Index attribution method and device, electronic equipment and readable medium

Abstract

The invention discloses an index attribution method and device, electronic equipment and a readable medium. In the method, in response to an operation on an attribution entry corresponding to a target index on a display page, attribution data of the target index is obtained, an attribution page of the target index is displayed, the attribution page is used for presenting the attribution data, and the attribution data is used for indicating an association relationship between factors influencing the target index and the target index. The attribution data comprises attribution numerical values corresponding to factors influencing the target index, and the attribution numerical values of the factors are determined based on the disassembling mode of the factors relative to the target index and the influence of the factors on the change of the target index. Thus, through the public control layout of the attribution entry corresponding to each index designed on the data analysis platform, a user only needs to operate the attribution entry on the display page to obtain the attribution page bearing the accurate attribution data, so that the user can quickly, accurately and comprehensively obtain the attribution data of the target index.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an indicator attribution method, device, electronic device and readable medium. Background Art

[0002] In the operation of an enterprise, a lot of indicator data will be generated. Indicator data can be understood as the value of the indicator. The trend of the indicator value change usually reflects the situation of the operation. Therefore, analyzing the reasons for the change of the indicator value (i.e. indicator attribution) can find the various factors that cause the change in the operation. Accurate and comprehensive indicator attribution can provide a reliable basis for the enterprise to make targeted operation decisions.

[0003] At present, there is an attribution module corresponding to each indicator on the data analysis platform. The attribution module displays the attribution result of the corresponding indicator. Each attribution module is displayed independently on the display page. In order to cover more indicators, the number of attribution modules continues to increase, and the display page of the data analysis platform becomes longer and longer. If the user wants to know the attribution result of a certain indicator, he needs to search for the attribution module corresponding to the indicator from the numerous attribution modules on the very long display page, which leads to poor user experience and waste of time.

[0004] Based on this, there is an urgent need to provide a solution that facilitates users to perform indicator attribution on the data analysis platform, so that users can quickly, accurately and comprehensively obtain the attribution results of the indicators they want to know. Summary of the invention

[0005] In order to solve the above technical problems, the present application provides an indicator attribution method, device, electronic device and readable medium, which can quickly, accurately and comprehensively obtain the attribution results of the desired indicators on a data analysis platform.

[0006] In order to achieve the above objectives, the technical solutions provided by this application are as follows:

[0007] In a first aspect, the present application provides an indicator attribution method, comprising:

[0008] In response to an operation on an attribution entry corresponding to a target indicator on a display page, attribution data of the target indicator is obtained, the display page including an attribution entry corresponding to each of a plurality of indicators, the attribution data being used to indicate an association relationship between a factor affecting the target indicator and the target indicator, the attribution data including an attribution value corresponding to the factor affecting the target indicator, the attribution value of the factor being determined based on a disassembly method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator;

[0009] An attribution page for the target indicator is displayed, wherein the attribution page is used to present the attribution data.

[0010] In a possible implementation, the display page includes a first area, a second area, and a third area, the first area includes a core attribution conclusion of at least one preconfigured core indicator and a first control corresponding to each core indicator in the at least one core indicator, the at least one core indicator includes the target indicator, the second area includes a plurality of indicator cards, each indicator card in the plurality of indicator cards includes a second control, and the third area includes a trend chart of a single indicator and / or multiple indicators;

[0011] Among them, the attribution entrance is any one of the following entrances: any point on the trend chart, the first control or the second control.

[0012] In a possible implementation, the attribution data includes detailed data of the target indicator, and the attribution page displaying the target indicator includes:

[0013] The first page is displayed, and the first page displays detailed data of the target indicator.

[0014] As an example, the attribution data also includes the core attribution conclusion of the target indicator, the attribution page includes a second page in addition to the first page, and the attribution page displaying the target indicator includes:

[0015] Displaying the second page, where the second page includes the core attribution conclusion of the target indicator and a third control;

[0016] In response to an operation on the third control on the second page, the first page is displayed.

[0017] In a possible implementation, the first page includes a fourth area, a fifth area, and a sixth area, wherein:

[0018] The fourth area includes the filter condition configuration options of the target indicator and the trend graph corresponding to the target indicator under the configured filter condition;

[0019] The fifth area includes an attribution tree, the root node of the attribution tree is the target indicator, the leaf nodes of the attribution tree include indicator nodes and / or dimension nodes, the indicator nodes are factors affecting the target indicator obtained by decomposing the target indicator, the dimension nodes are factors affecting the target indicator obtained by decomposing the target indicator, each leaf node has a corresponding attribution value, the attribution value of the leaf node is determined based on the decomposition method of the leaf node relative to the target indicator and the influence of the leaf node on the change of the target indicator, the link between the nodes is configured with a corresponding influence condition, and the influence condition is used to determine whether the detailed data of the backend node of the link is displayed;

[0020] The sixth area includes detailed data of each node on the path corresponding to the selected node in the attribution tree, and the path corresponding to the selected node is the path from the root node to the selected node.

[0021] In a possible implementation, the method further includes:

[0022] In response to an operation on a target node of the attribution tree on the first page, obtaining attribution algorithms corresponding to nodes on a path corresponding to the target node and influencing conditions corresponding to links on the path corresponding to the target node;

[0023] Based on the attribution values ​​corresponding to the nodes on the path corresponding to the target node, perform attribution analysis on the corresponding nodes to obtain detailed data of the nodes on the path corresponding to the target node;

[0024] Determine whether the detailed data of two nodes on each link on the path corresponding to the target node meets the influence condition of the corresponding link, and if so, determine to display the detailed data of the back-end node in the link;

[0025] In the sixth area, detailed data of the nodes corresponding to the factors on the path corresponding to the target node are displayed, and the factors are indicators corresponding to the nodes determined to be displayed on the path corresponding to the target node.

[0026] In one possible implementation,

[0027] The method for determining the attribution value corresponding to the indicator node includes at least one of the following algorithms: a rate of change algorithm, a factor ratio algorithm, or a target value difference algorithm;

[0028] The method for determining the attribution value corresponding to the dimension node includes at least one of the following algorithms: a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm or an occupancy rate algorithm.

[0029] In a possible implementation, the influencing condition includes any one of the following conditions:

[0030] Whether the absolute value of the detailed data of the front-end node of the corresponding link belongs to the preset target interval, and the front-end node is the indicator node;

[0031] Whether the front-end node of the corresponding link belongs to the preset target number of nodes with the largest value of detail data, the front-end node is a dimension node.

[0032] In a second aspect, the present application also provides an indicator attribution device, including:

[0033] A first acquisition unit is configured to acquire attribution data of the target indicator in response to an operation on an attribution entry corresponding to the target indicator on a display page, wherein the display page includes an attribution entry corresponding to each of a plurality of indicators, the attribution data is used to indicate an association relationship between a factor affecting the target indicator and the target indicator, the attribution data includes an attribution value corresponding to the factor affecting the target indicator, and the attribution value of the factor is determined based on a disassembly method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator;

[0034] The first display unit is used to display the attribution page of the target indicator, and the attribution page is used to present the attribution data.

[0035] In a possible implementation, the display page includes a first area, a second area, and a third area, the first area includes a core attribution conclusion of at least one preconfigured core indicator and a first control corresponding to each core indicator in the at least one core indicator, the at least one core indicator includes the target indicator, the second area includes a plurality of indicator cards, each indicator card in the plurality of indicator cards includes a second control, and the third area includes a trend chart of a single indicator and / or multiple indicators;

[0036] Among them, the attribution entrance is any one of the following entrances: any point on the trend chart, the first control or the second control.

[0037] In a possible implementation, the attribution data includes detailed data of the target indicator, and the first display unit is specifically used to:

[0038] The first page is displayed, and the first page displays detailed data of the target indicator.

[0039] As an example, the attribution data further includes a core attribution conclusion of the target indicator, the attribution page includes a second page in addition to the first page, and the first display unit includes:

[0040] A first display subunit, configured to display the second page, wherein the second page includes a core attribution conclusion of the target indicator and a third control;

[0041] The second display subunit is used to display the first page in response to an operation on the third control on the second page.

[0042] In a possible implementation, the first page includes a fourth area, a fifth area, and a sixth area, wherein:

[0043] The fourth area includes the filter condition configuration options of the target indicator and the trend graph corresponding to the target indicator under the configured filter condition;

[0044] The fifth area includes an attribution tree, the root node of the attribution tree is the target indicator, the leaf nodes of the attribution tree include indicator nodes and / or dimension nodes, the indicator nodes are factors affecting the target indicator obtained by decomposing the target indicator, the dimension nodes are factors affecting the target indicator obtained by decomposing the target indicator, each leaf node has a corresponding attribution value, the attribution value of the leaf node is determined based on the decomposition method of the leaf node relative to the target indicator and the influence of the leaf node on the change of the target indicator, the link between the nodes is configured with a corresponding influence condition, and the influence condition is used to determine whether the detailed data of the backend node of the link is displayed;

[0045] The sixth area includes detailed data of each node on the path corresponding to the selected node in the attribution tree, and the path corresponding to the selected node is the path from the root node to the selected node.

[0046] In a possible implementation manner, the device further includes:

[0047] A second acquisition unit is configured to obtain, in response to an operation on a target node of the attribution tree on the first page, an attribution algorithm corresponding to each node on a path corresponding to the target node and an influencing condition corresponding to each link on the path corresponding to the target node;

[0048] A third acquisition unit is used to perform attribution analysis on the corresponding node based on the attribution value corresponding to each node on the path corresponding to the target node, and obtain detailed data of each node on the path corresponding to the target node;

[0049] A determination unit, used to determine whether the detailed data of two nodes on each link on the path corresponding to the target node meets the influence condition of the corresponding link, and if so, determine to display the detailed data of the back-end node in the link;

[0050] The second display unit is used to display detailed data of the nodes corresponding to the factors on the path corresponding to the target node in the sixth area, where the factors are indicators corresponding to the nodes determined to be displayed on the path corresponding to the target node.

[0051] In one possible implementation,

[0052] The method for determining the attribution value corresponding to the indicator node includes at least one of the following algorithms: a rate of change algorithm, a factor ratio algorithm, or a target value difference algorithm;

[0053] The method for determining the attribution value corresponding to the dimension node includes at least one of the following algorithms: a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm or an occupancy rate algorithm.

