Data analysis method, device, electronic device and storage medium

Abstract

The present disclosure relates to a data analysis method, device, electronic device, and storage medium. The method includes: obtaining a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period; determining first target indicator data and second target indicator data; determining first weighted target indicator data and second weighted target indicator data based on the first display data set, the second display data set, the first click data set, and the second click data set; determining data volatility based on the first target indicator data and the second target indicator data; and determining a data influencing factor and a structural influencing factor based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility. The present disclosure implements analysis of the influence of structural factors and intrinsic factors on data volatility, thereby improving analysis accuracy.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a data analysis method, device, electronic device, and storage medium. Background Art

[0002] Target indicator data is a key indicator for measuring the effectiveness of information delivery, and it is also an aggregate indicator. When significant fluctuations in target indicator data are observed, we hope to find the reasons for the data fluctuations.

[0003] In related technologies, when the effectiveness of information delivery changes, in order to analyze the cause of the change, the impact factors can be split into multiple categories corresponding to the target indicator data to obtain the impact factors of the data fluctuation in each category, that is, the data fluctuation caused by each category can be obtained. For example, if the information delivery involves three categories: category 1, category 2, and category 3, when performing data fluctuation analysis, the impact factors can be split based on these three categories to obtain the data fluctuation caused by each category. However, this analysis method is not accurate. Summary of the Invention

[0004] The present disclosure provides a data analysis method, device, electronic device, and storage medium to at least address the problem of inaccurate data analysis in related technologies. The technical solutions of the present disclosure are as follows:

[0005] According to a first aspect of an embodiment of the present disclosure, there is provided a data analysis method, comprising:

[0006] Acquire data to be analyzed, the data to be analyzed comprising a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period;

[0007] Determining first target indicator data based on the first click data set and the first display data set, and determining second target indicator data based on the second click data set and the second display data set;

[0008] Determining first weighted target indicator data and second weighted target indicator data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set;

[0009] determining a data volatility rate according to the first target indicator data and the second target indicator data;

[0010] According to the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility, the data impact factor and the structure impact factor of the fluctuation of the first target indicator data relative to the second target indicator data are determined.

[0011] Optionally, determining a data impact factor and a structure impact factor of the fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility includes:

[0012] Determine the difference between the first weighted target indicator data and the second weighted target indicator data as a first difference, and determine the difference between the first target indicator data and the second target indicator data as a second difference;

[0013] determining a data impact weight of the data volatility according to the first difference and the second difference;

[0014] The data impact factor and the structure impact factor are determined according to the data volatility and the data impact weight.

[0015] Optionally, determining the data impact factor and the structure impact factor according to the data volatility and the data impact weight includes:

[0016] Determine the product of the data volatility and the data impact weight as the data impact factor;

[0017] The difference between the data volatility and the data impact factor is determined as the structural impact factor, or the difference between 1 and the data impact weight is determined as the structural impact weight, and the product of the data volatility and the structural impact weight is determined as the structural impact factor.

[0018] Optionally, determining first weighted target indicator data and second weighted target indicator data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set includes:

[0019] Determining a data influencing factor weight set of fluctuations of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set;

[0020] The first weighted target indicator data is determined based on the first click data set, the first display data set and the data influence factor weight set, and the second weighted target indicator data is determined based on the second click data set, the second display data set and the data influence factor weight set.

[0021] Optionally, the first click data set includes first click data under multiple categories, the first display data set includes first display data under the multiple categories, the second click data set includes second click data under the multiple categories, the second display data set includes second display data under the multiple categories, and the data influence factor weight set includes data influence factor weights under the multiple categories.

[0022] Optionally, determining first weighted target indicator data based on the first click data set, the first display data set, and the data influence factor weight set, and determining second weighted target indicator data based on the second click data set, the second display data set, and the data influence factor weight set includes:

[0023] Determine target indicator data for each category in a first statistical period based on the first click data and the first display data under the multiple categories, and determine the target indicator data for each category in the first statistical period as first category indicator data;

[0024] Determining target index data for each category in a second statistical period based on the second click data and the second display data under the multiple categories, and determining the target index data for each category in the second statistical period as second category index data;

[0025] The first classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the first weighted target indicator data, and the second classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the second weighted target indicator data.

[0026] Optionally, determining a set of data influencing factor weights of fluctuations of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set includes:

[0027] Determining data factor influencing data under each category based on the first display data and the second display data under the multiple categories;

[0028] The sum of the data factors affecting all categories is determined as the total data factors affecting data;

[0029] The ratio of the data factor influence data under each category to the total data factor influence data is determined as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

[0030] Optionally, determining the data factor influencing data under each category based on the first display data and the second display data under the multiple categories includes:

[0031] The ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category is determined as the data factor influence data under the category.

