Data analysis method and device, computer equipment and storage medium

Abstract

The invention discloses a data analysis method and device, computer equipment and a storage medium. The data analysis method comprises the following steps: firstly, dividing business data into first data under different dimensions; obtaining a first explanation value of the first data according to corresponding service success rates and service data proportions of different service seats under the first data; the first explanation value represents the association degree of the first data and the business data of the business seat; obtaining a first amazing degree value of the first data according to the difference between the business data proportions corresponding to different business seats; determining a second interpretation value corresponding to each dimension based on the first interpretation value of the first data, and determining a second amazing degree value corresponding to each dimension based on the first amazing degree value of the first data; and finally determining at least one service optimization strategy corresponding to the service data according to the first interpretation value, the first amazing degree value, the second interpretation value and the second amazing degree value. According to the invention, the efficiency and accuracy of business data analysis can be improved.

Description

Technical Field

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

[0002] Today, traditional business data analysis usually relies on manual simple statistics and analysis of basic indicators. For example, the performance of business agents is evaluated by manually counting indicators such as the number of calls, call duration, and customer conversion rate.

[0003] However, manual analysis of large-scale, multi-dimensional data is inefficient, and it is difficult to accurately assess the contribution of data in each dimension to the business, resulting in a lack of scientific basis for the formulation of business strategies and an inability to identify hidden business opportunities or potential risks in a timely manner. Summary of the invention

[0004] In response to the above technical problems, the embodiments of the present application provide a data analysis method, apparatus, computer device and storage medium.

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

[0006] Divide the business data into first data under different dimensions;

[0007] According to the business success rates and business data proportions corresponding to different business seats under the first data, a first interpretation value of the first data is obtained; the first interpretation value represents the degree of association between the first data and the business data of the business seat;

[0008] obtaining a first surprise value of the first data according to a difference between business data proportions corresponding to different business seats under the first data;

[0009] Determine a second interpretation value corresponding to each dimension based on the first interpretation value of the first data, and determine a second surprise value corresponding to each dimension based on the first surprise value of the first data;

[0010] At least one business optimization strategy corresponding to the business data is determined according to the first explanation value, the first surprise value, the second explanation value, and the second surprise value.

[0011] In a second aspect, an embodiment of the present application provides a data analysis device, the device comprising:

[0012] A data division module, used to divide the business data into first data under different dimensions;

[0013] A first analysis module, configured to obtain a first interpretation value of the first data according to the business success rates and business data proportions corresponding to different business agents under the first data; the first interpretation value represents the degree of association between the first data and the business data of the business agent;

[0014] A second analysis module is used to obtain a first surprise value of the first data according to a difference between business data proportions corresponding to different business seats under the first data;

[0015] a third analysis module, configured to determine a second interpretation value corresponding to each dimension based on the first interpretation value of the first data, and to determine a second surprise value corresponding to each dimension based on the first surprise value of the first data;

[0016] A strategy formulation module is used to determine at least one business optimization strategy corresponding to the business data according to the first interpretation value, the first surprise value, the second interpretation value and the second surprise value.

[0017] In a third aspect, an embodiment of the present application further provides a computer device, comprising a memory storing a plurality of instructions; a processor loads instructions from the memory to execute the steps of any one of the data analysis methods provided in the embodiments of the present application.

[0018] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a plurality of instructions suitable for loading by a processor to execute the steps of any data analysis method provided in the embodiment of the present application.

[0019] In a fifth aspect, an embodiment of the present application further provides a computer program product, including a computer program or instructions, which, when executed by a processor, implements the steps of any data analysis method provided in the embodiment of the present application.

[0020] By adopting the scheme of the embodiment of the present application, the business data is first divided into first data under different dimensions, and each dimension has at least two first data. Compared with business analysis of data of a single dimension, it has a more comprehensive data analysis basis, which is helpful to formulate more accurate and targeted business optimization strategies in the future. Then, according to the business success rate and business data proportion corresponding to different business seats under the first data, the first interpretation value of the first data is obtained. The first interpretation value represents the degree of association between the first data and the business data of the business seat, so that the performance differences generated by the first data of multiple dimensions can be analyzed based on the first interpretation value for interpretation; then, according to the difference between the business data proportions corresponding to different business seats under the first data, the first surprise value of the first data is obtained, so that the first data that has a significant impact on business performance can be identified without relying on manual judgment. Then, based on the first interpretation value and the first surprise value of the first data, the second interpretation value and the second surprise value corresponding to each dimension are obtained, so that the impact of the overall interpretation value and surprise value of the dimension on the business can be considered. Finally, based on the first explanation value, the first surprise value, the second explanation value and the second surprise value, at least one business optimization strategy corresponding to the business data is obtained, so that key dimensions that are highly explanatory and have a positive impact on business improvement, as well as key first data that are highly explanatory and have a positive impact on the business under the key dimensions can be selected, and the corresponding business optimization strategies can be efficiently and accurately generated based on the selected key dimensions and key first data. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 It is a schematic diagram of the application environment of the data analysis method provided in the embodiment of the present application;