[0054] In a possible implementation, the influencing condition includes any one of the following conditions:

[0055] Whether the absolute value of the detailed data of the front-end node of the corresponding link belongs to the preset target interval, and the front-end node is the indicator node;

[0056] Whether the front-end node of the corresponding link belongs to the preset target number of nodes with the largest value of detail data, the front-end node is a dimension node.

[0057] It should be noted that the specific implementation method of the device and the technical effects achieved can refer to the relevant description of the method provided in the first aspect or any implementation method of the first aspect.

[0058] In a third aspect, the present application further provides an electronic device, the electronic device comprising: a processor and a memory;

[0059] The memory is used to store instructions or programs;

[0060] The processor is used to execute the instructions or programs in the memory so that the electronic device executes the method provided by the first aspect or any one of the implementations of the first aspect.

[0061] In a fourth aspect, the present application further provides a readable medium, in which instructions or programs are stored. When the instructions or programs are executed on a processor, the processor executes the method provided in the first aspect or any one of the implementations of the first aspect.

[0062] Compared with the prior art, the embodiments of the present application have at least the following advantages:

[0063] In the technical solution provided by the present application, an attribution entry corresponding to each of a plurality of indicators is designed on a display page of a data analysis platform. When there is a need to attribute a target indicator, the method may include: in response to an operation on an attribution entry corresponding to a target indicator on a display page of the data analysis platform, obtaining attribution data of the target indicator, and displaying an attribution page of the target indicator, wherein the attribution page is used to present the attribution data, and the attribution data is used to indicate the correlation between factors affecting the target indicator and the target indicator, wherein the attribution data includes attribution values ​​corresponding to the factors affecting the target indicator, and the attribution values ​​of the factors are determined based on the decomposition method of the factor relative to the target indicator and the influence of the factor on the change of the target indicator. In this way, an attribution portal corresponding to each of the multiple indicators is designed on the display page of the data analysis platform. If the user wants to understand the attribution of the target indicator, he only needs to operate the attribution portal corresponding to the target indicator on the display page of the data analysis platform, and he can see the attribution page carrying the attribution data of the target indicator on the display page. The user does not need to spend time and effort to find out whether there is an attribution module corresponding to the target indicator among the numerous attribution modules on the current data analysis platform. The public control panel including the attribution portal corresponding to each of the multiple indicators on the improved data analysis platform provides the attribution function of each indicator. Moreover, the attribution data obtained after the attribution portal is triggered is determined by adaptive calculation based on the specific situation of the factors affecting the target indicator, and is more accurate attribution data for the target indicator. It can be seen that this method can facilitate users to quickly, accurately and comprehensively obtain the attribution data of the target indicator they want to know, and improves the user experience of the indicator attribution function on the data analysis platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0065] Figure 1 A flowchart of an indicator attribution method provided in an embodiment of the present application;

[0066] Figure 2 A schematic diagram of a display page of a data analysis platform provided in an embodiment of the present application;

[0067] Figure 3 A schematic diagram of a second page provided in an embodiment of the present application;

[0068] Figure 4A schematic diagram of a first page provided in an embodiment of the present application;

[0069] Figure 5 A schematic diagram of a flow chart of attribution analysis of a target indicator provided in an embodiment of the present application;

[0070] Figure 6 A schematic diagram of an example of indicator attribution provided in an embodiment of the present application;

[0071] Figure 7 A schematic diagram of the structure of an indicator attribution device 700 provided in an embodiment of the present application;

[0072] Figure 8 A schematic diagram of the structure of an electronic device 800 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0073] With the rise of mobile Internet, business forms are becoming more and more complex, and the demand for indicator attribution is increasing. The so-called indicator attribution refers to the process of accurately locating the key factors that play an important role in the fluctuation of a certain indicator when multiple factors work together to cause the fluctuation of the above indicators.

[0074] At present, although the data analysis platform provides the function of indicator attribution, the display page only displays the attribution modules of the developed indicators in sequence. Users can slide the display page up and down to locate the display page to the attribution module corresponding to the indicator they want to know and view the attribution data of the indicator. As the number of attribution modules for the developed indicators on the display page increases, the display page will become longer and longer, and it is more difficult for users to view the attribution indicators they want to know. Moreover, the attribution data displayed in each attribution module is relatively complicated. Users need to view the attribution module corresponding to the indicator for a long time to understand the key factors that cause the fluctuation of the indicator, which is not clear enough for users. Moreover, if the indicator that the user wants to know has not yet developed a corresponding attribution module, or the attribution module developed for the indicator does not meet the user's needs for the disassembly of the indicator, then technical personnel need to develop a new attribution module according to the user's needs. The development cycle will be relatively long, and users need to wait until the new attribution module is developed, which leads to poor user experience.

[0075] Based on this, in order to make the indicator attribution function of the data analysis platform more in line with the user's expectations, the embodiment of the present application designs a display page of the data analysis platform, adds a public control layout corresponding to the indicator attribution function, and the user can understand the indicator he wants to know by operating on the public control layout; on the other hand, for each factor affecting the target indicator, a method of calculating, analyzing and determining the attribution data adapted to it is designed, so that the obtained attribution data is more accurate. In specific implementation, the indicator attribution method provided by the embodiment of the present application may include: in response to the operation of the attribution entrance corresponding to the target indicator on the display page of the data analysis platform, obtaining the attribution data of the target indicator, and displaying the attribution page of the target indicator, the attribution page is used to present the attribution data, the attribution data is used to indicate the correlation between the factors affecting the target indicator and the target indicator, the attribution data includes the attribution value corresponding to the factor affecting the target indicator, and the attribution value of the factor is determined based on the disassembly method of the factor relative to the target indicator and the influence of the factor on the change of the target indicator.

[0076] In this way, an attribution entry corresponding to each of the multiple indicators is designed on the display page of the data analysis platform. When there is a need to attribution the target indicator, the user only needs to operate the attribution entry corresponding to the target indicator on the display page of the data analysis platform, and the attribution page carrying the attribution data of the target indicator can be seen on the display page. The user does not need to spend time and effort to search one by one among the numerous attribution modules on the current data analysis platform to see whether there is an attribution module corresponding to the target indicator. The public control layout including the attribution entry corresponding to each of the multiple indicators on the improved data analysis platform provides the attribution function of each indicator. Moreover, the attribution data obtained after the attribution entry is triggered is adaptively calculated, analyzed and determined based on the specific circumstances of the factors affecting the target indicator, and is more accurate attribution data for the target indicator. It can be seen that this method can facilitate users to quickly, accurately and comprehensively obtain the attribution data of the target indicator they want to know, and improves the user experience of the indicator attribution function on the data analysis platform.

[0077] It should be noted that the data analysis platform can be any platform that can provide data analysis functions to users, and the specific form can be a client or a website. The data analysis function provided by the data analysis platform at least includes an indicator attribution function. The display page of the data analysis platform can be a page corresponding to the indicator attribution function on the client or a web page corresponding to the indicator attribution function of the website.

[0078] It should be noted that the subject implementing the index attribution method may be the index attribution device provided in the embodiment of the present application, and the index attribution device may include a data analysis platform, or the index attribution device may be connected to the data analysis platform. The index attribution device may be carried in an electronic device or a functional module of an electronic device. The electronic device may be installed with a client corresponding to a wired data analysis platform, or may access a webpage corresponding to a data analysis platform.

[0079] Figure 1 A flow chart of an indicator attribution method provided in an embodiment of the present application. The method can be applied to an indicator attribution device, which can be, for example, Figure 7 The indicator attribution device 700 is shown. Since the indicator attribution device may include a data analysis platform or be connected to a data analysis platform, it can also be considered that the method is applied to a data analysis platform.

[0080] like Figure 1 As shown, the method may include, for example, the following S101 to S102:

[0081] S101, in response to an operation on an attribution entry corresponding to a target indicator on a display page, obtaining attribution data of the target indicator, wherein the display page includes an attribution entry corresponding to each indicator among a plurality of indicators, and the attribution data is used to indicate an association relationship between a factor affecting the target indicator and the target indicator, and the attribution data includes an attribution value corresponding to the factor affecting the target indicator, and the attribution value of the factor is determined based on a decomposition method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator.

[0082] Among them, the display page can be a page corresponding to the indicator attribution function on the interactive interface provided to the user by the data analysis platform. For example, the display page can be the homepage corresponding to the indicator attribution function on the data analysis platform, that is, when the user opens the data analysis platform and selects the indicator attribution function of the data analysis platform, the page first presented to the user by the interactive interface of the data analysis platform is the display page in S101. For another example, the display page can be a page corresponding to a certain type or scenario under the indicator attribution function on the data analysis platform, that is, when the user opens the data analysis platform and selects the indicator attribution function of the data analysis platform, he can continue to select the type or scenario of the indicator he wants to know, and the page presented to the user by the interactive interface of the data analysis platform is the display page in S101.

[0083] The display page has a public control panel for indicator attribution, which provides at least one attribution entry for various indicators, so that users can understand the attribution of any indicator they want to know on the public control panel. As the name suggests, the attribution entry is the entry that triggers the attribution of indicators. Users can enter the attribution page of the corresponding indicator by operating the attribution entry.

[0084] As an example, the presentation page may include a first area, a second area, and a third area.

[0085] The first area may include a core attribution conclusion of at least one pre-configured core indicator and a first control corresponding to each core indicator in the at least one core indicator. The first control may be understood as an attribution entry for the corresponding core indicator. Figure 2 In the schematic diagram of the display page shown, the first area may correspond to the core conclusion area, and the first control corresponds to the "view all" control after each core indicator in the core conclusion area. At least one core indicator may include the target indicator in S101.

[0086] It should be noted that the user can flexibly configure the core indicators included in the first area according to the focus requirements. When the core attribution conclusions of the core indicators configured by the user cannot be fully displayed in the first area, the user can view the core attribution conclusions of all configured core indicators through the slider, "Expand / Collapse" control or "View More Core Indicators" control; or, only the core attribution conclusions of the N core indicators with the largest changes in the configured core indicators can be displayed to the user in the first area, where N is an integer greater than or equal to 1.