[0032] According to a second aspect of an embodiment of the present disclosure, there is provided a data analysis device, comprising:

[0033] a data acquisition module configured to acquire data to be analyzed, wherein the data to be analyzed includes a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period;

[0034] a target indicator data determination module, configured to determine first target indicator data based on the first click data set and the first display data set, and determine second target indicator data based on the second click data set and the second display data set;

[0035] a weighted index data determination module, configured to determine first weighted target index data and second weighted target index data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set;

[0036] a data volatility determination module, configured to determine the data volatility based on the first target indicator data and the second target indicator data;

[0037] The impact factor splitting module is configured to determine the data impact factor and structural impact factor of the fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility.

[0038] Optionally, the impact factor splitting module includes:

[0039] a difference determining unit configured to determine a difference between the first weighted target indicator data and the second weighted target indicator data as a first difference, and to determine a difference between the first target indicator data and the second target indicator data as a second difference;

[0040] a self-weight determination unit, configured to determine a data impact weight of the data volatility based on the first difference and the second difference;

[0041] The impact factor splitting unit is configured to determine the data impact factor and the structure impact factor according to the data volatility and the self-influence weight.

[0042] Optionally, the impact factor splitting unit is configured to execute:

[0043] Determine the product of the data volatility and the data impact weight as the data impact factor;

[0044] The difference between the data volatility and the data impact factor is determined as the structural impact factor, or the difference between 1 and the data impact weight is determined as the structural impact weight, and the product of the data volatility and the structural impact weight is determined as the structural impact factor.

[0045] Optionally, the weighted index data determination module includes:

[0046] a factor weight set determining unit configured to determine a data influencing factor weight set of a fluctuation of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set;

[0047] The weighted indicator data determination unit is configured to determine the first weighted target indicator data based on the first click data set, the first display data set and the data influence factor weight set, and to determine the second weighted target indicator data based on the second click data set, the second display data set and the data influence factor weight set.

[0048] Optionally, the first click data set includes first click data under multiple categories, the first display data set includes first display data under the multiple categories, the second click data set includes second click data under the multiple categories, the second display data set includes second display data under the multiple categories, and the data influence factor weight set includes data influence factor weights under the multiple categories.

[0049] Optionally, the weighted index data determining unit is configured to execute:

[0050] Determine target indicator data for each category in a first statistical period based on the first click data and the first display data under the multiple categories, and determine the target indicator data for each category in the first statistical period as first category indicator data;

[0051] Determining target index data for each category in a second statistical period based on the second click data and the second display data under the multiple categories, and determining the target index data for each category in the second statistical period as second category index data;

[0052] The first classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the first weighted target indicator data, and the second classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the second weighted target indicator data.

[0053] Optionally, the factor weight set determination unit includes:

[0054] a data factor influence data determination subunit, configured to determine the data factor influence data under each category based on the first display data and the second display data under the multiple categories;

[0055] The total self-influence data determining subunit is configured to determine the sum of the data factor influence data under all categories as the total data factor influence data;

[0056] The factor weight set determination subunit is configured to determine the ratio of the data factor influence data under each category to the total data factor influence data as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

[0057] Optionally, the data factor impact data determination subunit is configured to execute:

[0058] The ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category is determined as the data factor influence data under the category.

[0059] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0060] processor;

[0061] a memory for storing instructions executable by the processor;

[0062] The processor is configured to execute the instructions to implement the data analysis method as described in the first aspect.

[0063] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When instructions in the computer storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the data analysis method as described in the first aspect.

[0064] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, comprising a computer program / instruction, which implements the data analysis method described in the first aspect when executed by a processor.

[0065] The technical solutions provided by the embodiments of the present disclosure bring at least the following beneficial effects:

[0066] In the embodiment of the present disclosure, since the first weighted target indicator data and the second weighted target indicator data associated with the data are determined to be target indicator data unrelated to structural changes, the data volatility is split into structural influencing factors and data influencing factors based on the weighted target indicator data and the target indicator data, thereby realizing the analysis of the structural factor influence and the data's own factor influence on data fluctuation, solving the problem in related technologies that the structural factor influence and the data's own factor influence on data fluctuation cannot be analyzed, and improving the accuracy of data analysis.

[0067] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] The accompanying drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute an improper limitation of the present disclosure.

[0069] Figure 1 is a flow chart showing a data analysis method according to an exemplary embodiment;

[0070] Figure 2 is a block diagram of a data analysis device according to an exemplary embodiment;

[0071] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0072] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0073] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0074] Figure 1 is a flow chart of a data analysis method according to an exemplary embodiment. Figure 1 As shown, the data analysis method is used in electronic devices such as computers or servers, and includes the following steps.

[0075] In step S11 , data to be analyzed is obtained, where the data to be analyzed includes a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period.