[0023] Figure 2 It is a flow chart of the data analysis method provided in the embodiments of the present application;

[0024] Figure 3 It is a schematic diagram of dimensional division of business data provided in an embodiment of the present application;

[0025] Figure 4 It is a schematic diagram of dividing business data of each dimension provided in an embodiment of the present application;

[0026] Figure 5 is a schematic diagram of selecting first data of a first set provided in an embodiment of the present application;

[0027] Figure 6 It is a schematic diagram of the second surprise value and the second explanation value of the overall statistical dimension provided in the embodiments of the present application;

[0028] Figure 7 is a schematic diagram of selecting first data of a second set provided in an embodiment of the present application;

[0029] Figure 8 is a schematic diagram of selecting target data provided in an embodiment of the present application;

[0030] Fig. 9 is a schematic diagram of the structure of a resource allocation device provided in an embodiment of the present application;

[0031] Fig.10 It is a schematic diagram of the internal structure of the computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application. At the same time, in the description of the embodiments of the present application, the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0033] In one embodiment of the present application, the data analysis method can be run on a local terminal device or a server. When the data analysis method is run on a server, the method can be implemented and executed based on a data analysis system, wherein the data analysis system includes a server and a client device.

[0034] In order to better understand a data analysis method, device, electronic device and storage medium provided in an embodiment of the present application, an application environment applicable to the embodiment of the present application is first described below.

[0035] See also Figure 1 , Figure 1 FIG. 1 is a schematic diagram showing an application environment of a data analysis method provided in an embodiment of the present application. As an implementation mode, the data analysis method provided in an embodiment of the present application can be applied to an electronic device. The electronic device can be, for example, Figure 1The server 110 shown in the figure can communicate with the terminal device 120 through a network. The network is a medium for providing a communication link between the server 110 and the terminal device 120. The network can include various connection types, such as wired communication links, wireless communication links, etc., which are not limited in the embodiments of the present application. Optionally, in other embodiments, the electronic device can also be a smart phone, a laptop computer, etc.

[0036] It should be understood that Figure 1 The server 110, network and terminal device 120 are merely illustrative, and any number of servers, networks and terminal devices may be provided as required. For example, the server 110 may be a physical server or a server cluster composed of multiple servers, and the terminal device 120 may be a mobile phone, a tablet, a desktop computer, a laptop computer, and the like. It is understood that the embodiments of the present application may also allow multiple terminal devices 120 to access the server 110 at the same time.

[0037] In some embodiments, the terminal device 120 can obtain the first data of each dimension, as well as the first surprise value and the first interpretation value of each first data. Further, the terminal device 120 can send the first data of each dimension, as well as the first surprise value and the first interpretation value of each first data to the server 110 through the network. After receiving the above data, the server 110 can calculate the second surprise value and the second interpretation value of each dimension as a whole, and analyze the above business data based on the above first surprise value, the first interpretation value, the second surprise value and the second interpretation value, determine the target data, and determine the business optimization strategy for the business based on the target data, so that the efficiency and accuracy of the business data analysis can be improved through the data analysis method described in the embodiment of the present application.

[0038] The following is a detailed description in conjunction with the accompanying drawings. In this embodiment, the execution subject is a terminal device as an example. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments. Although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from that shown in the accompanying drawings.

[0039] The data analysis method of this embodiment can be applied to any scenario that needs to be executed, and the method can be applied to products in these scenarios, such as data analysis systems, business analysis systems, etc.

