Data analysis method, device, equipment, medium and program product

Abstract

The embodiment of the invention relates to a data analysis method and device, equipment, a medium and a program product. The method includes determining a set of attribution links for a target indicator based on the target indicator in an experimental result of a target experiment, where the attribution links describe changes in the indicator and attribution logic for the indicator. The method further includes determining a set of indicator values for a set of associated indicators associated with the target indicator based on the set of attribution links. The method further includes generating an analysis result for the target indicator based on the target indicator value of the target indicator, the set of attribution links, and the set of indicator values. Through the method, the target index in the experimental result and the corresponding group of attribution links are utilized to obtain the analysis result for the target index, so that the index analysis and explanation efficiency is improved, and the user experience is improved.

Description

Technical Field

[0001] Embodiments of the present disclosure generally relate to the field of data analysis, and more particularly to methods, devices, equipment, media, and program products for analyzing data. Background Art

[0002] At present, more and more software applications are known and used by people, and various software applications are constantly improving the diversity and expansibility of application functions to attract more users to download and use their products. However, before improving the diversity and expansibility of product functions, it is necessary to conduct certain data analysis on the improved products and further optimize the products based on the conclusions of the analysis.

[0003] With the gradual development of software application product development, product experience is becoming more and more important and has become an important indicator to attract users. Therefore, product data analysis needs to be conducted from multiple aspects, that is, from multiple indicators, and comprehensive consideration should be given to the analysis results of multiple indicators in order to better optimize product experience and improve user experience. Summary of the invention

[0004] Embodiments of the present disclosure provide a method, apparatus, device, medium, and program product for analyzing data.

[0005] According to a first aspect of the present disclosure, a method for analyzing data is provided. The method includes determining a set of attribution links for the target indicator based on the target indicator in the experimental results of the target experiment, wherein the attribution link describes the change of the indicator and the attribution logic for the indicator. The method also includes determining a set of indicator values ​​of a set of associated indicators associated with the target indicator based on the set of attribution links. The method also includes generating an analysis result for the target indicator based on the target indicator value of the target indicator, a set of attribution links and a set of indicator values.

[0006] In a second aspect of the present disclosure, a device for analyzing data is provided. The device includes a set of attribution link determination modules, configured to determine a set of attribution links for a target indicator based on a target indicator in an experimental result of a target experiment, wherein the attribution link describes a change in the indicator and an attribution logic for the indicator; a set of indicator value determination modules, configured to determine a set of indicator values ​​of a set of associated indicators associated with the target indicator based on a set of attribution links; and an analysis result generation module, configured to generate an analysis result for the target indicator based on a target indicator value of the target indicator, a set of attribution links, and a set of indicator values.

[0007] In a third aspect of the present disclosure, an electronic device is provided, comprising at least one processor; and a storage device for storing at least one program, wherein when the at least one program is executed by the at least one processor, the at least one processor implements the method according to the first aspect of the present disclosure.

[0008] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0009] In a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method according to the first aspect of the present disclosure when executed by a processor.

[0010] It should be understood that the contents described in this content section are not intended to limit the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other objects, features and advantages of the present disclosure will become more apparent through a more detailed description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present disclosure.

[0012] Figure 1 A schematic diagram illustrating an example environment in which the apparatus and / or methods of some embodiments of the present disclosure may be implemented;

[0013] Figure 2 A schematic diagram illustrating an example method of analyzing data according to some embodiments of the present disclosure;

[0014] Figure 3 A schematic diagram illustrating an example of a system architecture for analyzing data according to some embodiments of the present disclosure;

[0015] Figure 4 A schematic diagram illustrating a flow chart for analyzing data according to some embodiments of the present disclosure;

[0016] Figure 5 A schematic diagram illustrating another flow chart for analyzing data according to some embodiments of the present disclosure;

[0017] Figure 6 A schematic block diagram of an apparatus for analyzing data according to some embodiments of the present disclosure is illustrated;

[0018] Figure 7 A schematic block diagram of an example device suitable for implementing various embodiments of the present disclosure is illustrated.

[0019] In the various drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION

[0020] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0021] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0022] For example, when receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0023] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0024] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0025] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0026] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0027] In the process of analyzing data, there are still many problems that need to be solved. For example, in the research and development of Internet products, it is necessary to conduct certain internal tests on the developed products. Then, the developed products are analyzed based on the experimental results obtained from the internal tests, such as functional and extensibility analysis, and then improvement plans for the products are obtained to make it easier for users to download and use the corresponding products. Usually, in traditional solutions, the analysis of developed products is carried out through experimental methods. After the experiment, the results are processed, such as using statistical methods to classify and summarize the obtained data. Then summarize the differentiated analysis results for the products to further determine how to better optimize the products.