[0076] Click data and display data are the foundational data for determining target indicator data. Display data can be, for example, the number of times a piece of information has been exposed, and click data can be, for example, the number of times a piece of information has been clicked by users after exposure. A first click data set is a set of first click data, which is click data within the first statistical period. A first display data set is a set of first display data, which is display data within the first statistical period. A second click data set is a set of second click data, which is click data within the second statistical period. A second display data set is a set of second display data, which is display data within the second statistical period.

[0077] A statistical period is a period for data statistics, and may be, for example, a moment, a day, a week, or a month. The first statistical period and the second statistical period are two statistical periods for determining data volatility, and the units of these two statistical periods are the same, namely, a moment, a day, a week, or a month. The data volatility may be, for example, a month-on-month change or a year-on-year change. The first statistical period and the second statistical period may be, for example, two statistical periods for calculating a month-on-month change, e.g., the first statistical period is the current day and the second statistical period is the previous day, or two statistical periods for calculating a year-on-year change, e.g., the first statistical period is Tuesday of the current week and the second statistical period is Tuesday of the previous week, etc.

[0078] In step S12, first target indicator data is determined based on the first click data set and the first display data set, and second target indicator data is determined based on the second click data set and the second display data set.

[0079] The first click data in the first click data set are summed up to obtain the first total click data, the first display data in the first display data set are summed up to obtain the first total display data, and the quotient of the first total click data and the first total display data is determined as the first target indicator data.

[0080] The second click data in the second click data set are summed up to obtain the second total click data, the second display data in the second display data set are summed up to obtain the second total display data, and the quotient of the second total click data and the second total display data is determined as the second target indicator data.

[0081] In step S13, first weighted target indicator data and second weighted target indicator data associated with the data are determined based on the first display data set, the second display data set, the first click data set, and the second click data set.

[0082] The first and second weighted target indicator data are target indicator data obtained based on the influence of factors within the data itself, excluding the influence of structural changes. Structural changes refer to changes in the aggregation dimensions of the target indicator data.

[0083] Since the display data is data that is unrelated to structural changes and is associated with the data's own factors, the weight associated with the data can be determined based on the first display data set and the second display data set, and the first weighted target indicator data associated with the data can be obtained based on the weight, the first click data set and the first display data set, and the second weighted target indicator data associated with the data can be obtained based on the weight, the second click data set and the second display data set.

[0084] In an exemplary embodiment, determining first weighted target metric data and second weighted target metric data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set includes:

[0085] Based on the first display data set and the second display data set, determine the data influencing factor weight set for the fluctuation of the first target indicator data relative to the second target indicator data; based on the first click data set, the first display data set and the data influencing factor weight set, determine the first weighted target indicator data, and based on the second click data set, the second display data set and the data influencing factor weight set, determine the second weighted target indicator data.

[0086] The data influence factor weight set includes multiple data influence factor weights, each of which corresponds to the first display data in the first display data set and the second display data in the second display data set. The data influence factor weight is the weight that the data's own factors have on the data's fluctuation.

[0087] The first display data in the first display data set and the second display data in the second display data set are calculated according to a weight calculation model to obtain a data influence factor weight set of fluctuation of the first target indicator data relative to the second target indicator data.

[0088] According to the first click data set and the first display data set, the quotient of the corresponding first click data and the first display data is determined, and the determined quotient is weighted and summed with the corresponding data influence factor weight in the data influence factor weight set to obtain the first weighted target indicator data.

[0089] According to the second click data set and the second display data set, the quotient of the corresponding second click data and the second display data is determined, and the determined quotient is weightedly summed with the corresponding data influence factor weight in the data influence factor weight set to obtain the second weighted target indicator data.

[0090] By determining the data influence factor weight set based on the first display data set and the second display data set, and then determining the first weighted target indicator data and the second weighted target indicator data based on the data influence factor weight set, more accurate first weighted target indicator data and second weighted target indicator data associated with the data can be obtained.

[0091] In step S14, the data volatility is determined according to the first target indicator data and the second target indicator data.

[0092] A difference between the first target indicator data and the second target indicator data is determined, and a ratio of the difference to the second target indicator data is determined as the data volatility.

[0093] In step S15, the data impact factor and the structural impact factor of the fluctuation of the first target indicator data relative to the second target indicator data are determined based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility.

[0094] The data impact factor represents the volatility caused by changes in data factors, and the structure impact factor represents the volatility caused by changes in data structure.

[0095] Based on the first target indicator data, the second target indicator data, the first weighted target indicator data, and the second weighted target indicator data, the data volatility is split into a data impact factor and a structural impact factor. The first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility are substituted into the data impact factor calculation model to obtain the data impact factor of the first target indicator data relative to the second target indicator data. The first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility are substituted into the structural impact factor calculation model to obtain the structural impact factor of the first target indicator data relative to the second target indicator data.