[0040] In many business scenarios, such as sales scenarios, sales teams usually rely on manual statistics and analysis to evaluate the performance of sales agents and formulate corresponding sales strategies. This traditional method is usually based on single or fewer dimensions of business data for analysis. For example, the performance of business agents is judged by only one or two basic sales indicators such as the number of calls, call duration, and customer conversion rate. Since the dimensions of the analysis data are too narrow, it is impossible to fully reveal the complex relationship between different data in the sales process, resulting in a lack of sufficient relevance and effectiveness of sales strategies.

[0041] It is easy to overlook the complexity of the sales process when only relying on data from a single or a few dimensions for business analysis. It is understandable that sales performance is often affected by multiple dimensions of data and multiple factors, and there are complex interactions between these data and factors. For example, a customer's purchase of goods depends not only on the communication skills of the sales agent, but also on multiple dimensions such as customer background information, market environment, and product features. Manual analysis cannot fully capture the interactive relationship between these first-level data, resulting in sales strategies that often lack high accuracy and pertinence, and cannot effectively improve sales performance.

[0042] In addition, the traditional manual data analysis method usually relies on the analyst's own experience and expertise. The analysis efficiency is extremely low when dealing with large-scale, multi-dimensional business data, and it is easy to ignore the potential opportunities to improve business performance. For example, the business team needs to record a large amount of customer feedback, call records, agent behavior and other data every day. Manual analysis of such massive data is not only inefficient, but also difficult to accurately find the real reasons for the large differences in business performance between different departments, which will affect the formulation and adjustment of business strategies. For example, when there are subtle changes in the behavior of customer groups, the sales team cannot detect these changes in time, and thus miss the opportunity to optimize sales strategies.

[0043] Based on this, in order to solve the problems of low efficiency and poor accuracy in existing business data analysis, the data analysis method, device, computer equipment and storage medium of the present application are proposed.

[0044] Please refer to Figure 2 The specific process of the data analysis method provided in the embodiment of the present application may include the following steps 201 to 205, which are specifically introduced below.

[0045] Step 201: Divide the business data into first data under different dimensions.

[0046] In the scenario of this application, business data can be information generated during the business sales process, such as the behavior of business agents and customers, customer characteristics, call content, market environment, etc. Business data is usually difficult to analyze using a single indicator, so it is necessary to divide it into multiple dimensions according to the different attributes of the business data for processing. Each dimension can be regarded as an independent factor that affects the business process and results.

[0047] In some embodiments, taking the business data generated in the process of telephone sales of financial products as an example, dimensions related to business seats may include call duration, call frequency, communication methods, etc., and dimensions related to customers may include age, credit limit, occupation, region, and consumption history, etc. Figure 3 As shown, the business data can be divided into P dimensions, each dimension having corresponding first data, for example, the first dimension is the customer's age, the second dimension is the customer's credit limit, and so on.

[0048] It should be noted that the basis for dimensional division of business data is not limited to the two types of business seats and customers. For example, business data can also be dimensionally divided from aspects such as business content and market environment. This embodiment does not limit the specific type of dimension.

[0049] In some embodiments, each dimension may have at least two first data, so as to provide comprehensive and detailed data for in-depth analysis. Figure 4 As shown, assuming that after the business data is divided into dimensions, the first dimension is the age dimension, and the age dimension can include the first data 1 to the first data q. For example, the first data 1 can be customers aged "18-25", the first data 2 can be customers aged "26-35", and the first data q can be customers aged "75-80". This ensures that each divided dimension has a sufficient number of first data, so that the business data can be comprehensively and accurately analyzed based on the sufficient number of first data.

[0050] Through the above method, the embodiment of the present application can perform structured processing on the business data by dimensional division, and then the divided first data can be combined for comprehensive analysis, so that the different impacts of each dimension and its first data on the business can be accurately identified, and the source of performance differences can be fully reflected, which is helpful to identify business data that plays a key role in business improvement.

[0051] Step 202: Obtain a first interpretation value of the first data according to the business success rates and business data proportions corresponding to different business seats under the first data.

[0052] Among them, the business success rate refers to the success rate of different business seats in completing a certain type of business, which can be measured by indicators such as customer conversion rate and business transaction rate. For example, the business success rate of an excellent seat for a certain type of customer is 50%, while the business success rate of an ordinary seat for this type of customer can be 30%.