[0028] In the above traditional scheme, the experiment can be a comparative experiment, such as an AB experiment. When conducting an AB experiment, two different strategies are usually formulated first, such as the original strategy A and the new strategy B. After the two strategies are formulated, a small portion of the traffic with lower risk is taken from the online traffic and randomly assigned to the original strategy A and the new strategy B. The interference of other factors will be eliminated in the random assignment process, and finally a certain statistical method is combined to obtain an accurate estimate of the relative effect of the two strategies, that is, to obtain the quantitative results for the two strategies. The above traditional scheme is widely used in Internet product development and scientific research due to its low risk, anti-interference and quantitative results. In addition, after the experiment in the above traditional scheme is run for a period of time with a small amount of online traffic, a significant difference between the control group and the experimental group can be obtained through statistical calculations. Subsequently, certain conclusions and suggestions are given manually or by pre-defined rules. Certain report conclusions can also be automatically output through automated rules. However, these conclusions are often rigid and can only support similar experiments, without transferability, which affects user experience. In addition, there is also the problem that the manual experience of report interpretation is difficult to transfer and cannot be replicated, which further causes the problem of high learning cost.

[0029] At least to solve the above and other potential problems, an embodiment of the present disclosure proposes a method for analyzing data. In this method, a set of attribution links for the target indicator can be determined at the computing device according to the target indicator in the experimental results of the target experiment. The attribution link describes the change of the indicator and the attribution logic for the indicator. Then, after determining a set of attribution links, the computing device will further determine a set of indicator values ​​for a set of associated indicators associated with the target indicator based on a set of attribution links. Finally, the computing device generates an analysis result for the target indicator based on the target indicator value of the target indicator, a set of attribution links and a set of indicator values. Through this method, the target indicator in the experimental results and the corresponding set of attribution links are used to obtain the analysis result for the target indicator, which improves the efficiency of indicator analysis and interpretation and improves the user experience.

[0030] The embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. Figure 1 An example environment in which the device and / or method of an embodiment of the present disclosure can be implemented is shown. In environment 100, computing device 102 first determines a set of attribution links 110 for target indicator 108 based on target indicator 108 in experimental result 106 of target experiment 104. One of the attribution links describes the change of an indicator and the attribution logic for the indicator. Then, after determining the set of attribution links 110, computing device 102 further determines a set of indicator values ​​114 of a set of associated indicators 112 associated with target indicator 108 based on the set of attribution links 110. Finally, computing device 102 generates analysis result 118 for target indicator 108 based on target indicator value 116 of target indicator 108, a set of attribution links 110 and a set of indicator values ​​114.

[0031] Examples of computing device 102 include, but are not limited to, personal computers, server computers, handheld or laptop devices, mobile devices (such as mobile phones, personal digital assistants (PDAs), media players, etc.), multiprocessor systems, consumer electronics, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

[0032] like Figure 1 As shown, the computing device 102 first obtains the experimental result 106 for the target experiment 104. In one example, the experimental result 106 is an experimental result obtained for a comparative experiment. In another example, the experimental result 106 is an experimental result obtained for a non-comparative experiment. The above examples are only used to describe the present disclosure, but are not specifically limited to the present disclosure.

[0033] The experimental result 106 includes a set of indicators, and each indicator in the set of indicators has a corresponding indicator value. For example, the indicator value is the degree of change of the indicator obtained for the two strategies in the comparative experiment, such as the percentage change of the indicator or the amount of change of the indicator. The computing device 102 can then select the target indicator 108 to be analyzed based on the indicator value. For example, if a user conducts an AB experiment, experimental data of multiple indicators such as the average number of users per capita in the past seven days, the average number of active days per capita, and the average number of active days per capita in the past 30 days under two different strategies can be obtained, so that the corresponding degree of change can be determined for the experimental data of the same indicator under the two strategies to be used as the indicator value.

[0034] The computing device 102 can analyze the indicators that meet the predetermined conditions. In some embodiments, the computing device 102 can analyze each indicator value in the experimental results. In some embodiments, the computing device 102 obtains the indicator value of each indicator in a set of indicators. Then determine whether to analyze the indicator based on the direction of the indicator. In one example, if the indicator value is positive, such as a positive number, the indicator does not need to be analyzed. If the indicator value is negative, such as a negative number, the indicator needs to be analyzed. For example, if it is determined that the number of submissions in the past seven days is -1.4%, the indicator needs to be analyzed. If it is determined that the number of submissions in the past seven days is +1.4%, the indicator does not need to be analyzed. In another example, only indicators with positive indicator values ​​can be analyzed, and indicators with negative indicator values ​​do not need to be analyzed.