[0087] It should be noted that the core attribution conclusion is only a simplified version of the attribution data of the core indicator, specifically the attribution conclusion corresponding to the key factors affecting the core indicator. By operating the first control corresponding to the core indicator, the overall attribution conclusion of the core indicator can be presented on the display page. The overall attribution conclusion can be carried on the first page described below. For details, please refer to the relevant description of the first page below.

[0088] The second area may include multiple indicator cards, each of which includes a second control, and the second control may be understood as an attribution entry corresponding to the indicator on the indicator card. Figure 2 In the schematic diagram of the display page shown, the second area may correspond to the core goals & trends area, and the second control corresponds to the "attribution" control in each indicator card in the core goals & trends area.

[0089] It should be noted that when the indicator cards of all indicators cannot be fully displayed in the second area, the user can view all indicator cards through a slider, a page selection control, an "expand / collapse" control or a "view more indicator cards" control; alternatively, only the M indicator cards with the largest changes in indicators among all the indicator cards can be displayed in the second area, where M is an integer greater than or equal to 1.

[0090] It should be noted that, in addition to the second control, the indicator card may also include: the indicator corresponding to the indicator card, the change range of the indicator, and other related information of the indicator. By operating the second control in the indicator card, the simplified version of the attribution data of the indicator (also called the core attribution conclusion) and / or the overall attribution conclusion of the indicator can be presented on the display page. The core attribution conclusion can be carried on the second page described below, and the overall attribution conclusion can be carried on the first page described below. For details, please refer to the following description of the second page and the first page.

[0091] Among them, the third area includes trend charts of single indicators and / or multiple indicators. Among them, the trend chart of a single indicator can be a trend chart corresponding to the indicator selected by the user through the drop-down menu. The trend chart of multiple indicators can refer to a trend chart including at least two indicators. The horizontal axis of the trend chart is time, and each point of the horizontal axis corresponds to the time of an observation period and the time of a base period. The base period can also be called a comparison period or comparison time, and the observation period can also be called an analysis period or statistical time; the vertical axis of the trend chart is the indicator value. The user can trigger the attribution of the indicator to the time interval corresponding to the point by clicking on a point on the trend chart, or can trigger the attribution of the indicator to the time interval corresponding to the area in the trend chart by selecting an area on the trend chart. The attribution time interval can also be adjusted through the time selection control provided on the subsequent first page or second page. Corresponding to Figure 2 In the schematic diagram of the display page shown, the third area can correspond to a single-indicator trend area and / or a multi-indicator trend area, and the attribution entry can correspond to a point on the trend chart or an area on the trend chart, or, a corresponding "attribution" control can be added to the legend on the trend chart for each indicator, and the "attribution" control serves as an attribution entry for the corresponding indicator.

[0092] It should be noted that in the third area, the trend graph of a single indicator includes two broken lines, one of which corresponds to the change in the indicator value in the base period, and the other corresponds to the change in the indicator value in the observation period. If the horizontal coordinate of the change in the indicator value in the observation period includes the future time, then the broken line also includes the change in the indicator forecast value corresponding to the future time.

[0093] It should be noted that, in the third area, the indicators specifically included in the multi-indicator trend chart can be flexibly configured by the user according to the needs of attention.

[0094] For example, the display page can be seen Figure 2 As shown, the display page includes a core conclusion area, a core goal & trend area, a single indicator trend area, and a multi-indicator trend area. Among them, the core conclusion area may include: relevant information 1 of core indicator 1 corresponding to core indicator 1, core attribution conclusion 1, and "View All" control 1, as well as relevant information 2 of core indicator 2 corresponding to core indicator 2, core attribution conclusion 2, and "View All" control 2. The core goal & trend area may include: indicator card 1, indicator card 2, ..., indicator card 5, each indicator card includes relevant information of the corresponding indicator and an "attribution" control. The single indicator trend area includes a drop-down menu control and a trend Figure 1 ,trend Figure 1 Used to present the trend of the indicator selected in the drop-down menu control during the observation period and the base period. The multi-indicator trend area includes trend Figure 2 ,trend Figure 2 Used to present the trends of multiple indicators during the observation period and the baseline period.

[0095] It should be noted that the relevant information of the indicator may include, but is not limited to: at least one of: the indicator name, the cut-off analysis time, the indicator time progress, or the year-on-year comparison with the base period.

[0096] As an example, when a user wants to know the attribution of an indicator (i.e., a target indicator) on a display page, he can operate any attribution entry corresponding to the target indicator on the display page, for example, he can click on the first control corresponding to the target indicator in the first area, for example, he can click on the second control corresponding to the target indicator in the second area, for example, he can also click on a point in the trend chart in the third area or select an area in the trend chart in the third area. Then, for the background of the data analysis platform, S101 can be executed to obtain the attribution data of the target indicator. The attribution data includes any data that can indicate the correlation between the factors affecting the target indicator and the target indicator, see the relevant instructions in S102 for details.

[0097] The target indicator can be decomposed in multiple layers, and the decomposition method of each layer can include indicator decomposition and / or dimension decomposition. For the decomposition of different indicators, the impact of the factors after decomposition on the change of the indicators before decomposition is also different, and the impact of the factors after decomposition on the change of the indicators before decomposition will also have a certain impact on the attribution analysis. Therefore, the attribution data obtained in S101 includes the attribution values ​​corresponding to each factor based on the target indicator. The attribution value of the factor is determined based on the decomposition method of the factor relative to the target indicator and the impact of the factor on the change of the target indicator. "Each factor based on the target indicator" corresponds to the indicators corresponding to each node determined and displayed on the path corresponding to the target node below. If the factor is decomposed into an indicator, then the factor corresponds to an indicator node, and the method for determining the attribution value corresponding to the indicator node may include at least one of the following algorithms: a rate of change algorithm, a factor ratio algorithm, or a target value difference algorithm; if the factor is decomposed into a dimension, then the factor corresponds to a dimension node, and the method for determining the attribution value corresponding to the dimension node may include at least one of the following algorithms: a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm, or an occupancy rate algorithm. As for which specific attribution value determination method is selected for the indicator node or dimension node, it can be further determined based on the impact of the corresponding factor on the change of the target indicator. It should be noted that for the relevant introduction to the specific attribution value determination method (hereinafter also referred to as the attribution algorithm), please refer to the description of the relevant part below.

[0098] S102, displaying an attribution page of the target indicator, wherein the attribution page is used to present the attribution data.

[0099] Among them, the attribution data may include detailed data of the target indicator and the core attribution conclusion of the target indicator.

[0100] As an example, S102 may include: displaying a first page, where the first page displays detailed data of the target indicator. The first page may be a "target indicator attribution details" page displayed in the form of a drawer, a pop-up window, a full screen, or an embedded card based on the display page. If the first page is displayed in the form of a drawer or an embedded card, then S102 may include: displaying the first page on the display page. The following description takes the first page displayed in the form of a drawer as an example.

[0101] For example, when a user clicks on a first control corresponding to a target indicator in a first area on a display page, the first page is displayed on the display page. For another example, when a user clicks on a second control on an indicator card corresponding to a target indicator in a second area on a display page, the first page is displayed on the display page. For another example, when a user clicks on a midpoint in a trend graph of a target indicator on a display page, or selects an area in a trend graph of a target indicator on a display page, the first page is displayed on the display page.

[0102] In some possible implementations, the first page may include a fourth area, a fifth area, and a sixth area.

[0103] The fourth area may include filter condition configuration options for the target indicator and a trend chart corresponding to the target indicator under the configured filter condition. The filter condition configuration options may include time configuration options for the base period, time configuration options for the observation period, and a control for "adding filter conditions." In addition, the filter condition configuration options may also include attribution time unit options provided by a drop-down menu control. The attribution time units available for selection provided by the drop-down menu control include: daily average, weekly average, monthly average, annual average, etc.

[0104] Among them, the fifth area may include an attribution tree (also known as an analysis tree), the root node of which is the target indicator, and the leaf nodes of the attribution tree include indicator nodes and / or dimension nodes, the indicator nodes are factors affecting the target indicator obtained by decomposing the target indicator, and the dimension nodes are factors affecting the target indicator obtained by decomposing the target indicator. In the background of the data analysis platform, a corresponding attribution algorithm is configured for each leaf node, and the attribution algorithm is used to determine the attribution value of the corresponding leaf node, and the attribution value of the leaf node is determined based on the decomposition method of the leaf node relative to the target indicator and the influence of the leaf node on the change of the target indicator; and corresponding influence conditions (also known as attribution conditions or threshold judgment conditions) are configured for the links between nodes, and the influence conditions are used to determine whether the detailed data of the back-end node of the link is displayed.

[0105] Indicator decomposition can be understood as decomposing the target indicator into the product or sum of multiple different sub-indicators. For example, the target indicator is A. Since A=B*C, the indicator nodes that can be obtained by decomposing the target indicator A include B and C.

[0106] Dimension decomposition can be understood as decomposing the target indicator according to a certain dimension into the sum of multiple different dimension values ​​under the dimension. For example, if the target indicator is A, the dimensions can include B, C, D, etc. Taking the decomposition of B dimension as an example, the dimension nodes obtained by decomposing A can include: B1, B2 and B3, so it can be understood as A=B1+B2+B3; taking the decomposition of C dimension as an example, the dimension nodes obtained by decomposing A can include: C1 and C1, so it can be understood as A=C1+C2.

[0107] The sixth area may include detailed data of each node on the path corresponding to the selected node in the attribution tree, and the path corresponding to the selected node is the path from the root node to the selected node. When no node in the attribution tree is selected, the sixth area may display the core attribution conclusion of the target indicator by default. The content in the sixth area may be presented in the form of text, table or figure.