[0053] Among them, the business data proportion refers to the proportion of customer groups of business seats under a certain first data, which can reflect the correlation between business seats and a certain type of business data. For example, under the dimension of customer type, the customer proportion of different business seats may be different. Some business seats may interact more with high-value customers. For example, high-value customers served by excellent seats account for 50% of all customers they serve, while high-value customers served by ordinary seats may only account for 10% of all customers they serve.

[0054] It is understandable that the performance difference of each business seat is closely related to different first data. For example, different business seats have different success rates when handling different types of customers, or some business seats perform well in certain types of call content.

[0055] The first interpretation value is calculated by comprehensively analyzing the business success rate and business data proportion corresponding to the first data, and is used to characterize the degree of correlation between the first data and the business data of the business agent.

[0056] It can be understood that if the business success rate of a certain business agent under the first data of a specific dimension is significantly higher than that of other agents, it means that the first data has a great positive impact on the business performance of the business agent, and the difference in business success rate will be included in the calculation of the first explanation value.

[0057] Correspondingly, if the business data ratio of a certain business agent to a specific first data is significantly higher, that is, the agent has a higher degree of participation in such business, which also indicates that the business performance of the agent is strongly correlated with the first data. Therefore, the higher the ratio of a first data to the business data of a business agent, the greater the impact of the first data on the business performance of the business agent.

[0058] Combining the above two factors, the first explanation value can reflect the impact of the first data on the agent's business performance. By analyzing the explanation value of business data, we can reveal the advantages and disadvantages of different agents in handling business in different business dimensions.

[0059] In a specific implementation, the first interpretation value represents the degree of association between the first data and the business agent, that is, it quantifies the impact of a specific first data (such as "the user's age is 25 years old" or "the call duration is 5 minutes") on the business performance of different business agents (such as excellent business agents and ordinary business agents). For example, when the first interpretation value of business agent A corresponding to the first data "the user's age is 25 years old" is relatively high, it means that the business success rate of this type of customer is highly correlated with that of business agent A, and business agent A has a significant effect on the service of this type of customer. For another example, if the first interpretation value of an excellent agent under a certain first data is significantly higher than that of an ordinary agent, it means that the first data is the key first data for the excellent agent's business performance to surpass that of an ordinary agent, and subsequent business strategies can be developed around the key first data with a higher first interpretation value. Similarly, first data with a lower first interpretation value represent that the first data has a weaker correlation with business performance, and the business team can adjust resource allocation or take improvement measures based on the characteristics of these first data.

[0060] In some embodiments, step 202 may include the following steps:

[0061] First, a first service success rate corresponding to the first service seat under the first data and a second service success rate corresponding to the second service seat are obtained;

[0062] Next, the first business data ratio corresponding to the first business seat under the first data and the second business data ratio corresponding to the second business seat are obtained;

[0063] Finally, based on the first difference between the first business success rate and the second business success rate, the second difference between the first business data proportion and the second business data proportion, the first average value of the first business success rate and the second business success rate, and the second average value of the first business data proportion and the second business data proportion, the first interpretation value of the first data is obtained.

[0064] Specifically, a first product value of the first difference and the second average value, and a second product value of the second difference and the first average value may be calculated, and then the interpretation value of the first data may be determined according to the sum of the first product value and the second product value.

[0065] Among them, the business performance of the first business seat may be better than that of the second business seat, so the first business seat may be called an excellent business seat, and the second business seat may be called an ordinary business seat.

[0066] In some embodiments, the first interpretation value may be calculated by the following formula:

[0067]

[0068] Among them, EP ijis the first interpretation value of the jth first data in the i-th dimension, Pop′ ij The first business data percentage of the first business seat, Pop ij is the proportion of the second business data of the second business seat, R′ ij is the first business success rate corresponding to the first business agent, R ij It is the second business success rate corresponding to the second business agent.

[0069] The following describes the specific calculation process of the first interpretation value. Assuming that the first data is "customers aged 26-35 years old", the first business success rate R of the first business agent i ′ j 30%, the first business data accounts for Pop' ij The second business success rate of the second business agent is 27%. ij 20%, the second business data accounts for Pop ij is 15%.

[0070] First, a first difference R′ representing the difference in the service success rate of two service agents under the first data can be calculated. ij -R ij =0.3-0.2=0.1, and then calculate the second difference Pop′ of the difference in the proportion of business data of the two business seats ij -Pop ij =0.27-0.15=0.12. Then, the first average value representing the average value of the service success rate of the two service seats can be calculated: Then calculate the second average value representing the average proportion of business data of the two business seats

[0071] Finally, the first interpretation value of customers aged 26-35 can be calculated as EP 26-25 =(0.27-0.15)×(0.3+0.2) / 2+(0.3-0.2)×(0.27+0.15) / 2=0.102.