[0035] In some embodiments, the computing device 102 needs to determine not only the direction of the indicator value, but also the size of the indicator value. Then, the computing device 102 selects a target indicator from a set of indicator values ​​according to the size and direction of the indicator. For example, the indicator is determined as the target indicator only when the direction of the indicator value of the indicator is negative and its absolute value is greater than a threshold value. The above examples are only used to describe the present disclosure, and are not specific limitations of the present disclosure.

[0036] Then, the computing device 102 further determines a set of attribution links 110 associated with the target indicator 108. For example, a set of attribution links 110 corresponding to the target indicator 108 is preset. The attribution links describe the changes in the indicator and the attribution logic for the indicator.

[0037] Among them, the attribution logic mainly describes which factors (such as advertising, products, prices, competition, etc.) cause the change of a certain indicator (such as sales, conversion rate, user satisfaction, etc.) and / or the impact of the change of the indicator. For example, for the decline in sales, the corresponding attribution logic describes that the decline in sales is caused by the negative impact of advertising and prices. The attribution link combines various indicator data and the corresponding attribution logic to form a link, and the link can be stored in the relevant link library so that it can be directly used to analyze the experimental results during the analysis of the experimental results.

[0038] The following are some examples of attribution links. For example, the decline in "the proportion of 1 minute+ video duration" has affected the consumption structure of medium-long videos and short videos. For example, medium-long videos are promoted more and short videos are promoted less. The duration may increase, but it does not mean that the user experience has improved; the decline in long-term (LT) retention and duration has led to a decline in all business indicators; the consumption of medium-long videos has increased, and the distribution of user-generated content (UGC) / short videos has decreased, which has reduced the number of short video submissions.

[0039] In some embodiments, a set of attribution links is obtained from a link library storing attribution links. The link library may be updated. For example, attribution links related to indicators that no longer need to be analyzed are deleted from the link library, or new attribution links that need to be analyzed are added to the link library.

[0040] After determining a set of attribution links 110, the computing device 102 further determines a set of indicator values ​​114 for a set of associated indicators 112 in the set of attribution links 110 according to the set of attribution links 110. Since the indicators obtained in the experiment usually have corresponding indicator values, a set of indicator values ​​114 corresponding to the set of associated indicators 112 can be found in the experimental results 106.

[0041] Finally, the computing device 102 generates an analysis result 118 for the target indicator 108 according to the target indicator value 116, the set of attribution links 110, and the set of indicator values ​​114 of the target indicator 108. Additionally, it is necessary to screen the associated indicators in each attribution link in the set of attribution links for the target indicator, select the associated indicators and their indicator values ​​that meet the conditions, and then combine the indicator values ​​of the associated indicators that meet the requirements and the target indicator values ​​with the set of attribution links 110 to analyze the target indicator.

[0042] In some embodiments, the target indicator 108 is an indicator in the experimental result 106. When performing analysis, only one target indicator may be analyzed, or multiple indicators in the experimental result may be analyzed. It is understood that the above is only used to describe the present disclosure and is not intended to limit the present disclosure.

[0043] Figure 1 It is shown that the computing device 102 obtains the experimental result 106 of the target experiment 104. The experimental result 106 may be received by other computing devices, or may be generated by the computing device 102 executing the target experiment 104.

[0044] Through this method, the target indicator in the experimental results and a corresponding set of attribution links are used to obtain the analysis results for the target indicator, which improves the analysis and interpretation efficiency of the indicator and improves the user experience.

[0045] Combined with the above Figure 1 A schematic diagram of an example environment in which the devices and / or methods of some embodiments of the present disclosure may be implemented is described below. Figure 2 A schematic diagram depicting an example method of analyzing data according to some embodiments of the present disclosure. Figure 2 The method in can be Figure 1 Executed by computing device 102 or any suitable computing device.

[0046] like Figure 2 As shown, in the example method 200, at box 202, the computing device 102 determines a set of attribution links for the target indicator based on the target indicator in the experimental results of the target experiment, wherein the attribution link describes the change of the indicator and the attribution logic for the indicator.

[0047] In some embodiments, a set of attribution links corresponding to the target indicator is obtained from a link library. Typically, a user can determine different attribution links corresponding to different indicators based on historical experimental data or experimental reports. Then, the user can store the obtained attribution links in the link library for analysis of subsequent experimental results. In one example, a user can store the attribution links in the link library through an interface in a web page. The link library can be updated. For example, the attribution links related to indicators that no longer need to be analyzed are deleted from the link library, or new attribution links that need to be analyzed are added to the link library.

[0048] In some embodiments, when performing analysis, the computing device 102 first obtains the experimental results 106 for the target experiment 104. The experimental results 106 include a set of indicators obtained by performing the experiment and the corresponding indicator values. The computing device 102 then analyzes the indicator values ​​of the set of indicators, and then determines whether it is selected as the target indicator 108 to be analyzed based on whether the indicator value meets the predetermined standard. For example, if it is determined that the number of submissions in the past seven days is -1.4%, if the selection criteria for the target indicator require that the data is negative and the absolute value of the data needs to be greater than 1%, then the number of submissions in the past seven days is selected as the target indicator 108 to be analyzed.