[0108] When the attribution page is the first page, the method may further include: S11, in response to the operation of the target node of the attribution tree on the first page, obtaining the attribution algorithm corresponding to each node on the path corresponding to the target node and the influence condition corresponding to each link on the path corresponding to the target node; S12, based on the attribution value corresponding to each node on the path corresponding to the target node, performing attribution analysis on the corresponding node to obtain the detailed data of each node on the path corresponding to the target node; S13, judging whether the detailed data of two nodes on each link on the path corresponding to the target node meets the influence condition of the corresponding link, if so, determining to display the detailed data of the back-end node in the link, if not, determining not to display the detailed data of the back-end node in the link and each subsequent jump node; S14, after completing the above S13 judgment for each link on the path corresponding to the target node, displaying the detailed data of each node determined to be displayed on the path corresponding to the target node in the sixth area. Among them, the attribution value corresponding to each node on the path corresponding to the target node in S13 can be a value determined based on the attribution algorithm corresponding to each node on the path corresponding to the target node.

[0109] Among them, the attribution algorithm corresponding to the indicator node includes at least one of the following algorithms: change rate algorithm, decomposition factor proportion algorithm or target value difference algorithm; the attribution algorithm corresponding to the dimension node includes at least one of the following algorithms: single dimension value contribution rate algorithm, single dimension specified value contribution rate algorithm, multi-dimensional contribution rate algorithm, structural attribution algorithm or crowding out rate algorithm.

[0110] Among them, the influencing conditions include any one of the following conditions: whether the absolute value of the detail data of the front-end node of the corresponding link belongs to the preset target interval, and this influencing condition is applicable to the case where the front-end node is an indicator node; whether the front-end node of the corresponding link belongs to the preset target number of nodes with the largest detail data value, and this influencing condition is applicable to the case where the front-end node is a dimension node.

[0111] It should be noted that, in the first page, the fourth area may correspond to the following Figure 4 Corresponding indicator area, the fifth area can correspond to the following Figure 4 The corresponding attribution tree area, the sixth area can correspond to the following Figure 4 The corresponding detail attribution area.

[0112] As another example, S102 may include: displaying a second page on a display page, the second page including the core attribution conclusions of the target indicator and a third control; in response to an operation on the third control on the second page, displaying the first page on the display page. The third control is any control that can realize a jump from the second page to the first page, such as a "view more" or "view details" button. The second page may be a "simple attribution of target indicators" page displayed in the form of a drawer, pop-up window, full screen or embedded card on the basis of the display page. The core attribution conclusions on the second page may be displayed in the form of text, table or figure. The following description will be given by taking the second page as a drawer and the core attribution conclusions of the second page as an example.

[0113] For example, when a user clicks on the second control on the indicator card corresponding to the target indicator in the second area on the display page, the second page is first displayed on the display page. For another example, when a user clicks on the midpoint of the trend chart of the target indicator on the display page, or selects an area in the trend chart of the target indicator on the display page, the second page is displayed on the display page. When the second page is displayed on the display page, if the user wants to learn more about the attribution details of the target indicator, the user can click on the third control on the second page to display the first page on the display page. At this time, the second page may not be displayed, may be fully or partially covered by the first page, or may be displayed simultaneously with the first page.

[0114] For example, the second page can be found at Figure 3 shown. Figure 3 The second page may include: the name of the target indicator, the baseline period, the observation period, relevant information of the target indicator, the trend chart of the target indicator, and the core attribution conclusion of the target indicator.

[0115] For example, the first page can be found at Figure 4 shown. Figure 4In the figure, the first page may include: an indicator area, an attribution tree area, and a detailed attribution area. Among them, the indicator area may include: the name of the target indicator, the baseline period, the observation period, the relevant information of the target indicator, or the trend chart of the target indicator. The attribution tree area includes the attribution tree corresponding to the target indicator. The attribution tree may include sub-indicators obtained by decomposing the indicator (i.e., the factors mentioned above) and / or sub-indicators obtained by decomposing the dimension. In addition to displaying the indicator name, each node may also display the corresponding attribution algorithm name. The detailed attribution area may include: the selected attribution path and the detailed data corresponding to the attribution path (which can also be understood as the attribution result). Figure 4 In the attribution tree, the target indicator is search, and the first-level leaf node includes three dimensional nodes, namely: search channel, search intent, and search vertical category; under search intent, the second-level leaf node includes two groups of indicator nodes, one group of indicator nodes is search page views (PV) and orders per thousand searches (OPMS), and the other group of indicator nodes is orders per thousand searches and average price per search order; under search vertical category, the second-level leaf node is an indicator node, namely the search leaf category; under OPMS, the third-level leaf node includes three dimensional nodes, namely: Simpr, Sctr, and Scvr. When clicking on OPMS, the attribution path includes from "search" through "search intent" to "OPMS" and from "search" through "search intent" to "search PV". The detail data corresponding to the attribution path may include the detail data of the search intention. Whether it includes the detail data of the OPMS depends on whether the detail data of the search intention satisfies the influence condition of the link 1 from the search intention to the OPMS. If so, the detail data corresponding to the attribution path includes the detail data of the OPMS. Similarly, whether the detail data corresponding to the attribution path includes the detail data of the search PV depends on whether the detail data of the search intention satisfies the influence condition of the link 2 from the search intention to the search PV. If so, the detail data corresponding to the attribution path includes the detail data of the search PV.

[0116] It should be noted that Figure 3 and Figure 4 The observation period and the benchmark period can be set as required. The first page and / or the second page may also include an "add filter condition" control. The user can click the "add filter condition" control to configure the filter condition in the displayed pop-up window for adding filter conditions. Then, the target indicator displayed on the first page and / or the second page is the indicator after the filter condition is added to the original target indicator. For example, if the original target indicator is "search" and the added filter condition is "mall", then the target indicator is updated to "mall search".

[0117] It should be noted that Figure 4In the attribution tree, you can also double-click the root node of the attribution tree or click the "View Large Image" control in the attribution tree area to display the large image display page of the attribution tree in a pop-up window, allowing users to better understand the various influencing factors of the target indicator.

[0118] In order to more clearly illustrate the attribution data displayed in the attribution page in the embodiment of the present application, Figure 5 The flow shown illustrates the process of attribution analysis.

[0119] like Figure 5 As shown, the process of attribution analysis of target indicators can include:

[0120] S21, determining detailed data of the target indicator based on the target indicator on the root node of the attribution tree and the attribution value corresponding to the target indicator;

[0121] S22, determining whether the target indicator has a next-hop node and whether the detailed data of the target indicator satisfies the influence condition to the next-hop node. If yes, marking the target indicator as the current node and executing S23 and / or S24; otherwise, executing S26;

[0122] S23, obtaining an indicator node after the current node is decomposed into indicators, determining detailed data of the indicator node based on the indicator node and the attribution algorithm corresponding to the indicator node, recording the indicator node as the current node, and executing S25;

[0123] S24, obtaining a dimension node after the current node is dimensionally decomposed, determining detailed data of the dimension node based on the dimension node and the attribution algorithm corresponding to the dimension node, recording the dimension node as the current node, and executing S25;

[0124] S25, determining whether the current node has a next-hop node and whether the detailed data of the current node meets the influence condition to the next-hop node, if yes, returning to execute S23 and / or S24, otherwise, executing S26;

[0125] S26, generating attribution data of the target indicator.

[0126] In this way, a comprehensive and accurate attribution analysis of any node in the attribution tree can be achieved.

[0127] The construction of the attribution tree for each target indicator may specifically include: the first step is to decompose the target indicator as comprehensively as possible to obtain the architecture of the attribution tree of the target indicator, and the leaf nodes in the architecture may include indicator nodes and / or dimension nodes; the second step is to set corresponding influencing conditions for each line segment in the architecture of the attribution tree; the third step is to set corresponding attribution algorithms for each node in the architecture of the attribution tree.

[0128] In the second step, if the front end node of the line segment is an index node, then the influencing condition may include whether the absolute value of the detail data of the front end node belongs to the preset first interval; if the front end node of the line segment is a dimension node, then the influencing condition may include whether the front end node belongs to a preset target number of nodes with the largest detail data value, or whether the contribution rate of the front end node belongs to the preset second interval.

[0129] In the third step, for the root node of the attribution tree (i.e., the target indicator), the default analysis is to analyze the changes in the target indicator during the observation period and the baseline period; for the dimension node, according to the characteristics of the dimension node, select an attribution algorithm suitable for the dimension node from at least one attribution algorithm applicable to the dimension node (which may include a change rate algorithm, a decomposition factor proportion algorithm, or a target value difference algorithm); for the indicator node, according to the characteristics of the indicator node, select at least one attribution algorithm suitable for the indicator node from the attribution algorithms applicable to the indicator node (which may include a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm, or an occupancy rate algorithm).

[0130] The various attribution algorithms are introduced below.

[0131] The change rate algorithm focuses on the change of the indicator node itself by comparing the change rate of the same indicator node in the observation period and the base period. The change rate can be equal to the quotient of the change of the indicator node in the observation period divided by the change of the indicator node in the base period minus one. This algorithm can calculate the change rate of the target indicator and the change rate of each indicator node.

[0132] The factor ratio algorithm is used to determine which indicator node after the decomposition caused the change of the decomposed node by calculating the ratio of the change rate of the two indicator nodes decomposed from the same node in the addition or multiplication indicator decomposition. For example, Y = X1*X2 or Y = X1+X2, y represents the change rate of Y, x1 represents the change rate of X1, and x2 represents the change rate of X2. Then, through the factor ratio algorithm, it can be known that the factor ratio of X1 is equal to x1 divided by y, and the factor ratio of X2 is equal to x2 divided by y.

[0133] The target value difference algorithm is used to determine whether the gap between the indicator node and the target value meets expectations by calculating the data difference between the indicator node and the target value at the same time point. Among them, expectations can be achieved through the set influencing conditions.

[0134] The contribution rate algorithm of each value in a single dimension calculates the contribution rate of different values ​​(also called enumeration values) under a single dimension, and analyzes which value under a single dimension has the greatest impact on the analyzed decomposed indicator. For example, under the industry dimension, the contribution rate of different industries to indicator A is analyzed to determine which industry is the biggest influencing factor causing indicator A to rise or fall.