[0072] By calculating the first interpretation value of the first data through the above process, it is possible to capture the different impacts of differences in business success rates and differences in business data proportions on the performance of different business agents under the first data. If the calculated first interpretation value is high, it means that the first data has made a significant contribution to the performance of excellent agents, and the business strategies subsequently formulated can be optimized based on the first data. The present application can quantify the degree of correlation between the first data and the performance differences between different business agents through the first interpretation value, thereby helping to identify the key factors affecting sales performance to optimize business strategies.

[0073] Step 203: Obtain a first surprise value of the first data according to the difference between the business data proportions corresponding to different business seats under the first data.

[0074] Among them, the first surprise value is used to measure the difference between the proportions of business data of different business seats under a certain first data. By quantifying the distribution difference between two business seats (such as an excellent seat and an ordinary seat) on the first data, it reflects whether the first data can significantly affect the performance differences between business seats.

[0075] Specifically, taking the difference in the proportion of business data between the first business agent and the second business agent on a certain first data as an example, the difference can indicate whether there is a significant difference in the interaction or investment of the two business agents on this dimension. If the difference in the proportion of business data is large, it means that the coverage or investment of the above two business agents under the first data is significantly different, indicating that the first data has a greater impact on the performance of the business agents. Correspondingly, if the difference in the proportion of business data is small, it means that the first data has similar coverage of the two business agents and fails to cause a significant difference between the two agents.

[0076] After understanding the impact of the difference in the proportion of business data on different business seats and businesses, it can be understood that if the calculated first surprise value of the first data is high, it means that the proportion of business data of the first data is significantly different between the two business seats, and the first data has a greater impact on the business performance of the two business seats, and is the key factor that causes the difference in performance between excellent seats and ordinary seats, which can prompt subsequent business strategies to pay special attention to the first data. Correspondingly, if the calculated first surprise value of the first data is low, it means that the proportion of business data of the first data is relatively uniform between the two business seats, and is not the main reason for the performance difference between the two business seats, which means that the reference value of the first data for subsequent business strategy adjustments is low.

[0077] For example, suppose that in a telephone sales business scenario, there are two first data under the dimension of "customer type", namely high-value customers and ordinary customers. Now we hope to analyze the performance differences between the first business agent and the second business agent under different customer types by calculating the first surprise value. Assuming that high-value customers account for 60% of the customers served by the first business agent and 30% of the customers served by the second business agent, it can be calculated that the difference in the proportion of business data between the two is 30%. The large difference in the proportion of business data indicates that excellent agents have invested more resources in high-value customer types or obtained more customers, while ordinary agents have relatively less performance in this type. This difference may lead to a high surprise value, indicating that this customer type has a significant impact on the performance of the two agents, and the excellent agent has gained an advantage through this type of customers.

[0078] As can be seen from the above, the first surprise value of the first data of this application is calculated by analyzing the difference in business share of different business seats on a certain first data. The surprise value measures whether the performance of two seats on the first data is significantly different, reflecting the degree of influence of the first data on the performance difference between the two seats. A high surprise value indicates that the first data has made an important contribution to the performance difference and may become a key factor in optimizing sales strategies.

[0079] In some embodiments, the first surprise value may be determined by:

[0080] First, determining a first difference between the first business data proportion and the second average value, and a second difference between the second business data proportion and the second average value;

[0081] Next, determining a first relative entropy between the first business data proportion and the second average value;

[0082] Next, determining a second relative entropy between the second business data proportion and the second average value;

[0083] Finally, the surprise value of the first data is determined according to the product of the first relative entropy and the second relative entropy.

[0084] In some embodiments, the first surprise value may be calculated using the following formula:

[0085]

[0086] Among them, S ij represents the first surprise value of the jth first data in the i-th dimension, m is the second average value, Pop′ ij The first business data percentage of the first business seat, Pop ij is the proportion of the second business data of the second business seat, D KL represents relative entropy, D KL (Pop′ ij ‖m) is the first relative entropy, D KL (Pop ij ‖m) is the second relative entropy, and x and y are the two elements for relative entropy calculation.