[0096] The absolute value of the impact factor reflects the relative size of the impact factor, and the positive or negative value of the impact factor reflects whether the impact factor and the data volatility change in the same direction (same sign means same direction change, opposite sign means opposite direction change). The sum of the data impact factor and the structural impact factor is exactly equal to the data volatility. In this way, it is possible to quickly determine whether the fluctuation of the target indicator data is caused by structural factors or the fluctuation of the target indicator data itself.

[0097] In an exemplary embodiment, based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility, the data impact factor and the structural impact factor of the fluctuation of the first target indicator data relative to the second target indicator data are determined, including: determining the difference between the first weighted target indicator data and the second weighted target indicator data as the first difference, and determining the difference between the first target indicator data and the second target indicator data as the second difference; determining the data impact weight of the data volatility based on the first difference and the second difference; determining the data impact factor and the structural impact factor based on the data volatility and the data impact weight.

[0098] Calculate the difference between the first weighted target indicator data and the second weighted target indicator data to obtain the first difference, which is the fluctuation value caused by data fluctuation; calculate the difference between the first target indicator data and the second target indicator data to obtain the second difference, which is the fluctuation value caused by both data fluctuation and structural fluctuation; calculate the ratio of the first difference to the second difference, which is the weight of the data fluctuation, that is, the data impact weight; multiply the data volatility by the data impact weight to obtain the data impact factor, and the difference between the data volatility and the data impact factor is the structural impact factor.

[0099] By determining the ratio of the first difference to the second difference as the data impact weight of the data volatility, the data impact factor and the structure impact factor can be accurately determined.

[0100] In an exemplary embodiment, the data impact factor and the structural impact factor are determined based on the data volatility and the data impact weight, including: determining the product of the data volatility and the data impact weight as the data impact factor; determining the difference between the data volatility and the data impact factor as the structural impact factor, or determining the difference between 1 and the data impact weight as the structural impact weight, and determining the product of the data volatility and the structural impact weight as the structural impact factor.

[0101] After obtaining the data impact weight, the data volatility can be multiplied by the data impact weight to obtain the data impact factor. That is, the data impact factor can be calculated using the following formula:

[0102]

[0103] Among them, CPM part Indicates the data impact factor, MHcpm t is the first weighted target indicator data, MHcpm y The second weighted target indicator data, cpm t The first target indicator data, cpm y is the second target indicator data, is the data impact weight, is the data volatility.

[0104] Since data volatility can be split into data impact factor and structural impact factor, the difference between data volatility and data impact factor can be determined as the structural impact factor. Alternatively, the difference between 1 and the data impact weight can be calculated to obtain the structural impact weight, and the data volatility and structural impact weight can be multiplied to obtain the structural impact factor. The structural impact factor can be calculated using the following formula:

[0105]

[0106] Among them, W part represents the structural impact factor, MHcpm t is the first weighted target indicator data, MHcpm y The second weighted target indicator data, cpm t The first target indicator data, cpm y is the second target indicator data, is the data impact weight, is the data volatility.

[0107] By determining the data impact factor based on data volatility and data impact weight, and obtaining the structural impact factor based on the relationship between the data impact factor and the structural impact factor that together constitute the data volatility, more accurate data impact factors and structural impact factors can be obtained, thereby improving the accuracy of data fluctuation analysis.

[0108] The data analysis method provided by this exemplary embodiment determines the first target indicator data according to the first click data set and the first display data set after obtaining the data to be analyzed, and determines the second target indicator data according to the second click data set and the second display data set, and determines the first weighted target indicator data and the second weighted target indicator data associated with the data according to the first display data set, the second display data set, the first click data set and the second click data set, so that the first weighted target indicator data and the second weighted target indicator data that are only affected by the data itself can be obtained, the data volatility is determined according to the first target indicator data and the second target indicator data, and the first target indicator data and the second target indicator data are used to determine the data volatility. The indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility can determine the data influencing factor and the structural influencing factor of the fluctuation of the first target indicator data relative to the second target indicator data. Since the first weighted target indicator data and the second weighted target indicator data associated with the data are target indicator data unrelated to structural changes, the data volatility is split into the structural influencing factor and the data influencing factor based on the weighted target indicator data and the target indicator data, thereby realizing the analysis of the structural factor influence and the data's own factor influence on the data fluctuation, solving the problem in related technologies that the structural factor influence and the data's own factor influence on the data fluctuation cannot be analyzed, thereby improving the accuracy of data analysis.

[0109] Based on the above technical solution, the first click data set includes the first click data under multiple categories, the first display data set includes the first display data under the multiple categories, the second click data set includes the second click data under the multiple categories, the second display data set includes the second display data under the multiple categories, and the data influence factor weight set includes the data influence factor weights under the multiple categories.