[0049] The computing device 102 can determine a group of attribution links 110 corresponding to the target indicator 108. In some embodiments, the correspondence between the target indicator 108 and the group of attribution links 110 is pre-set, and the computing device 102 obtains the group of attribution links 110 from the link library based on the target indicator 108. In other embodiments, the group of attribution links 110 corresponding to the target indicator 108 is obtained by searching from the link library, for example, based on the degree of association between the target indicator 108 and the attribution links in the link library. For example, the correlation between the target indicator 108 and the keywords in the return logic in the attribution link is calculated to determine its degree of association with the attribution link. Then, the obtained degree of association is sorted to select a group of attribution links 110 that meet predetermined conditions. For example, a group of attribution links with the highest degree of association, such as 3 or 5, is selected. The above examples are only used to describe the present disclosure, and are not specific limitations of the present disclosure.

[0050] In some embodiments, each attribution link in the set of attribution links is directly or indirectly associated with the target indicator. In other words, an attribution link in the determined set of attribution links may be directly associated with the target indicator or indirectly associated with the target indicator.

[0051] Then, at block 204, the computing device 102 determines a set of indicator values ​​of a set of associated indicators associated with the target indicator based on a set of attribution links. In the process of analyzing the experimental results, in addition to obtaining the indicator value of the target indicator, it is also necessary to obtain the indicator values ​​of a set of associated indicators associated with the target indicator. Then, the computing device 102 uses the target indicator value of the target indicator and the set of indicator values ​​of the set of associated indicators for analyzing the target indicator.

[0052] In some embodiments, the computing device 102 obtains an associated indicator for each attribution link from each attribution link in the set of attribution links. Then, the computing device 102 may obtain an indicator value corresponding to the associated indicator in the experimental result 106. Thus, the computing device 102 may determine a set of indicator values ​​for a set of associated indicators in the set of attribution links.

[0053] In some embodiments, the indicator value of the indicator is the rate of change of the experimental data obtained in the experiment, which is within a certain threshold range, for example, within the range of -100% to 100%, and includes the endpoint value. Additionally, the numerical value of the indicator value is a floating point value and retains two decimal places, for example, one of the indicator values ​​in a set of indicator values ​​is 7.56%. In some embodiments, the indicator value of the indicator is the amount of change of the experimental data obtained in the experiment. The range of the amount of change can be set by the user as needed.

[0054] In some embodiments, the indicator value of each attribution link in a group of attribution links has a corresponding threshold range. Additionally, the threshold ranges of the indicator values ​​of some indicators are different, and the threshold ranges of the indicator values ​​of another part of indicators are the same.

[0055] Additionally, a set of indicator values ​​in the experimental results also includes a direction, and the direction of the indicator value of the indicator in the experimental results may be consistent with or inconsistent with the reference direction in a set of attribution links. Additionally, the direction of the indicator is divided into positive and negative. For example, when the indicator value is positive, the direction of the indicator is positive, and when the indicator value is negative, the direction of the indicator is negative.

[0056] Finally, at block 206, the computing device 102 generates an analysis result for the target indicator based on the target indicator value, the set of attribution links, and the set of indicator values ​​for the target indicator. After determining the target indicator value for the target indicator and the set of indicator values ​​for the set of associated indicators, the target indicator is also analyzed in conjunction with the set of attribution links.

[0057] In some embodiments, the computing device 102 applies the target indicator value, a set of attribution links, and a set of indicator values ​​to a machine learning model to generate an analysis result for the target indicator. For example, the machine learning model can be any suitable neural network model, such as a large language model. In some embodiments, a mapping relationship between indicator values, attribution links, and analysis results is preset. Therefore, based on the mapping relationship, the analysis result of the target indicator corresponding to the target indicator value, a set of attribution links, and a set of indicator values ​​of the target indicator can be determined. The above examples are only used to describe the present disclosure, and are not specific limitations of the present disclosure.

[0058] In some embodiments, when generating analysis results for a target indicator, a set of indicator values ​​of a set of associated indicators associated with the target indicator will also be filtered, and then the analysis results are generated based on the filtered set of indicator values, the target indicator value and a set of attribution links.