[0087] In some embodiments, the business data proportion Pop′ of a first business agent (eg, an excellent agent) on a first data may be obtained first. ij , and compared with the second mean value m of this dimension, m can be specifically expressed as: A first difference between the first business data proportion and the average value is obtained, where the first difference indicates a degree of deviation between the data proportion of the first business agent in the dimension and the average level.

[0088] In some embodiments, the business data proportion Pop of the second business seat (eg, ordinary seat) on the same first data can be obtained. ij The second difference between the second business data proportion and the average value is obtained by comparing the second difference with the second average value m. The second difference is calculated in a similar manner to the first difference, and is used to measure the deviation of ordinary seats from the average distribution.

[0089] In some embodiments, the first difference can be reflected by a first relative entropy, which is KL divergence and is used to measure the difference between one distribution and another distribution. The first relative entropy is the proportion of business data of the first business seat Pop′ ij The difference between the second mean value m and Pop′ ij Substituting η and m into the calculation formula of the first relative entropy, it is specifically expressed as:

[0090]

[0091] It can be understood that the first relative entropy is used to measure the difference between the business proportion of the first business agent under a certain first data and the average distribution of the business proportion. The larger the value of the first relative entropy, the greater the deviation of the distribution of the first business agent on the first data from the average value.

[0092] Correspondingly, the second relative entropy can be calculated, that is, the business data proportion Pop of the second business seat ij The difference between the average distribution m. Its calculation formula is the same as the first relative entropy:

[0093]

[0094] It can be understood that the second relative entropy is used to measure the difference between the business proportion of ordinary seats on a certain first data and the average distribution of business proportions. The larger the value of the second relative entropy, the greater the deviation of the distribution of the second business seat on the first data from the average value.

[0095] In some embodiments, the product of the first relative entropy and the second relative entropy may be averaged to calculate the first surprise value of the first data, characterizing the distribution difference of the proportion of business data of different business agents on the first data. The larger the first surprise value, the more significant the difference in business performance of the first data between the two business agents.

[0096] It can be understood that relative entropy is used to measure the information loss or difference between two distributions. For the proportion of two business seats in a certain first data, the larger the relative entropy, the greater the difference between the distribution of the seat in this dimension and the average distribution. The present application reflects whether there is a significant difference in the distribution of the two business seats in the first data through the first surprise value of the first data. If the business proportions of the two business seats are significantly deviated from the average, the first surprise value is large, indicating that the first data plays a key role in the performance difference between the two business seats.

[0097] like Figure 5 As shown, after completing the above calculation process, the first explanation value and the first surprise value corresponding to the first data of each dimension can be summarized, for example Figure 5 The first surprise value corresponding to the first data "26-35 years old" under the dimension "age group" is 0.3, and the first explanation value is 5% or 0.05, so as to facilitate the subsequent determination of the first data that has a key impact on the business according to the first explanation value and the first surprise value of each first data, and optimize the business.

[0098] This application calculates the first surprise value of each first data based on the difference between the business data proportion of different business seats on a certain first data and the average distribution of the business data proportion. By focusing on the first surprise value with a higher numerical value, the first data that has an important impact on the performance difference between two business seats is captured, and the important first data is subsequently applied in business strategy formulation.

[0099] Step 204: Obtain a second explanation value and a second surprise value corresponding to each dimension based on the first explanation value and the first surprise value of the first data.

[0100] In some embodiments, the second interpretation value of each dimension can be obtained by summing up the first interpretation value of the first data under the dimension, and the second surprise value of each dimension can be obtained by summing up the first surprise value of the first data under the dimension.

[0101] It can be understood that the first explanation value is calculated based on the difference between the business success rate and business share of a certain business seat (such as an excellent seat and an ordinary seat) under the first data (such as customer type, call duration, etc.), which represents the degree of correlation between the first data and the performance difference between the two business seats. Correspondingly, the second explanation value is the overall explanation value of the dimension, which is calculated by accumulating the first explanation value of each specific first data under the dimension. Therefore, the explanation values ​​of all first data under a dimension jointly determine the role of this dimension in the performance difference between the two business seats.

[0102] The second explanation value indicates the overall contribution of the entire dimension to the performance difference between the two business agents. By accumulating the first explanation value of each first data, the total influence of the dimension can be obtained. If the second explanation value of a dimension is high, it means that all the first data in the dimension have a significant impact on the performance difference of the business agents. Subsequent business strategies can give priority to optimizing these dimensions with higher second explanation values.