[0110] In an exemplary embodiment, first weighted target indicator data is determined based on the first click data set, the first display data set and the data influence factor weight set, and second weighted target indicator data is determined based on the second click data set, the second display data set and the data influence factor weight set, including: determining the target indicator data of each category in the first statistical period based on the first click data and the first display data under the multiple categories, and determining the target indicator data of each category in the first statistical period as the first category indicator data; determining the target indicator data of each category in the second statistical period based on the second click data and the second display data under the multiple categories, and determining the target indicator data of each category in the second statistical period as the second category indicator data; performing weighted summation on the first category indicator data under the multiple categories and the data influence factor weight to obtain the first weighted target indicator data, and performing weighted summation on the second category indicator data under the multiple categories and the data influence factor weight to obtain the second weighted target indicator data.

[0111] The quotient of the first click data and the first display data under each category is determined as the first classification index data under the corresponding category, and the quotient of the second click data and the second display data under each category is determined as the second classification index data under the corresponding category.

[0112] According to the following formula, the first classification indicator data and the weight of the data impact factor under multiple categories are weighted and summed to obtain the first weighted target indicator data:

[0113]

[0114] Among them, MHcpm t is the first weighted target indicator data, a represents one of the multiple categories, cpm t,a is the first classification indicator data under classification a, weight a is the data impact factor weight under category a.

[0115] According to the following formula, the second classification indicator data under multiple categories and the weight of the data impact factor are weighted and summed to obtain the second weighted target indicator data:

[0116]

[0117] Among them, MHcpm y is the first weighted target indicator data, a represents one of the multiple categories, cpm y,a is the second classification indicator data under classification a, weight a is the data impact factor weight under category a.

[0118] The first weighted target indicator data and the second weighted target indicator data determined in the above manner are target indicator data that are irrelevant to structural changes, which facilitates the separation of data influencing factors and structural influencing factors.

[0119] On the basis of the above technical solution, according to the first display data set and the second display data set, a data influence factor weight set of the fluctuation of the first target indicator data relative to the second target indicator data is determined, including: according to the first display data and the second display data under the multiple categories, determining the data factor influence data under each category; determining the sum of the data factor influence data under all categories as the total data factor influence data; determining the ratio of the data factor influence data under each category to the total data factor influence data as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

[0120] The data impact factor weight for each category is calculated according to the following formula:

[0121]

[0122] Among them, weight a Indicates the weight of the data impact factor under category a, This is the first display data under category a. The second display data under category a, where a is one of multiple categories. The data factors under category a affect the data. The factors affecting the total data are:

[0123] Since the display data is unrelated to the change of the data structure, the data influence factor weight can be determined based on the first display data and the second display data, thereby improving the accuracy of determining the data influence factor weight.

[0124] On the basis of the above technical solution, the data factor influence data under each category is determined according to the first display data and the second display data under the multiple categories, including: determining the ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category as the data factor influence data under the category.

[0125] When determining the data factors affecting the data, the data is obtained by calculating the ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the corresponding category.

[0126] The following describes the derivation process of the above formula:

[0127] The calculation of weighted target indicator data is derived from the expanded formula of the ratio of target indicator data:

[0128]

[0129] Among them, cpm t The first target indicator data, cpm y The second target indicator data, cost_total t The first total click data under multiple categories, ad_show t The first total display data under multiple categories, cost_total y The second total click data under multiple categories, ad_show y It is the second total display data under multiple categories. is the first click data under category a, This is the first display data under category a. This is the second click data under category a. The second display data under category a, cpm t,a is the first classification indicator data under classification a, cpm y,a is the second classification indicator data under classification a, weight t,a is the first data weight under category a, weight y,a is the second data weight under category a.

[0130] In order to separate the fluctuations caused purely by changes in the target indicator data itself (excluding structural changes), the weights in the numerator and denominator are kept consistent. The resulting data impact factor weights are as follows:

[0131]

[0132] Among them, weight a Indicates the data impact factor weight under category a.

[0133] The first data weight and the second data weight in the expanded formula of the ratio of the above target indicator data are replaced by the data influence factor weights respectively, and the ratio of the first weighted target indicator data to the second weighted target indicator data is obtained as follows:

[0134]

[0135] Among them, MHcpm ratio It is the ratio of the first weighted target indicator data to the second weighted target indicator data.

[0136] Therefore, the first weighted target indicator data is expressed as follows:

[0137]

[0138] The second weighted target indicator data is expressed as follows:

[0139]

[0140] After obtaining the first weighted target indicator data and the second weighted target indicator data, the above calculation formulas for the data impact factor and the structure impact factor can be obtained.

[0141] The following is a specific example to illustrate the data fluctuation analysis method in the present disclosure:

[0142] Assume that the acquired data to be analyzed is as shown in Table 1. In Table 1, column A represents the first click data set, column B represents the second click data set, column C represents the first display data set, and column D represents the second display data set.

[0143] Table 1 Data to be analyzed

[0144]

[0145]

[0146] After obtaining the data to be analyzed, determine the first target indicator data and the second target indicator data:

[0147]

[0148]

[0149] Among them, cpm t The first target indicator data, cpm y This is the second target indicator data.