[0059] In some embodiments, when filtering a set of indicator values, the indicator values ​​with significant changes can be filtered out from the set of indicator values. For example, when filtering a set of indicator values, first determine whether the indicator value in a set of indicator values ​​is less than a threshold value. For the convenience of description, the indicator value may also be referred to as a first indicator value. If the first indicator value is less than the threshold value, it indicates that the change amplitude of the first indicator value has not reached the threshold value, indicating that the change of the first indicator value is not significant, and then the first indicator value is removed from the set of indicator values. Additionally, the attribution link corresponding to it may also be removed. When the first indicator value is greater than the threshold value, it indicates that the change amplitude of the first indicator value has reached the threshold value, indicating that the change of the first indicator value is significant, and then the first indicator value is retained in a set of indicator values ​​for subsequent analysis. Additionally, the threshold value can be set by itself according to the analysis requirements.

[0060] In some embodiments, when filtering a set of indicator values, the computing device 102 may first determine the actual direction of the indicator value in a set of indicator values. For the convenience of description, the indicator value may also be referred to as a second indicator value. Then, the computing device 102 determines whether the reference direction in the attribution link associated with the indicator corresponding to the second indicator value is consistent. If not, the second indicator value is removed from a set of indicator values. Additionally, the attribution link corresponding to it may also be removed. For example, if the indicator in the attribution link is negative, and the actual indicator value obtained is positive, it indicates that the two are inconsistent. At this time, the attribution link and the corresponding indicator value may not be used in the analysis process. If the actual direction in the second indicator value is consistent with the reference direction in the attribution link associated with the indicator corresponding to the second indicator value, the second indicator value is retained for subsequent analysis.

[0061] Additionally, indicator values ​​whose indicator values ​​are less than a threshold value and indicator values ​​whose actual directions are inconsistent with reference directions may be filtered to be removed from a group of indicator values, so as to better perform targeted analysis on the experimental results.

[0062] In some embodiments, only target indicators with positive changes in indicator values ​​can be selected for analysis. In other embodiments, only target indicators with negative changes in indicator values ​​can be selected for analysis.

[0063] In some embodiments, a set of dimensional information associated with the target indicator can also be determined based on a set of attribution links, and then the target indicator value, a filtered set of indicator values, a set of attribution links and a set of dimensional information are used to analyze the target indicator in the experimental results, thereby generating analysis results for the target indicator.

[0064] The dimension information in a set of dimension information includes some limiting information, such as geographic area, preference information or education level. Therefore, when performing analysis, the above-mentioned dimension information can be combined for result analysis.

[0065] In some embodiments, when the analysis results do not meet expectations, the analysis process needs to be repeated, and the corresponding background information is supplemented before the analysis process is repeated. After the corresponding background information for the target experiment is supplemented, the analysis results for the target indicator are regenerated.

[0066] In some embodiments, before performing the analysis, the computing device 102 obtains background information associated with the target experiment. Then, the computing device 102 generates an analysis result for the target indicator based on the background information, the target indicator value, a set of filtered indicator values, a set of attribution links, and a set of dimension information.

[0067] Additionally, the computing device 102 generates an analysis result for the target indicator by applying the background information, the target indicator value, the filtered set of indicator values, the set of attribution links, and the set of dimension information to the machine learning model. In one example, the machine learning model is a large language model. In another example, the machine learning model can also be other types of machine learning models that can be used to implement the present disclosure.

[0068] In some embodiments, a mapping relationship between various information and analysis results is preset. Then, the computing device 102 can use the mapping relationship to obtain analysis results for the target indicator corresponding to the background information, the target indicator value, the filtered set of indicator values, the set of attribution links, and the set of dimension information. The above examples are only used to describe the present disclosure, and are not specific limitations of the present disclosure.

[0069] Through this method, by utilizing the target indicators in the experimental results, a set of attribution links is obtained from the link library, and a set of associated indicators and a set of indicator values ​​are further determined therefrom. The obtained data is cleaned and filtered, and then applied to the machine learning model to obtain analysis results for the target indicators, thereby improving the user experience.

[0070] Combined with the above Figure 2 A schematic diagram of an example method for analyzing data according to some disclosed embodiments is described below. Figure 3 A schematic diagram depicting an example of a system architecture for analyzing data according to some embodiments of the present disclosure. Figure 3 An example of Figure 1 The system may be executed by the computing device 102 shown in FIG. 1 or any suitable device.

[0071] like Figure 3 In the example 300 shown, first, an analysis link page 302 is displayed to the user on the user interface. Then, at 304, the user can enter the attribution links obtained from the historical experiment report and the analysis of the historical experiment into the link library 306 through the analysis link page 302. Therefore, a large number of attribution links are entered into the link library 306.

[0072] When the analysis of the target experiment begins, the report page 314 will obtain the experimental results of the target experiment. Then, the computing device 102 also determines whether to analyze the indicator according to the indicator value of the indicator in the experimental result. For example, the computing device 102 determines whether to analyze the indicator according to the indicator value of the indicator. If analysis is required, at 316, a request is triggered.