[0103] It can be understood that the first surprise value can be used to measure the difference in business proportion of a specific first data between different business seats, and through relative entropy calculation, it indicates the degree of surprise of the first data to the performance difference of different business seats. Correspondingly, the second surprise value refers to the overall surprise value of a certain dimension, which is calculated by accumulating the first surprise values ​​of all first data under the dimension. The second surprise value is used to reflect the difference in the performance of the entire dimension between two business seats.

[0104] The second surprise value indicates the total surprise degree of the distribution difference of all the first data under the entire dimension to the two business agents. If the second surprise value of a dimension is high, it means that there is a significant deviation in the distribution of multiple first data under this dimension between business agents, indicating that this dimension has a greater potential impact on business performance. Therefore, the first data with high surprise values ​​are the data that need special attention when optimizing business strategies and business performance.

[0105] As can be seen from the above, the first explanation value and the first surprise value are calculated for specific first data (such as a certain type of customer, a certain communication method, etc.), and are used to measure the degree of association and difference of a single data to business differences. The second explanation value and the second surprise value are obtained by accumulating the explanation value and surprise value of these single first data, combining the contribution of all data under one dimension to derive the overall influence and difference of the dimension. Therefore, the second explanation value and the second surprise value provide a comprehensive indicator at the dimensional level, which is used to identify the first data that has played a significant role in the performance differences of different business agents.

[0106] like Figure 6 As shown, taking the dimension "age group" as an example, this dimension may include multiple first data such as "26-35 years old", "36-45 years old", etc., and the first explanation value and the first surprise value of the above first data may be obtained one by one, and then the above multiple first explanation values ​​and multiple first surprise values ​​are accumulated to obtain and calculate the overall second explanation value of the age group dimension as 90% and the second surprise value as 4, thereby reflecting the contribution and influence of the age group dimension on the overall performance differences of different business agents.

[0107] The second explanation value can help identify dimensions that have a significant overall impact on business performance differences. Dimensions with higher second explanation values ​​can prompt the business team to focus on these dimensions for optimization and strategy adjustment. The second surprise value can help discover dimensions that have significant differences in the proportion of business data. Dimensions with higher second surprise values ​​usually indicate that different business agents have significant differences in performance on these dimensions, and there may be potential market opportunities or areas that need improvement.

[0108] The second explanation value and the second surprise value of each dimension of the present application can be calculated by accumulating the first explanation value and the first surprise value of all the first data under the dimension. The second explanation value is used to measure the overall contribution of a certain dimension to the performance difference between two business seats, while the second surprise value reflects the overall surprise degree of the distribution difference of the first data between the two seats. The dimensions with high explanation value and high surprise value and the corresponding first data can become the key data to focus on when optimizing business strategies.

[0109] Step 205: Obtain at least one service optimization strategy corresponding to the service data according to the first explanation value, the first surprise value, the second explanation value, and the second surprise value.

[0110] Among them, business optimization strategy refers to identifying the key factors that affect business success through analysis of business data, and formulating a plan to optimize business operations based on the key factors to improve business performance and customer conversion rate. For example, in a telephone sales scenario, business optimization strategy can usually be formulated based on different first data (such as customer age, call duration, credit limit, etc.) and the impact of these first data on the success rate and proportion of the business.

[0111] In some embodiments, the contribution of each first data to the business performance can be identified by performing multi-dimensional analysis on the business data of different business agents, such as customer characteristics, salesperson behavior, call content, etc. By calculating the explanation value and surprise value for attribution analysis, the system can find out the key operations of business agents who perform well in certain dimensions, and summarize these operations into optimization strategies to improve the performance of the overall sales team.

[0112] In some embodiments, step 205 may include the following steps:

[0113] First, first data whose first interpretation value is greater than a first threshold value is added to the first set;

[0114] Next, the first data of the first set whose second interpretation value is greater than the second threshold value is added to the second set;

[0115] Finally, according to the first surprise value of the first data and the second surprise value of the dimension, target data is selected from the second set, and at least one business optimization strategy is determined based on the target data.