[0150] The process of determining the first weighted target indicator data and the second weighted target indicator data is shown in Table 2.

[0151] Table 2 Determining weighted target indicator data

[0152]

[0153]

[0154]

[0155] In Table 2, column H: is the data impact factor weight a The sum of the H column is the weight of the data impact factor. a The denominator of the E column is the weight of the data impact factor. a Column F is cpm t,a = Column A / Column C, which is the first classification indicator data under each category a, where columns A and C are the columns in Table 1; column G is cpm y,a = Column B / Column D, which is the second classification indicator data under each category a, where columns B and D are the columns in Table 1; the first weighted target indicator data MHcpm t =∑ a (cpm t,a *weight a ) = sum (column E * column F) = 19.89792; the second weighted target indicator data MHcpm y =∑ a (cpm t,a *weight a )=sum(column E*column G)=20.26437.

[0156] Determine the data volatility as:

[0157]

[0158] The data impact factors are:

[0159]

[0160] The structural impact factors are:

[0161]

[0162] in,

[0163] At this point, the data volatility is split into data influencing factors and structural influencing factors, that is, it is split into two parts: the influence of data factors themselves and the influence of structural factors.

[0164] By splitting the data volatility into data impact factors and structural impact factors, it can be determined that the main reason why the data volatility of the first target indicator data relative to the second target indicator data is negative -2.9661% is caused by the decline of the target indicator data itself, because the data impact factor CPM part The absolute value of the structural influence factor W part The absolute value of the data impact factor CPM part The sign is the same as that of the data volatility, which is negative, indicating that the data impact factor and the data volatility change in the same direction.

[0165] Compared with the calculation of traditional target indicator data (including the first target indicator data and the second target indicator data), the calculation of weighted target indicator data (including the first target indicator data and the second target indicator data) in the embodiment of the present disclosure can eliminate the impact of changes in the aggregation dimension (i.e., structure); the weighted target indicator data is used to split the data volatility into two parts: data impact factor and structural impact factor, thereby achieving effective attribution of data volatility; the data analysis in the embodiment of the present disclosure can be applied to a variety of scenarios, and the data impact factor weights of aggregation indicators of different dimensions are calculated and split according to different target indicators and dimensions of concern, which is flexible and universal; through the split data impact factor and structural impact factor, it can be effectively explained whether the data volatility is caused by structural changes or changes in the data itself, thereby achieving quantification of data fluctuation attribution.

[0166] Figure 2 FIG. 1 is a block diagram of a data analysis device according to an exemplary embodiment. Figure 2 The device includes a data acquisition module 21, a target indicator data determination module 22, a weighted indicator data determination module 23, a data volatility determination module 24 and an impact factor splitting module 25.

[0167] The data acquisition module 21 is configured to acquire data to be analyzed, wherein the data to be analyzed includes a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period;

[0168] The target indicator data determination module 22 is configured to determine first target indicator data based on the first click data set and the first display data set, and determine second target indicator data based on the second click data set and the second display data set;

[0169] The weighted index data determination module 23 is configured to determine first weighted target index data and second weighted target index data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set;

[0170] The data volatility determination module 24 is configured to determine the data volatility based on the first target indicator data and the second target indicator data;

[0171] The impact factor splitting module 25 is configured to determine the data impact factor and structural impact factor of the fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data and the data volatility.

[0172] Optionally, the impact factor splitting module includes:

[0173] a difference determining unit configured to determine a difference between the first weighted target indicator data and the second weighted target indicator data as a first difference, and to determine a difference between the first target indicator data and the second target indicator data as a second difference;

[0174] a self-weight determination unit, configured to determine a data impact weight of the data volatility based on the first difference and the second difference;

[0175] The impact factor splitting unit is configured to determine the data impact factor and the structure impact factor according to the data volatility and the data impact weight.

[0176] Optionally, the impact factor splitting unit is configured to execute:

[0177] Determine the product of the data volatility and the data impact weight as the data impact factor;

[0178] The difference between the data volatility and the data impact factor is determined as the structural impact factor, or the difference between 1 and the data impact weight is determined as the structural impact weight, and the product of the data volatility and the data impact weight is determined as the structural impact factor.

[0179] Optionally, the weighted index data determination module includes:

[0180] a factor weight set determining unit configured to determine a data influencing factor weight set of a fluctuation of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set;

[0181] The weighted indicator data determination unit is configured to determine the first weighted target indicator data based on the first click data set, the first display data set and the data influence factor weight set, and to determine the second weighted target indicator data based on the second click data set, the second display data set and the data influence factor weight set.

[0182] Optionally, the first click data set includes first click data under multiple categories, the first display data set includes first display data under the multiple categories, the second click data set includes second click data under the multiple categories, the second display data set includes second display data under the multiple categories, and the data influence factor weight set includes data influence factor weights under the multiple categories.