[0073] Then, at 308, the analysis service 320 obtains a set of attribution links corresponding to the target indicator from the link library 306. The associated data of the experimental results for the target experiment are stored in the database 312. At 310, the computing device 102 determines that the link library 306 and the analysis service 320 rely on the database 312 to obtain data.

[0074] When the analysis service 320 obtains a set of data on return links, target indicators, and a set of data corresponding to indicators in the attribution links, it will call the large language model 324 for processing at 322, and then return the conclusion to the report page 314 at 318 to generate analysis results for the target experiment.

[0075] In some embodiments, when the analysis result does not meet the expected result, a request 316 is triggered to the analysis service 320, and the analysis service 320 obtains the background information for the target experiment to re-analyze the target experiment.

[0076] Combined with the above Figure 3 A schematic diagram of an example of a system architecture for analyzing data according to some disclosed embodiments is described below. Figure 4 A schematic diagram of a flow chart for analyzing data according to some embodiments of the present disclosure is shown. The flow chart may be composed of Figure 1 Executed by computing device 102 or any suitable computing device.

[0077] like Figure 4In the example 400 shown, at block 402, the computing device 102 starts the experimental analysis, and then the computing device 102 analyzes the experimental results. At block 404, the computing device 102 obtains the indicator data to be analyzed from the experimental results, for example, determining the target indicator and the target indicator value to be analyzed based on the indicator values ​​of a group of indicators.

[0078] Then, at block 406, the computing device 102 obtains relevant attribution links from the experience library, that is, obtains a set of attribution links associated with the target indicator. After obtaining a set of attribution links, at block 408, the computing device 102 continues to obtain associated indicator data, that is, obtains indicator data of the indicators in a set of attribution links.

[0079] After obtaining the target indicator value, a set of attribution links, and a set of indicator values, data aggregation is performed at box 410. One step of data aggregation is to filter the obtained data to obtain data that meets the analysis requirements. For example, indicator values ​​whose indicator values ​​are less than a threshold and indicator values ​​whose actual directions and reference directions are inconsistent can be filtered to remove them from a set of indicator values, so as to better conduct targeted analysis of the experimental results.

[0080] Another step in data aggregation is to clean the acquired data, such as removing some dirty data and screening duplicate data, so as to ensure the quality and accuracy of the data and the completeness and correctness of the analysis results.

[0081] After the data aggregation is completed, at box 412, it is necessary to call the large language model to apply the obtained data to the large language model to analyze the experimental results. It is understandable that other machine learning models that can achieve the same effect can be used to replace the large language model in the present disclosure.

[0082] Then, after being processed by the large language model, at box 414, the computing device 102 analyzes the conclusion output. When the conclusion does not meet expectations, at box 416, the user background information can be supplemented, and then the analysis process for the experimental results is repeated. When the analysis conclusion meets the requirements, at box 418, the conclusion report is exported. Then, at box 420, the analysis process for the experimental results ends.

[0083] Combined with the above Figure 4 A schematic diagram of a flow chart for analyzing data according to some disclosed embodiments is described below. Figure 5 A schematic diagram of another flow chart describing some embodiments of the present disclosure. The flow chart may be composed of Figure 1 Executed by computing device 102 or any suitable device.

[0084] like Figure 5 As shown in example 500, user 502 can enter the analysis experience into experience library 506 at 510, wherein the analysis experience includes the attribution link, and the experience library can also be referred to as a link library. After receiving the analysis experience entry, experience library 506 can save the attribution link in the analysis experience. The attribution link can also be referred to as an analysis link. Then, at 512, a response is returned to user 502 to indicate that the entry is complete. If user 502 obtains experimental results for an executed experiment, such as a comparative experiment, at 514, user 502 triggers indicator analysis and starts analyzing the experimental results.

[0085] In this process, first at 516, the analysis service 504 sends a request to obtain the analysis link to the experience library 506. Then, at 518, the experience library 506 returns the analysis link according to the request, and at this time, returns a set of analysis links or attribution links to the analysis service 504. Afterwards, at 520, the analysis service 504 queries the relevant indicator data, that is, queries a set of indicator values ​​of a set of associated indicators associated with the target indicator.

[0086] After data aggregation of the obtained indicator values ​​and a set of attribution links, at block 522, the analysis service 504 sends a reasoning analysis request including the aggregated data to the machine learning model 508. In this process, the aggregated structured data can be converted into semantic data and then applied to the machine learning model 508 for analysis, and then at 524, the machine learning model 508 returns the reasoning conclusion.

[0087] After obtaining the inference conclusion, at 526, the analysis service 504 generates an analysis conclusion based on the inference conclusion and sends it to the user, for example, using the inference conclusion as the analysis conclusion. When the user 502 is not satisfied with the analysis conclusion or the analysis conclusion does not meet the user's expectations, at 528, the user 502 may also add other background information 528, and the added background information may be information related to the purpose of the experiment or the operation of the experiment.