[0135] As an example, the contribution rate of a certain value in a single dimension can be equal to the percentage of the change in the value in the total change. The sum of the contribution rates of all values ​​in a single dimension is equal to 100%. However, the attribution page can only display the attribution data corresponding to the values ​​with the largest contribution rates. Therefore, the sum of the contribution rates of the values ​​displayed on the attribution page in a single dimension is usually less than 100%. Taking the following Table 1 as an example, the values ​​in a certain dimension include: A, B and C. After being processed by the single-dimensional value contribution rate algorithm, the contribution rates of each value are as follows:

[0136] Table 1

[0137]

[0138]

[0139] In this example, the influence condition can be selected as 0 or 27%, so that the change direction of the selected dimension value is always the same as the change direction of the disassembled node. Among at least one dimension node that meets the influence condition, the dimension node corresponding to the value with the largest contribution rate (such as A in Table 1 above) can be considered to be the main factor affecting the disassembled node.

[0140] It should be noted that dimensional decomposition can also include division-type decomposition. For division-type decomposition, the change of the decomposed node is not equal to the sum of the changes of the nodes of each dimension after decomposition. Therefore, for the nodes of division-type decomposition, the fixed-base substitution method or the chain substitution method can be used. Taking the fixed-base substitution method as an example, it can be understood as replacing the actual value with the standard value (historical, peer enterprise or budget standard) to measure the impact of each factor on the node before decomposition. The attribution algorithm corresponding to the division-type decomposition can be, for example, the structural attribution algorithm introduced below. Among them, the division-type decomposition can be, for example: or, or, or,

[0141] It should be noted that dimensional decomposition can also include multiplication decomposition. Multiplication decomposition belongs to process decomposition and is applicable to nodes with funnel processes. The corresponding attribution algorithm can be a single-dimensional value contribution rate algorithm. For example, assuming that indicator A = B*C, if A decreases by 30%, B increases by 10%, and C decreases by 36%, then the contribution rate of B is 10% / (-30%) = -33.3%, and the contribution rate of C is (-36%) / (-30%) = 120%. It can be seen that the main reason for the decrease in A is C.

[0142] The single-dimension specified value contribution rate algorithm calculates the situation of the specified value under a single dimension, which value of the dimension is most affected, and analyzes the impact of the specified value under a single dimension on other values ​​of the dimension after the decomposed indicator is decomposed.

[0143] The multi-dimensional contribution rate algorithm breaks down the indicator into multiple dimensions to find the dimension combination that has the greatest impact on the indicator. For example, the frequent pattern growth (Fp-growth) algorithm can be used to traverse all combinations of dimension values ​​to find the dimension with the greatest impact and the value combination under this dimension. Taking the configuration of three dimensions as an example, all combination types of the three dimensions (i.e., single dimension, two dimensions, and three dimensions) will be traversed to calculate the contribution rate of each combination type. In the multi-dimensional contribution rate algorithm, any combination of the configured multiple dimensions may be used as a determined dimension combination. For example, the configured multiple dimensions include: A, B, and C, and the returned dimension combination can be<A=A1,B=B2,C=C1> , or<A=A1,C=C1> , or<C=C1> All dimensional combinations screened out by the multi-dimensional contribution rate algorithm can be used as indicators for further decomposition and filter drilling (which can be understood as continued decomposition).

[0144] As an example, the Gini coefficient can be reused to implement a multi-dimensional contribution rate algorithm. The Gini coefficient is a parameter that measures the inequality of the distribution of fluctuation contribution values ​​according to a certain dimension, that is, a measure of whether the fluctuations between different values ​​in the same dimension are uniform, and the value range is [0, 1]. Taking the population and income dimensions as an example, the approximate calculation formula of the Gini coefficient can be found in the following formula (1):

[0145]

[0146] Among them, X represents the population proportion of each group, Y represents the income proportion of each group, V represents the cumulative income proportion of each group, i = 1, 2, 3, ..., n, n represents the number of groups, and G is the Gini coefficient of the split indicator.

[0147] The Gini coefficient can quickly locate a dimension and find the value with the greatest contribution under the current dimension. The larger the Gini coefficient, the greater the fluctuation of each value under this dimension. It is reasonable to prioritize the value under this dimension. In Gini (y, x), y is the sequence of the average contribution of each sample to the fluctuation of the decomposed indicator under each dimension, and x is the sequence of the proportion of each dimension value in the base period. The Gini coefficient can also be calculated by the following formula (2):

[0148]

[0149] Among them, G RSV is the Gini coefficient of the disassembled indicator; N represents the number of samples of the disassembled indicator in the base period, and N can be expressed as base(M), where M refers to the number of samples of the disassembled indicator; p i For Y i The weight of the association, p j For Y j The weight of the association; Y i Y is the contribution of dimension value i to the fluctuation of the decomposed indicator, j T is the contribution value of dimension value j to the fluctuation of the decomposed indicator; + is for all positive Y j Find sum(Y j *p j ), T - is for all negative Y j Find abs(sum(Y j *p j )), where sum is the sum function and abs is the absolute value function.

[0150] It should be noted that through the Gini coefficient calculation, the Gini coefficient set corresponding to each dimension can be obtained. On the attribution page displayed to the user, the N dimensions with the largest Gini coefficients and M values ​​in each of the selected N dimensions can be selected. N and M can be integers greater than or equal to 1, and N and M can be the same or different.

[0151] The structural attribution algorithm is suitable for division-type (also called ratio-type) decomposition. For example, the behavior rate can be equal to the number of users with behavior divided by the number of users visiting the mall. Therefore, the behavior rate can be decomposed into the number of users with behavior and the number of users visiting the mall. The structural attribution algorithm is used to locate whether the change of the dimension node is mainly affected by the change of the dimension node itself or the structural change (also called structural proportion). It can eliminate the Simpson's paradox that may appear in the above-mentioned various contribution rate-related algorithms.

[0152] In order to better understand structural attribution, the following example is given using the class average score.

[0153] Assume that a set has 20 objects, each with a corresponding score value. The set can be divided into two subsets: subset 1 and subset 2. The objects included in each subset account for 50% of the total number of objects in the set, that is, subset 1 and subset 2 each include 10 objects. If the average score corresponding to subset 1 is 90 points and the average score corresponding to subset 2 is 60 points, then the total average score of the set is (60×10+90×10) / 20=75 points. On this basis, if the total average score of the set drops to 67.5 points, there are two possible situations: In the first case, the average score corresponding to subset 1 drops from 90 points to 75 points, and the number of people in each subset remains unchanged, resulting in a decline in the total average score of the set. This situation can be understood as the change of the dimension node being affected by the change of the dimension node itself; In the second case, the number of objects included in subset 1 changes to 5, and the number of objects included in subset 2 changes to 15. The average score values ​​corresponding to each subset remain unchanged. The reason for the decline in the total average score of the set is the change in the number of objects included in each subset. This situation can be understood as the change of the dimension node being affected by the change of the dimension node structure.

[0154] Taking the cost per thousand impressions (CPM2) = (cost / number of impressions) * 1000, the analysis group is day t, and the comparison group is day (t-1) as an example, the structural attribution algorithm is illustrated. In the structural attribution algorithm, after the indicator is decomposed into dimensions according to the division type, the quantitative value of the impact of the change of a certain dimension node itself on the indicator before decomposition can be called the self-impact change, and the self-impact change can be shown in the following formula (3); after the indicator is decomposed into dimensions according to the division type, the quantitative value of the impact of the structural change of one or some dimension nodes on the indicator before decomposition can be called the structural impact change, and the structural impact change can be shown in the following formula (4):

[0155]

[0156]

[0157] Among them, show t It can refer to the overall value of show during the observation period, for example, it can be the sum of the show values ​​of the observation period under the different values ​​of the decomposition dimension; show t-1 It can refer to the overall value of show in the base period, for example, it can be the sum of show in the base period under each value of the decomposed dimension; i is the value after decomposition by dimension, for example, show i，t It can be the number of values ​​​​of i in the observation period after dimension decomposition, showi，t-1 It can be the number of values ​​i in the base period after dimension decomposition; CPM2 i，t It can be the cost per thousand impressions of the value i in the observation period after the dimension is decomposed, CPM2 i，t-1 It can be the cost per thousand impressions of the value i in the base period after the dimension is decomposed.

[0158] Then, the proportion of self-change can be equal to the change in self-influence of a certain dimensional node after decomposition divided by the first value. Similarly, the proportion of structural change can be equal to the change in structural influence of a certain dimensional node after decomposition divided by the first value. The first value indicates the overall change of the indicator before decomposition, and the first value can be equal to the sum of the cumulative sum of the changes in self-influence of each dimensional node after decomposition of the indicator according to the division type and the cumulative sum of the changes in structural influence.

[0159] The crowding-out rate algorithm is suitable for disassembling the same dimension of the indicator. There is a competitive relationship between the obtained dimension nodes (under limited conditions, two dimension nodes will have a situation where one increases while the other decreases). At this time, the crowding-out ratio of each dimension node is judged by the crowding-out rate of the dimension node. For example, cost fluctuations are caused by internal advertising crowding out sales, or brands crowding out bidding. The cost indicator has the dimension of advertising classification. Among the dimension nodes obtained by disassembling this dimension, sales and internal advertising usually account for a relatively large proportion. Since advertising charges are limited, more internal advertising will lead to sales revenue, which reduces sales revenue.

[0160] For example, the crowding rate can be equal to the change of the crowding dimension node divided by the change of the crowded dimension node. Since the change of the crowding dimension node and the change of the crowded dimension node are usually one positive and one negative, the crowding rate is a negative number. Therefore, in the influencing conditions corresponding to the crowding rate, the crowding rate threshold is usually a negative number. Taking the crowding rate threshold = -0.7 as an example, when the crowding rate of a certain dimension node is less than or equal to -0.7, it can be determined that the indicator change is caused by the crowding of the dimension node; when the crowding rate of a certain dimension node is greater than -0.7, it can be determined that the indicator change is not caused by the crowding of the dimension node, and you can continue to drill down (that is, perform the next hop disassembly) to find the reason for the indicator change.