[0183] Optionally, the weighted index data determining unit is configured to execute:

[0184] Determine target indicator data for each category in a first statistical period based on the first click data and the first display data under the multiple categories, and determine the target indicator data for each category in the first statistical period as first category indicator data;

[0185] Determining target index data for each category in a second statistical period based on the second click data and the second display data under the multiple categories, and determining the target index data for each category in the second statistical period as second category index data;

[0186] The first classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the first weighted target indicator data, and the second classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the second weighted target indicator data.

[0187] Optionally, the factor weight set determination unit includes:

[0188] a data factor influence data determination subunit, configured to determine the data factor influence data under each category based on the first display data and the second display data under the multiple categories;

[0189] The total data impact data determining subunit is configured to determine the sum of the data factor impact data under all categories as the total data factor impact data;

[0190] The factor weight set determination subunit is configured to determine the ratio of the data factor influence data under each category to the total data factor influence data as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

[0191] Optionally, the self-influence data determination subunit is configured to execute:

[0192] The ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category is determined as the data factor influence data under the category.

[0193] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0194] Figure 3 3 is a block diagram of an electronic device according to an exemplary embodiment. For example, the electronic device 300 can be provided as a server. Figure 3 The electronic device 300 includes a processing component 322, which further includes one or more processors, and a memory resource represented by a memory 332 for storing instructions, such as applications, that can be executed by the processing component 322. The application stored in the memory 332 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 322 is configured to execute the instructions to perform the above-mentioned data analysis method.

[0195] The electronic device 300 may further include a power supply component 326 configured to perform power management of the electronic device 300, a wired or wireless network interface 350 configured to connect the electronic device 300 to a network, and an input / output (I / O) interface 358. The electronic device 300 may operate based on an operating system stored in the memory 332, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or the like.

[0196] In an exemplary embodiment, a computer-readable storage medium including instructions is also provided, such as a memory 332 including instructions. The instructions can be executed by the processing component 322 of the electronic device 300 to perform the above-mentioned data fluctuation analysis method. Alternatively, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0197] In an exemplary embodiment, a computer program product is further provided, comprising a computer program or computer instructions, wherein the computer program or computer instructions implement the above-mentioned data analysis method when executed by a processor.

[0198] Other embodiments of the present disclosure will readily occur to those skilled in the art after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, with the true scope and spirit of the present disclosure being indicated by the following claims.

[0199] It should be understood that the present invention is not limited to the exact construction described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A data analysis method, characterized in that: include: Acquire data to be analyzed, the data to be analyzed comprising a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period; Determining first target indicator data based on the first click data set and the first display data set, and determining second target indicator data based on the second click data set and the second display data set; Determining first weighted target indicator data and second weighted target indicator data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set; determining a data volatility rate according to the first target indicator data and the second target indicator data; Determining a data impact factor and a structure impact factor of a fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility; Determining a data impact factor and a structure impact factor of the fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility includes: Determine the difference between the first weighted target indicator data and the second weighted target indicator data as a first difference, and determine the difference between the first target indicator data and the second target indicator data as a second difference; determining a data impact weight of the data volatility according to the first difference and the second difference; The data impact factor and the structure impact factor are determined according to the data volatility and the data impact weight, and the sum of the data impact factor and the structure impact factor is equal to the data volatility.

2. The method according to claim 1, characterized in that Determining the data impact factor and the structure impact factor according to the data volatility and the data impact weight includes: Determine the product of the data volatility and the data impact weight as the data impact factor; The difference between the data volatility and the data impact factor is determined as the structural impact factor, or the difference between 1 and the data impact weight is determined as the structural impact weight, and the product of the data volatility and the structural impact weight is determined as the structural impact factor.

3. The method according to claim 1 or 2, characterized in that Determining first weighted target indicator data and second weighted target indicator data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set includes: Determining a data influencing factor weight set of fluctuations of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set; The first weighted target indicator data is determined based on the first click data set, the first display data set and the data influence factor weight set, and the second weighted target indicator data is determined based on the second click data set, the second display data set and the data influence factor weight set.

4. The method according to claim 3, characterized in that The first click data set includes the first click data under multiple categories, the first display data set includes the first display data under the multiple categories, the second click data set includes the second click data under the multiple categories, the second display data set includes the second display data under the multiple categories, and the data influence factor weight set includes the data influence factor weights under the multiple categories.

5. The method according to claim 4, characterized in that Determining first weighted target indicator data based on the first click data set, the first display data set, and the data influence factor weight set, and determining second weighted target indicator data based on the second click data set, the second display data set, and the data influence factor weight set, including: Determine target indicator data for each category in a first statistical period based on the first click data and the first display data under the multiple categories, and determine the target indicator data for each category in the first statistical period as first category indicator data; Determining target index data for each category in a second statistical period based on the second click data and the second display data under the multiple categories, and determining the target index data for each category in the second statistical period as second category index data; The first classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the first weighted target indicator data, and the second classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the second weighted target indicator data.