[0088] After supplementing the background information, at 530, the user triggers the indicator analysis again to re-analyze the target indicator in the experimental results. Subsequently, at 532, the analysis service 504 performs the analysis process 532 again. After repeating the previous analysis process, at 534, the analysis conclusion is returned again.

[0089] Subsequently, at 536 , the user 502 may send a request for exporting an analysis report to the analysis service 504 , and then the analysis service 504 returns an analysis report 538 to the user, completing the analysis of the target indicator of the experimental results.

[0090] like Figure 6As shown, the device 600 includes an attribution link determination module 602, which is configured to determine a set of attribution links for the target indicator based on the target indicator in the experimental results of the target experiment, wherein the attribution link describes the change of the indicator and the attribution logic for the indicator; an indicator value determination module 604, which is configured to determine a set of indicator values ​​of a set of associated indicators associated with the target indicator based on a set of attribution links; and an analysis result generation module 606, which is configured to generate an analysis result for the target indicator based on the target indicator value of the target indicator, a set of attribution links and a set of indicator values.

[0091] In some embodiments, the device 600 also includes: an experimental result determination module, configured to determine the experimental results for the target experiment, the experimental results including a set of indicators; and a target indicator selection module, configured to select a target indicator from a set of indicators based on the indicator value of the indicator in the set of indicators.

[0092] In some embodiments, the attribution link determination module 602 includes: a link library determination module, configured to determine a link library for the target indicator in response to the target indicator in the experimental results having been determined, the link library storing multiple attribution links; and a group of attribution link acquisition modules, configured to acquire a group of attribution links associated with the target indicator from the multiple attribution links.

[0093] In some embodiments, the indicator value determination module 604 includes: an associated indicator acquisition module, configured to obtain an associated indicator in each attribution link in a set of attribution links; and a set of indicator value determination modules, configured to determine the indicator value of the associated indicator from the experimental results based on the associated indicator.

[0094] In some embodiments, the analysis result generation module 606 includes: a filtering module configured to filter a set of indicator values; and an analysis result generation module configured to generate an analysis result for the target indicator based on the target indicator value, the filtered set of indicator values ​​and a set of attribution links.

[0095] In some embodiments, the filtering module includes: a first indicator value less than a threshold determination module, configured to determine whether a first indicator value in a set of indicator values ​​is less than a threshold; and a first indicator value removal module, configured to remove the first indicator value from a set of indicator values ​​in response to the first indicator value being less than the threshold.

[0096] In some embodiments, the filtering module includes: a module for determining whether the actual direction of the second indicator value is consistent with a reference direction in an attribution link, configured to determine whether the actual direction of a second indicator value in a set of indicator values ​​is consistent with the reference direction in an attribution link associated with the indicator corresponding to the second indicator value; and a second indicator value removal module, configured to remove the second indicator value from a set of indicator values ​​in response to the actual direction being inconsistent with the reference direction.

[0097] In some embodiments, the analysis result generation module includes: a set of dimensional information determination modules, configured to determine a set of dimensional information associated with the target indicator based on a set of attribution links; and an analysis result generation module, configured to generate analysis results for the target indicator based on the target indicator value, a filtered set of indicator values, a set of attribution links and a set of dimensional information.

[0098] In some embodiments, the device 600 also includes: a module for determining whether the analysis results meet expectations, configured to determine whether the analysis results meet expectations; and a background information adding module, configured to regenerate the analysis results by adding background information for the target experiment in response to the analysis results not meeting expectations.

[0099] In some embodiments, the analysis result generation module includes: a background information acquisition module, configured to acquire background information associated with a target experiment; and an analysis result generation module, configured to generate analysis results for the target indicator based on the background information, the target indicator value, a filtered set of indicator values, a set of attribution links, and a set of dimensional information.

[0100] In some embodiments, the analysis result generation module includes: a machine learning model application module, configured to generate analysis results for the target indicator by applying background information, target indicator values, a filtered set of indicator values, a set of attribution links, and a set of dimensional information to the machine learning model.

[0101] Figure 7 A schematic block diagram of an example device 700 that may be used to implement embodiments of the present disclosure is shown. Figure 1 The computing device 102 in the embodiment of the present invention can be implemented by using the device 700. As shown in the figure, the device 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 702 or computer program instructions loaded from a storage unit 808 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0102] A number of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage page 708, such as a disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks. High pitch technique, voice training, articulation, breathing, falsetto and correct singing method 2.

[0103] The various processes and processing described above, such as method 200 and examples 300, 400, and 500, may be performed by processing unit 701. For example, in some embodiments, method 200 and examples 300, 400, and 500 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and / or installed on device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by CPU 701, one or more actions of the example methods 200 and examples 300, 400, and 500 described above may be performed.