[0161] In some implementations, when a node drills down to multiple child nodes (such as dimension contribution rate nodes and / or structural attribution nodes), a child node with the greatest influence is selected from these child nodes to continue drilling down, instead of continuing to drill down on all child nodes. In fact, the attribution tree is reasonably pruned, and only the child nodes that are more meaningful to continue drilling down are retained as nodes to continue drilling down. Therefore, the attribution algorithm included in the node of the attribution tree in the embodiment of the present application can also include a parallel dimension selection algorithm. The parallel dimension selection algorithm can be understood as: when there are multiple analysis paths for the same indicator, only the analysis path with the greatest influence is wanted to be found, and the corresponding disassembly and drilling down are completed. The analysis path can correspond to the disassembly method. As an example, the parallel dimension selection algorithm can select the dimension with the greatest influence for disassembly and drilling down by comprehensively considering the contribution rate (also understood as the difference ratio) and fluctuation range (i.e., the difference ratio divided by the ratio of each value) of different values ​​in each dimensional node, and no longer continue to disassemble other nodes. For example, after the internal advertisement occupies the node in the attribution tree, it can be disassembled from the four dimensions of product, billing, industry, and group. However, in order to avoid the conclusion of the attribution page being too long and to reduce the computing resources of the data analysis platform, the parallel dimension selection algorithm configured on the internal advertisement occupies the node can be used to select the dimension with the greatest impact from these four dimensions for disassembly and drilling down.

[0162] It can be understood that the core of the parallel dimension selection algorithm is to measure the impact of different dimensional decompositions on the decomposed indicators. Taking the dimension node as an example, there are two measurement methods. Method 1 calculates the Gini coefficient after different analysis paths are decomposed into dimensions for the same indicator, and selects a dimension with the largest Gini coefficient for decomposition and drilling down; Method 2 calculates the contribution rate of different analysis paths after dimensional decomposition of the same indicator (see the above attribution algorithm for contribution rate), and selects a dimension with the largest contribution rate for decomposition and drilling down. The two measurement methods can be used together or separately. When the two measurement methods are used together, one dimension can be selected for decomposition and drilling down respectively. When the dimensions selected by the two methods are different, the decomposed node is decomposed according to the two dimensions determined by the two methods respectively.

[0163] It should be noted that the parallel dimension selection algorithm only selects the decomposition dimensions (which can also be understood as pruning the attribution tree), does not involve the attribution of nodes, and will not directly drill down the values ​​of each dimension to the next hop. The actual attribution is achieved by the child nodes after decomposition according to the determined dimensions.

[0164] It should be noted that the attribution conclusions obtained by different attribution algorithms can be displayed in different styles on the attribution page. The attribution page can also display the attribution conclusions in the form of tables, graphs or text based on the user's configuration.

[0165] The following is an introduction to examples of textual attribution conclusions corresponding to various attribution algorithms.

[0166] For single-dimension contribution rate algorithms (which may include single-dimension value contribution rate algorithms and single-dimension specified value contribution rate algorithms), the attribution conclusion can be: The increase in [indicator name] is caused by [value name] ↑X% (Y, contribution rate Z). For example, the increase in [search] is caused by [comprehensive search] ↑6.76% (18.96 million, contribution rate 31.23%).

[0167] For the multi-dimensional contribution rate algorithm, the attribution conclusion can be: [Indicator Name] The increase is due to the dimension combination [Dimension Name 1 = Value Name 1] ↑ X1% (Y1, Contribution Rate Z1), [Dimension Name 1 = Value Name 2] ↑ X2% (Y2, Contribution Rate Z2), [Dimension Name 2 = Value Name 3] ↑ X3% (Y3, Contribution Rate Z3). For example, the increase in [GMV] is due to the dimension combination: [Search Intent = E-commerce Search for People] ↑ 3.66% (4.1118 million, 2.80%), [Search Vertical Category = Clothing, Shoes and Bags] ↑ 3.66% (4.1118 million, 2.80%).

[0168] For the non-structural changes of the structural attribution algorithm, the attribution conclusion can be: The increase (or decrease) of [XXX indicator] is caused by the change of [YYY dimension] itself, and the proportion of each value of this dimension has not changed significantly. Among them, [YYY dimension value 1] itself explains the change ZZ1% (change rate WW1%), and [YYY dimension value 2] itself explains the change ZZ2% (change rate WW 2%). For structural changes, the attribution conclusion can be: The increase in [XXX indicator] is caused by the structural change of [dimension name] (structural change accounts for X%), among which, the proportion of [value 1] above (or below) the overall level increases (or decreases) by XX1% (structural change explanation YY1%), and the proportion of [enumeration value 2] above (or below) the overall level increases (or decreases) by XX2% (structural change explanation YY2%); or, the attribution conclusion can be: The increase in [XXX indicator] is caused by the increase (or decrease) of [value 1] above (or below) the overall level by XX1% (contribution rate YY3%), and the proportion of [value 2] above (or below) the overall level increases (or decreases) by XX2% (contribution rate YY4%). For example, the attribution conclusion output by the structural attribution algorithm corresponding to the structural change can be: The increase in [GMV] is caused by the structural change of [search vertical category] (structural change accounts for x%), among which the proportion of [clothing, shoes and bags] above the average value increases (structural change accounts for x%). For the structural attribution algorithm corresponding to non-structural changes, the output content may include: dimension name, value, own impact change, own change explanation and change rate; for the structural attribution algorithm corresponding to structural changes, the output content may include: dimension name, value, absolute value of analysis time, denominator proportion change rate, structural change explanation and change rate.

[0169] For the crowding rate algorithm, the attribution conclusion can be: [Indicator Name] increase is caused by [Value] ↑X% (Y, crowding rate Z), and the copy can be: [Indicator Name] under [Dimension Name], [Value 1] change XX, [Value 2] change YY, crowding rate ZZ. For example, the increase in [GMV] is caused by [Apparel, Shoes and Bags] ↑3.24% (201,200, crowding rate 200%).

[0170] For the proportion algorithm, the copy of the attribution conclusion can be: The change of [indicator name] under [dimension name] is mainly reflected in the proportion of [value 1] ↑X% (analysis group Y%, comparison group Z%).

[0171] For the rate of change algorithm, the attribution conclusion can be: [Indicator Name] ↑X% (Y, comparison group Z1, analysis group Z3). For example, [GMV] ↑3.21% (901 million, comparison group 7.051 billion, analysis group 7.952 billion).

[0172] For the algorithm of factor ratio breakdown, the attribution conclusion can be: The increase of [indicator name] is caused by the breakdown indicator [indicator name] ↑X% (Y, factor ratio Z).

[0173] For the target value difference algorithm, if there is a target value, then the attribution conclusion can be: [Indicator Name] (xx) is less than (or greater than or equal to) [Target Value Indicator Name] (yy); if there is no target value, then there is no attribution conclusion because no target value has been set. In order to distinguish, indicators within and outside the target value range can be marked with different colors.

[0174] The following is an introduction to the fields included in the attribution conclusions corresponding to various attribution algorithms and examples of related descriptions.

[0175] For a single-dimension-related contribution rate algorithm, the following fields may be included: value, change rate, change amount, and contribution rate.

[0176] For the multi-dimensional contribution rate algorithm, since it is not possible to drill down layer by layer across dimensions, the fields cannot be merged when displayed.

[0177] For the structural attribution algorithm, the following fields may be included: value, self-influence change, structural influence change, self-change ratio and structural change ratio. Among them, at least one of the following display contents may be added to the first page: analysis group numerator data, comparison group numerator data, numerator change rate, analysis group denominator data, comparison group denominator data, denominator change rate, analysis group denominator ratio, comparison group denominator ratio or denominator ratio change rate.

[0178] For the crowding rate algorithm, the following fields may be included: value, change amount, change amount, and crowding rate.

[0179] For the proportion algorithm, since the conclusion is simple and rarely used, field merging is not performed for the time being.

[0180] For the overall change rate in the change rate algorithm, the following fields may be included: indicator, change rate, change amount, analysis group absolute value, and comparison group absolute value.

[0181] For the change rate of a specified dimension in the change rate algorithm, the following fields can be included: value, indicator, change rate, change amount, analysis group absolute value, and comparison group absolute value. In this case, on the one hand, the merging of nodes after indicator decomposition is supported; on the other hand, when the drilled-down parallel indicator (i.e., the indicator obtained from the same decomposition) is an indicator node, regardless of whether all decomposition factors are attributed, the decomposition formula needs to be displayed on the title of the chart. For example, the increase in [Consumption] can be analyzed from the perspective of [Daily Active User (DAU) * Average Revenue Per User (ARPU)], the increase in [Consumption] can be analyzed from the perspective of [Brand Consumption + Bidding Consumption], and the increase in [Consumption] can be analyzed from the perspective of [DAU*ARPU and (Brand Consumption + Bidding Consumption)].

[0182] For the factor breakdown ratio algorithm, the following fields may be included: indicator, change rate, change amount, and indicator breakdown ratio.

[0183] For the target value difference algorithm, since the conclusion is simple and rarely used, field merging is not performed for the time being.

[0184] In order to make the method provided in the embodiment of the present application clearer, Figure 6 An example in a practical scenario is shown to illustrate the embodiments of the present application.

[0185] When an analysis requirement of the data analysis platform includes: first, determining whether the fluctuation of total advertising consumption is related to the fluctuation of DAU, and second, finding the groups that cause the largest fluctuation of total advertising consumption. Then, in the method provided in the embodiment of the present application, the attribution tree corresponding to the analysis requirement can be found in Figure 6As shown, the root node is the total advertising consumption (i.e., the target indicator, the ID of this node = 0), the indicator nodes obtained by disassembling the root node may include: daily active user node 1 (the ID of this node = 1) and daily active old customer number node 2 (the ID of this node = 2), and the dimension nodes obtained by disassembling the root node may include: customer first-level industry node 3 (the ID of this node = 3) and group name node 4 (the ID of this node = 4). Among them, the daily active user node 1 is used to analyze the impact of DAU on the total advertising consumption, and the influence condition 1 between the daily active user node 1 and the daily active old customer number node 2 may include that the proportion of the disassembly factor is greater than or equal to 0. If the daily active user node 1 meets the influence condition 1, then check the daily active old customer number node 2; the daily active old customer number node 2 is used to analyze the change rate and change amount of DAU, and the influence condition 2 between the daily active old customer number node 2 and the next hop may include that the overall change rate is greater than or equal to 0. If the daily active old customer number node 2 meets the influence condition 2, then continue to check the next hop of the daily active old customer number node 2. The customer-level industry node 3 is used to analyze the changes in advertising consumption of different industries. The influence condition 3 between the customer-level industry node 3 and the group name node 4 may include a preset number (such as 3) industries with a contribution rate greater than or equal to 0 and the largest contribution rate. If the customer-level industry node 3 meets the influence condition 3, continue to check the group name node 4; the group name node 4 is used to analyze the changes in advertising consumption of the group in the industry filtered by the influence condition 3. The influence condition 4 between the group name node 4 and the next hop may include a preset number (such as 5) group names with a contribution rate greater than or equal to 0 and the largest contribution rate. If the group name node 4 meets the influence condition 4, continue to check the next hop of the group name node 4.