6. The method according to claim 4, characterized in that Determining, based on the first display data set and the second display data set, a set of data influencing factor weights of a fluctuation of the first target indicator data relative to the second target indicator data, including: Determining data factor influencing data under each category based on the first display data and the second display data under the multiple categories; The sum of the data factors affecting all categories is determined as the total data factors affecting data; The ratio of the data factor influence data under each category to the total data factor influence data is determined as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

7. The method according to claim 6, characterized in that Determining data factor influencing data for each category based on the first display data and the second display data for the multiple categories includes: The ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category is determined as the data factor influence data under the category.

8. A data analysis device, characterized in that: include: a data acquisition module configured to acquire data to be analyzed, wherein the data to be analyzed includes a first click data set and a first display data set within a first statistical period, and a second click data set and a second display data set within a second statistical period; a target indicator data determination module, configured to determine first target indicator data based on the first click data set and the first display data set, and determine second target indicator data based on the second click data set and the second display data set; a weighted index data determination module, configured to determine first weighted target index data and second weighted target index data associated with the data based on the first display data set, the second display data set, the first click data set, and the second click data set; a data volatility determination module, configured to determine the data volatility based on the first target indicator data and the second target indicator data; an impact factor splitting module, configured to determine a data impact factor and a structural impact factor of the fluctuation of the first target indicator data relative to the second target indicator data based on the first target indicator data, the second target indicator data, the first weighted target indicator data, the second weighted target indicator data, and the data volatility; The impact factor splitting module includes: a difference determining unit configured to determine a difference between the first weighted target indicator data and the second weighted target indicator data as a first difference, and to determine a difference between the first target indicator data and the second target indicator data as a second difference; a self-weight determination unit, configured to determine a data impact weight of the data volatility based on the first difference and the second difference; The impact factor splitting unit is configured to determine the data impact factor and the structure impact factor according to the data volatility and the data impact weight, wherein the sum of the data impact factor and the structure impact factor is equal to the data volatility.

9. The device according to claim 8, characterized in that The impact factor splitting unit is configured to perform: Determine the product of the data volatility and the data impact weight as the data impact factor; The difference between the data volatility and the data impact factor is determined as the structural impact factor, or the difference between 1 and the data impact weight is determined as the structural impact weight, and the product of the data volatility and the structural impact weight is determined as the structural impact factor.

10. The device according to claim 8 or 9, characterized in that The weighted index data determination module includes: a factor weight set determining unit configured to determine a data influencing factor weight set of a fluctuation of the first target indicator data relative to the second target indicator data based on the first display data set and the second display data set; The weighted indicator data determination unit is configured to determine the first weighted target indicator data based on the first click data set, the first display data set and the data influence factor weight set, and to determine the second weighted target indicator data based on the second click data set, the second display data set and the data influence factor weight set.

11. The device according to claim 10, characterized in that The first click data set includes the first click data under multiple categories, the first display data set includes the first display data under the multiple categories, the second click data set includes the second click data under the multiple categories, the second display data set includes the second display data under the multiple categories, and the data influence factor weight set includes the data influence factor weights under the multiple categories.

12. The device according to claim 11, characterized in that The weighted index data determining unit is configured to execute: Determine target indicator data for each category in a first statistical period based on the first click data and the first display data under the multiple categories, and determine the target indicator data for each category in the first statistical period as first category indicator data; Determining target index data for each category in a second statistical period based on the second click data and the second display data under the multiple categories, and determining the target index data for each category in the second statistical period as second category index data; The first classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the first weighted target indicator data, and the second classification indicator data under the multiple categories and the data influence factor weights are weighted and summed to obtain the second weighted target indicator data.

13. The device according to claim 11, characterized in that The factor weight set determination unit includes: a data factor influence data determination subunit, configured to determine the data factor influence data under each category based on the first display data and the second display data under the multiple categories; The total data impact data determining subunit is configured to determine the sum of the data factor impact data under all categories as the total data factor impact data; The factor weight set determination subunit is configured to determine the ratio of the data factor influence data under each category to the total data factor influence data as the data influence factor weight under the category, and the data influence factor weights under all categories constitute the data influence factor weight set.

14. The device according to claim 13, characterized in that The data factor affecting data determining subunit is configured to execute: The ratio of the product of the first display data and the second display data under each category to the sum of the first display data and the second display data under the category is determined as the data factor influence data under the category.

15. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the data analysis method according to any one of claims 1 to 7. 16 . A computer-readable storage medium, wherein when instructions in the computer storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the data analysis method according to claim 1 .

17. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the data analysis method according to any one of claims 1 to 7 is implemented.

Images