in the user interface; If target information is obtained in the information input area, target nodes and target edges matching the target information are screened out from the indicator causal graph; In the indicator causal graph displayed on the user interface, the target node and the target edge are kept displayed, and nodes other than the target node and edges other than the target edge are canceled from display.

13. The method according to claim 1, wherein The method further comprises: During the display of the indicator cause-effect diagram, if there is a move operation for any element in the indicator cause-effect diagram, the element is moved according to the move operation to update the indicator cause-effect diagram; the element includes: the weight of any node or any edge; generating new second metadata according to the updated indicator causal graph, and converting the new second metadata into new first metadata; The new first metadata is stored so that when the indicator cause-effect diagram of the target object needs to be displayed next time, the new first metadata is converted into the new second metadata, and the updated indicator cause-effect diagram is displayed based on the new second metadata.

14. A data processing device, characterized in that: include: an acquisition unit, configured to acquire target user data of a plurality of users, wherein the target user data of any user is generated based on the usage behavior of any user for the target object; The target object is configured with M indicators, where M is an integer greater than 1; the target user data of any user includes: the indicator data of any user under each indicator; a processing unit, configured to analyze and quantify the causal relationship between each of the M indicators based on the indicator data of each user under each indicator, and obtain an analysis and quantification result; The processing unit is further configured to generate first metadata based on the analysis and quantification results, wherein the first metadata includes attribute information of N credible indicator pairs, where N is a positive integer; a credible indicator pair is an indicator pair consisting of two indicators having the causal relationship; The processing unit is further configured to convert the first metadata into second metadata when displaying the indicator cause-effect diagram of the target object in a front-end application, wherein the data format of the second metadata is compatible with the front-end application; The processing unit is also used to display the indicator causal graph based on the second metadata in the user interface displayed by the front-end application; the indicator causal graph includes M nodes and N directed edges, one node records one indicator, and one directed edge is used to connect the two nodes corresponding to a trusted indicator pair.

15. A computer device comprising an input interface and an output interface, characterized in that: Also includes: a processor adapted to implement one or more instructions; as well as, A computer storage medium storing one or more instructions, wherein the one or more instructions are suitable for being loaded by the processor and executing the data processing method according to any one of claims 1 to 13.

Images