[0186] Then, in this example, the analysis conclusion corresponding to the analysis requirement may include, for example:

[0187] The total advertising consumption increased (also expressed as ↑) by 6.66% (+1 million) on January 3, 2020 compared to January 2, 2020. The reasons may include:

[0188] a. The increase in [total advertising consumption] was caused by a 1.1% increase in [daily active users] (+2.2 million, with the breakdown factor accounting for 47%). Among them, [daily active users] were 10 million in the baseline period and 9 million in the observation period, an increase of 11% (+1 million);

[0189] b. The increase in [Total Advertising Consumption] was caused by a 70% increase in [Financial Industry] (+3 million, contribution rate 40%), a 50% increase in [Games] (+2.8 million, contribution rate 37%), and a 33% increase in [Retail] (+560,000, contribution rate 9%). Among them, the increase in [Financial Industry] was caused by a 60% increase in [Group A] (+660,000, contribution rate 66%), a 30% increase in [Group B] (+400,000, contribution rate 40%), and an 8% increase in [Group C] (+300,000, contribution rate 30%); the increase in [Games] was caused by a 120% increase in [Group D] (+200,000, contribution rate 20%), a 10% increase in [Group E] (+150,000, contribution rate 15%), and a 15% increase in [Group F] (+100,000, contribution rate 10%).

[0190] The core conclusion area on the data analysis platform about indicator attribution can be understood as a minimalist version of the attribution conclusion. This area simplifies the copy based on the output of the product function-attribution conclusion, decomposes the target indicator to obtain indicator nodes and dimension nodes, finds the value of the maximum contribution in the [first layer] or the dimension corresponding to the proportion of the decomposed factors, and takes the relevant information of the second largest value based on the original output rules; finds another layer based on the [first layer] and outputs the relevant information of the second largest value. For example: For the root node in the analysis tree, which is the e-commerce GMV, the nodes of the first layer include industry, field, and daily active customers (DailyActive Customers, DAC)*average order value. The industry is decomposed to obtain categories and DAC*average order value.

[0191] The full attribution conclusion of e-commerce indicator A can include: e-commerce indicator A↓10% (-100), the absolute value of the analysis group is 1000, and the absolute value of the comparison group is 900; on the one hand, the decline of [e-commerce indicator A] is caused by a 15% decline in [beauty] (-70, contribution rate 70%), a 10% decline in [personal care and household cleaning] (-50, contribution rate 50%), and a 5% decline in [food] (-20, contribution rate 20%). Among them, from the perspective of [category]: the decline in [beauty] was caused by a 15% decline in [luxury goods] (-50, contribution rate 71.4%), a 10% decline in [personal care] (-10, contribution rate 14.3%), and a 5% decline in [food] (-5, contribution rate 7.2%); the decline in [personal care and household cleaning products] was caused by a 15% decline in [personal care] (-50, contribution rate 71.4%), a 10% decline in [beauty and skin care] (-10, contribution rate 14.3%), and a 5% decline in [cleaners] (-5, contribution rate 7.2%); the decline in [food] was caused by a 15% decline in [fresh food] (-40, contribution rate 57.1%), a 10% decline in [alcoholic beverages] (-10, contribution rate 14.3%), and a 5% decline in [nuts] (-5, contribution rate 7.2%). From the perspective of [DAC*Average Order Value]: the decline in [Beauty] was caused by a 2% decline in [DAC] (-12.34 million, 77% of the dismantling factors), and a 1% decline in [Average Order Value] (-2.5, 23% of the dismantling factors); the decline in [Personal Care and Home Cleaning] was caused by a 1% decline in [DAC] (-5.67 million, 50% of the dismantling factors), and a 1% decline in [Average Order Value] (-2.5, 50% of the dismantling factors); the decline in [Food] was caused by a 1.1% decline in [DAC] (-6.67 million, 53% of the dismantling factors), and a 0.9% decline in [Average Order Value] (-2, 47% of the dismantling factors). On the other hand, the decline in [E-commerce Indicator A] was caused by a 15% decline in [Content Field] (-50, 50% contribution rate), a 10% decline in [Shelf Field] (-40, 40% contribution rate), and a 10% decline in [Others] (-10, 10% contribution rate). On the other hand, the decline in [e-commerce indicator A] was caused by a 1% drop in [DAC] (-12.34 million, with the breakdown factors accounting for 50%) and a 1% drop in [average order value] (-4, with the breakdown factors accounting for 50%).

[0192] Based on the above full attribution conclusions, the core conclusions (i.e. the content displayed in the core conclusion area) can include: the decline in [e-commerce indicator A] was caused by a 15% decline in [Beauty] (-70, contribution rate 70%) and a 10% decline in [Personal Care and Household Cleaning] (-50, contribution rate 50%). Among them, the decline in [Beauty] was caused by a 2% decline in [DAC] (-12.34 million, with the breakdown factors accounting for 77%) and a 1% decline in [average order value] (-2.5, with the breakdown factors accounting for 23%); the decline in [Personal Care and Household Cleaning] was caused by a 1% decline in [DAC] (-5.67 million, with the breakdown factors accounting for 50%) and a 1% decline in [average order value] (-2.5, with the breakdown factors accounting for 50%).

[0193] From the indicator attribution page on the data analysis platform, operate any entry of the target indicator, and display the first page or the second page of the attribution page. Among them, the content displayed on the first page may include the following five parts. (1) Global screening, at least the baseline period and observation period can be configured. When the baseline period and observation period are not configured, the time can inherit the time of the user's operation on the attribution page before the first page is displayed. (2) Indicator overview, which may include but is not limited to: the current value, month-on-month, month-on-month difference of the target indicator, and whether the current value is a historical high. (3) Indicator trend, for example, the trend of the target indicator in the past 30 days can be displayed. (4) Attribution tree, used to display all attribution nodes, where the node is clickable. After clicking, the attribution data of the path where the clicked node is located will be displayed in the subsequent (5) [Detailed Attribution]. For example, the current value and month-on-month change of each node can be displayed. The month-on-month change can retain two significant digits. The prerequisite for display may be that there is no dimension node in the previous jump of the node. (5) Detailed attribution: sorting function can be added through multi-level tables; attribution data displayed in the core conclusion can be highlighted; nodes where structural proportion is the main factor can add hover prompts to the values ​​of change rate, change amount, and change contribution rate. The prompts can be, for example, "Search for meaning" Figure 1 The average order value is lower than the overall value, and the number of unique visitors (UV) accounts for -10%, resulting in an increase in the overall average order value." The expansion of table data is not limited by the influencing conditions of the attribution tree. By default, the maximum preset number (such as 6) of attribution data is displayed. The number of attribution data can be increased by adding data controls; the data download function can be added to download all attribution data (including intermediate data and attribution conclusions). The downloaded attribution data can be carried by the table; the downloaded table can be displayed with a ceiling effect. The content displayed on the second page can be based on different attribution algorithms, corresponding to different attribution conclusion output styles.

[0194] It can be seen that in this method, an attribution entry corresponding to each of the multiple indicators is designed on the display page of the data analysis platform. When there is a need to attribute the target indicator, the user only needs to operate the attribution entry corresponding to the target indicator on the display page of the data analysis platform to see the attribution page carrying the attribution data of the target indicator. The attribution data is used to indicate the correlation between the factors affecting the target indicator and the target indicator. The attribution data includes the attribution values ​​corresponding to each of the factors based on the target indicator. The attribution value of the factor is determined based on the decomposition method of the factor relative to the target indicator and the influence of the factor on the change of the target indicator. Users do not need to waste time and effort searching one by one among the numerous attribution modules on the current data analysis platform to see whether there is an attribution module corresponding to the target indicator. The public control panel on the improved data analysis platform, which includes an attribution portal corresponding to each of the multiple indicators, provides the attribution function of each indicator. Moreover, the attribution data obtained after the attribution portal is triggered is adaptively calculated, analyzed and determined based on the specific circumstances of the factors affecting the target indicator, and is more accurate attribution data for the target indicator. Through this method, users can quickly, accurately and comprehensively obtain the attribution data of the target indicator they want to know, thereby improving the user experience of the indicator attribution function on the data analysis platform.

[0195] Accordingly, the present embodiment also provides an indicator attribution device 700, see Figure 7 As shown, the apparatus 700 may include:

[0196] A first acquisition unit 701 is configured to acquire attribution data of the target indicator in response to an operation on an attribution entry corresponding to the target indicator on a display page, wherein the display page includes an attribution entry corresponding to each of a plurality of indicators, and the attribution data is used to indicate an association relationship between a factor affecting the target indicator and the target indicator, and the attribution data includes an attribution value corresponding to the factor affecting the target indicator, and the attribution value of the factor is determined based on a disassembly method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator;

[0197] The first display unit 702 is used to display the attribution page of the target indicator, and the attribution page is used to present the attribution data.

[0198] In a possible implementation, the display page includes a first area, a second area, and a third area, the first area includes a core attribution conclusion of at least one preconfigured core indicator and a first control corresponding to each core indicator in the at least one core indicator, the at least one core indicator includes the target indicator, the second area includes a plurality of indicator cards, each indicator card in the plurality of indicator cards includes a second control, and the third area includes a trend chart of a single indicator and / or multiple indicators;

[0199] Among them, the attribution entrance is any one of the following entrances: any point on the trend chart, the first control or the second control.