[0104] The present disclosure may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present disclosure.

[0105] Computer readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (non-exhaustive list) of computer readable storage medium include: portable computer disk, hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disk read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanical encoding device, such as a punch card or a convex structure in a groove on which instructions are stored, and any suitable combination of the above. The computer readable storage medium used here is not interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagated by a waveguide or other transmission medium (e.g., a light pulse by an optical fiber cable), or an electrical signal transmitted by a wire.

[0106] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0107] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via 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., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0108] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0109] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0110] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0111] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse 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 the specified function or action, or can be implemented with a combination of special hardware and computer instructions.

[0112] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for analyzing data, comprising: Based on the target indicator in the experimental result of the target experiment, determine a set of attribution links for the target indicator, wherein the attribution link describes the change of the indicator and the attribution logic for the indicator; Determining a set of indicator values ​​for a set of associated indicators associated with the target indicator based on the set of attribution links; as well as Based on the target indicator value of the target indicator, the set of attribution links and the set of indicator values, an analysis result for the target indicator is generated.

2. The method according to claim 1, further comprising: Determine an experimental result for the target experiment, wherein the experimental result includes a set of indicators; as well as The target indicator is selected from the set of indicators based on indicator values ​​of indicators in the set of indicators.

3. The method of claim 1 , wherein determining a set of attribution links for the target metric comprises: In response to the target indicator in the experimental result being determined, determining a link library for the target indicator, the link library storing a plurality of attribution links; as well as The set of attribution links associated with the target metric is obtained from the multiple attribution links.

4. The method according to claim 1, wherein determining a set of indicator values ​​of a set of associated indicators associated with the target indicator comprises: Obtaining a correlation indicator in each attribution link in the set of attribution links; as well as Based on the correlation indicator, an indicator value for the correlation indicator is determined from the experimental result.

5. The method according to claim 1, wherein generating the analysis result for the target indicator comprises: filtering the set of indicator values; as well as An analysis result for the target indicator is generated based on the target indicator value, the filtered set of indicator values, and the set of attribution links.

6. The method according to claim 5, wherein filtering the set of indicator values ​​comprises: Determining whether a first indicator value in the set of indicator values ​​is less than a threshold value; as well as In response to the first indicator value being less than the threshold, the first indicator value is removed from the set of indicator values.

7. The method according to claim 5, wherein filtering the set of indicator values ​​comprises: determining whether an actual direction of a second indicator value in the set of indicator values ​​is consistent with a reference direction in an attribution link associated with an indicator corresponding to the second indicator value; as well as In response to the actual direction not being consistent with the reference direction, the second indicator value is removed from the set of indicator values.

8. The method according to claim 5, wherein generating an analysis result for the target indicator based on the target indicator value, the filtered set of indicator values ​​and the set of attribution links comprises: Based on the set of attribution links, determining a set of dimensional information associated with the target indicator; as well as Based on the target indicator value, the filtered set of indicator values, the set of attribution links and the set of dimension information, an analysis result for the target indicator is generated.

9. The method according to claim 8, further comprising: Determining whether the analysis results meet expectations; as well as In response to the analysis result not meeting expectations, the analysis result is regenerated by adding background information for the target experiment.

10. The method according to claim 8, wherein generating an analysis result for the target indicator based on the target indicator value, the filtered set of indicator values, the set of attribution links and the set of dimension information comprises: Obtaining background information associated with the target experiment; as well as Based on the background information, the target indicator value, the filtered set of indicator values, the set of attribution links and the set of dimension information, an analysis result for the target indicator is generated.

11. The method according to claim 10, wherein generating an analysis result for the target indicator based on the background information, the target indicator value, the filtered set of indicator values, the set of attribution links and the set of dimension information comprises: The analysis result for the target indicator is generated by applying the background information, the target indicator value, the filtered set of indicator values, the set of attribution links and the set of dimensional information to a machine learning model. The method according to claim 1 , wherein the target experiment is a comparative experiment.

13. A device for analyzing data, comprising: An attribution link determination module is configured to determine a set of attribution links for the target indicator based on the target indicator in the experimental result of the target experiment, wherein the attribution link describes the change of the indicator and the attribution logic for the indicator; an indicator value determination module, configured to determine a set of indicator values ​​of a set of associated indicators associated with the target indicator based on the set of attribution links; as well as The analysis result generating module is configured to generate an analysis result for the target indicator based on the target indicator value of the target indicator, the set of attribution links and the set of indicator values.

14. An electronic device, comprising: at least one processor; as well as A storage device, used for storing at least one program, when the at least one program is executed by the at least one processor, so that the at least one processor implements the method according to any one of claims 1-12.

15. A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method according to any one of claims 1 to 12 when executed by a processor.

16. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 12.

Images