[0200] In a possible implementation, the attribution data includes detailed data of the target indicator, and the first display unit 702 is specifically configured to:

[0201] The first page is displayed, and the first page displays detailed data of the target indicator.

[0202] As an example, the attribution data further includes the core attribution conclusion of the target indicator, the attribution page includes a second page in addition to the first page, and the first display unit 702 includes:

[0203] A first display subunit, configured to display the second page, wherein the second page includes a core attribution conclusion of the target indicator and a third control;

[0204] The second display subunit is used to display the first page in response to an operation on the third control on the second page.

[0205] In a possible implementation, the first page includes a fourth area, a fifth area, and a sixth area, wherein:

[0206] The fourth area includes the filter condition configuration options of the target indicator and the trend graph corresponding to the target indicator under the configured filter condition;

[0207] The fifth area includes an attribution tree, the root node of the attribution tree is the target indicator, the leaf nodes of the attribution tree include indicator nodes and / or dimension nodes, the indicator nodes are factors affecting the target indicator obtained by decomposing the target indicator, the dimension nodes are factors affecting the target indicator obtained by decomposing the target indicator, each leaf node has a corresponding attribution value, the attribution value of the leaf node is determined based on the decomposition method of the leaf node relative to the target indicator and the influence of the leaf node on the change of the target indicator, and the link between the nodes is configured with a corresponding influence condition, and the influence condition is used to determine whether the detailed data of the backend node of the link is displayed;

[0208] The sixth area includes detailed data of each node on the path corresponding to the selected node in the attribution tree, and the path corresponding to the selected node is the path from the root node to the selected node.

[0209] In a possible implementation manner, the apparatus 700 further includes:

[0210] A second acquisition unit is configured to obtain, in response to an operation on a target node of the attribution tree on the first page, an attribution algorithm corresponding to each node on a path corresponding to the target node and an influencing condition corresponding to each link on the path corresponding to the target node;

[0211] A third acquisition unit is used to perform attribution analysis on the corresponding node based on the attribution value corresponding to each node on the path corresponding to the target node, and obtain detailed data of each node on the path corresponding to the target node;

[0212] A determination unit, used to determine whether the detailed data of two nodes on each link on the path corresponding to the target node meets the influence condition of the corresponding link, and if so, determine to display the detailed data of the back-end node in the link;

[0213] The second display unit is used to display detailed data of the nodes corresponding to the factors on the path corresponding to the target node in the sixth area, where the factors are indicators corresponding to the nodes determined to be displayed on the path corresponding to the target node.

[0214] In one possible implementation,

[0215] The method for determining the attribution value corresponding to the indicator node includes at least one of the following algorithms: a rate of change algorithm, a factor ratio algorithm, or a target value difference algorithm;

[0216] The method for determining the attribution value corresponding to the dimension node includes at least one of the following algorithms: a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm or an occupancy rate algorithm.

[0217] In a possible implementation, the influencing condition includes any one of the following conditions:

[0218] Whether the absolute value of the detailed data of the front-end node of the corresponding link belongs to the preset target interval, and the front-end node is the indicator node;

[0219] Whether the front-end node of the corresponding link belongs to the preset target number of nodes with the largest value of detail data, the front-end node is a dimension node.

[0220] It should be noted that the specific implementation method of the device 700 and the technical effects achieved can be found in Figure 1 Description of the method shown.

[0221] In addition, an embodiment of the present application also provides an electronic device, which includes a processor and a memory: the memory is used to store instructions or computer programs; the processor is used to execute the instructions or computer programs in the memory, so that the electronic device executes any implementation of the method provided in the embodiment of the present application.

[0222] See also Figure 8 , which shows a schematic diagram of the structure of an electronic device 800 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0223] like Figure 8 As shown, the electronic device 800 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0224] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the electronic device 800 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 The electronic device 800 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0225] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 809, or installed from the storage device 808, or installed from the ROM 802. When the computer program is executed by the processing device 801, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0226] The electronic device provided by the embodiment of the present disclosure and the method provided by the above embodiment belong to the same inventive concept, and the technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0227] The embodiment of the present application also provides a computer-readable medium, in which instructions or computer programs are stored. When the instructions or computer programs are executed on a device, the device executes any implementation of the method provided in the embodiment of the present application.

[0228] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0229] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0230] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0231] The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device can execute the method.

[0232] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0233] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0234] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit / module does not, in some cases, constitute a limitation on the unit itself.

[0235] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0236] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0237] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the system or device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0238] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0239] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0240] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0241] It should be noted that in the embodiments of the present application, user sensitive information is not involved, and user-related information is obtained, used and determined after user authorization. In one example, before obtaining user-related information, the corresponding interface displays prompt information related to obtaining data use authorization. The prompt information is based on relevant laws and regulations in an appropriate manner to inform the user of the type, scope of use, usage scenarios, etc. of the personal information involved in this disclosure, so that the user can determine whether to agree to the authorization based on the prompt information. It can be understood that the above notification and the process of obtaining user authorization are only illustrative and do not constitute a limitation on the implementation method of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation method of the present disclosure.

[0242] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An indicator attribution method, characterized in that: include: In response to an operation on an attribution entry corresponding to a target indicator on a display page, attribution data of the target indicator is obtained, the display page including an attribution entry corresponding to each of a plurality of indicators, the attribution data being used to indicate an association relationship between a factor affecting the target indicator and the target indicator, the attribution data including an attribution value corresponding to the factor affecting the target indicator, the attribution value of the factor being determined based on a disassembly method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator; An attribution page for the target indicator is displayed, wherein the attribution page is used to present the attribution data.

2. The method according to claim 1, characterized in that The display page includes a first area, a second area, and a third area, the first area includes a core attribution conclusion of at least one pre-configured core indicator and a first control corresponding to each core indicator in the at least one core indicator, the at least one core indicator includes the target indicator, the second area includes a plurality of indicator cards, each indicator card in the plurality of indicator cards includes a second control, and the third area includes a trend chart of a single indicator and / or multiple indicators; Among them, the attribution entrance is any one of the following entrances: any point on the trend chart, the first control or the second control.

3. The method according to claim 1 or 2, characterized in that: The attribution data includes detailed data of the target indicator, and the attribution page showing the target indicator includes: The first page is displayed, and the first page displays detailed data of the target indicator.

4. The method according to claim 3, characterized in that The attribution data also includes the core attribution conclusion of the target indicator. The attribution page includes a second page in addition to the first page. The attribution page displaying the target indicator includes: Displaying the second page, where the second page includes the core attribution conclusion of the target indicator and a third control; In response to an operation on the third control on the second page, the first page is displayed.

5. The method according to claim 3 or 4, characterized in that: The first page includes a fourth area, a fifth area and a sixth area, wherein: The fourth area includes the filter condition configuration options of the target indicator and the trend graph corresponding to the target indicator under the configured filter condition; The fifth area includes an attribution tree, the root node of the attribution tree is the target indicator, the leaf nodes of the attribution tree include indicator nodes and / or dimension nodes, the indicator nodes are factors affecting the target indicator obtained by decomposing the target indicator, the dimension nodes are factors affecting the target indicator obtained by decomposing the target indicator, each leaf node has a corresponding attribution value, the attribution value of the leaf node is determined based on the decomposition method of the leaf node relative to the target indicator and the influence of the leaf node on the change of the target indicator, the link between the nodes is configured with a corresponding influence condition, and the influence condition is used to determine whether the detailed data of the backend node of the link is displayed; The sixth area includes detailed data of each node on the path corresponding to the selected node in the attribution tree, and the path corresponding to the selected node is the path from the root node to the selected node.

6. The method according to claim 5, characterized in that The method further comprises: In response to an operation on a target node of the attribution tree on the first page, obtaining attribution algorithms corresponding to nodes on a path corresponding to the target node and influencing conditions corresponding to links on the path corresponding to the target node; Based on the attribution values ​​corresponding to the nodes on the path corresponding to the target node, perform attribution analysis on the corresponding nodes to obtain detailed data of the nodes on the path corresponding to the target node; Determine whether the detailed data of two nodes on each link on the path corresponding to the target node meets the influence condition of the corresponding link, and if so, determine to display the detailed data of the back-end node in the link; In the sixth area, detailed data of the nodes corresponding to the factors on the path corresponding to the target node are displayed, and the factors are indicators corresponding to the nodes determined to be displayed on the path corresponding to the target node.

7. The method according to claim 5 or 6, characterized in that: The method for determining the attribution value corresponding to the indicator node includes at least one of the following algorithms: a rate of change algorithm, a factor ratio algorithm, or a target value difference algorithm; The method for determining the attribution value corresponding to the dimension node includes at least one of the following algorithms: a single-dimensional value contribution rate algorithm, a single-dimensional specified value contribution rate algorithm, a multi-dimensional contribution rate algorithm, a structural attribution algorithm or an occupancy rate algorithm.

8. The method according to claim 5 or 6, characterized in that: The influencing conditions include any one of the following conditions: Whether the absolute value of the detailed data of the front-end node of the corresponding link belongs to the preset target interval, the front-end node is the indicator node; Whether the front-end node of the corresponding link belongs to the preset target number of nodes with the largest value of detail data, the front-end node is a dimension node.

9. An indicator attribution device, characterized in that: include: an acquisition unit, configured to acquire attribution data of the target indicator in response to an operation on an attribution entry corresponding to the target indicator on a display page, wherein the display page includes an attribution entry corresponding to each of a plurality of indicators, wherein the attribution data is used to indicate an association relationship between a factor affecting the target indicator and the target indicator, wherein the attribution data includes an attribution value corresponding to the factor affecting the target indicator, wherein the attribution value of the factor is determined based on a disassembly method of the factor relative to the target indicator and an influence of the factor on a change in the target indicator; A display unit is used to display the attribution page of the target indicator on the display page, and the attribution page is used to present the attribution data.

10. An electronic device, characterized in that: The electronic device comprises: a processor and a memory; The memory is used to store instructions or programs; The processor is used to execute the instructions or programs in the memory so that the electronic device executes the method according to any one of claims 1 to 8.

11. A readable medium, characterized in that: The readable medium stores instructions or programs, and when the instructions or programs are executed on a processor, the processor is enabled to execute the method according to any one of claims 1 to 8.

Images