[0116] In some embodiments, firstly, by calculating the first interpretation value of each first data, the first data with greater influence can be screened out. If the first interpretation value of a certain first data is greater than the set first threshold, it means that the first data has made a significant contribution to the performance difference and deserves further attention. The screened out first data can be added to the first set, which can also be called a candidate set. The purpose of constructing the candidate set is to screen out potentially important first data in the first step, and add the first data whose first interpretation value is greater than the first threshold to the first set as a candidate for the target data.

[0117] Specifically, after obtaining the first data of each dimension, each first data can be grouped and displayed by dimension, and in order to facilitate the subsequent rapid screening of first data with high first explanation value and first surprise value and improve data analysis efficiency, each first data can be arranged in descending order according to the size of the first surprise value. After grouping each first data by dimension and arranging in descending order according to the first surprise value, a plurality of first data after arrangement are obtained as follows: Figure 5 shown.

[0118] As an example only, assuming that the first threshold is 0.9%, i.e., 0.09, then Figure 5 The first data "age group 26-35 years old, age group 36-45 years old, call duration 2-5 minutes, call duration 2-5 minutes, credit line 100,000-150,000, credit line 50,000-100,000" that are greater than the first threshold are added to the first set.

[0119] In some embodiments, first data with a second interpretation value greater than a second threshold value can be selected from the first data in the first set. First data with a significant overall impact on business differences can be screened out, retained and added to the second set, which is also called an interpretation set.

[0120] As an example only, assuming that the second threshold is 90%, i.e., 0.9, then Figure 7 The first data shown as "age, call duration, credit limit, number of calls and historical purchase amount" that are greater than the second threshold are added to the second set.

[0121] By analyzing the first surprise value and the second surprise value of the first data in the second set, the present application can further filter out target data with a higher second surprise value from the second set, so as to filter out data that has significant differences in the performance of business agents and has an important impact on the performance difference. After determining the target data, a specific business optimization strategy can be formulated based on the target data. The business optimization strategy can be formulated based on the significant impact of the target data on business performance, and the business process, speech or strategy can be optimized in a targeted manner.

[0122] The goal of the business optimization strategy is to use these target data to improve customer conversion rate and business efficiency. For example, by optimizing the communication strategy of business seats and adjusting resource allocation, we can respond to different customer groups or market demands. For example, if the target data is "customers aged 26-35 years old" and "credit limit greater than 100,000", as well as "2-5 minutes" call duration, the above target data can be combined to obtain the final optimization strategy: for customers aged 26-35 years old and "credit limit greater than 100,000", a more concise and efficient 2-5 minute call duration is adopted, or some feasible optimization measures are added, such as emphasizing the long-term returns and financial security of the product in the sales talk, and providing personalized discounts or value-added services to such customers.

[0123] In some embodiments, the target data may be determined by:

[0124] From the multiple dimensions of the second set, select M dimensions whose second surprise value is greater than the third threshold; from each dimension of the M dimensions, select N first data whose first surprise value is greater than the fourth threshold to obtain target data.

[0125] In some embodiments, these dimensions can be further screened by a second surprise value, and dimensions whose second surprise values ​​are greater than a third threshold are selected, indicating that there are significant distribution differences in the first data in these dimensions among business agents. Therefore, these dimensions play an important role in the formulation of business optimization strategies.

[0126] In some embodiments, M dimensions may be screened out, such as Figure 7 As shown, since each dimension is arranged in descending order according to the second surprise value during the screening process, the first M dimensions can be directly screened out. For example, if M is equal to 3, the three dimensions of age group, call duration and credit limit can be screened out.

[0127] In the selected M dimensions, the specific first data under each dimension can be further analyzed to select the first data whose first surprise value is greater than the fourth threshold. The first surprise value greater than the fourth threshold indicates that the first data has a significant impact on the differences between different business seats. Figure 8As shown, assuming that N is 2, the two first data of "2-5 minutes" and "5-8 minutes" can be filtered out from the dimension of call duration as target data.

[0128] Through the above methods, the selected target data can be more clearly reflected in specific dimensions. The performance differences between excellent business seats and ordinary business seats in terms of customer groups, communication methods, etc. provide a strong basis for formulating targeted business optimization strategies. Based on the target data, precise optimization plans can also be formulated for specific customer groups or specific business scenarios. For example, for high-value customers or customers of a certain age group, operations such as optimizing speech, adjusting resource allocation, or adjusting communication strategies can be performed to improve the overall business effect.

[0129] In some embodiments, the service optimization strategy may be determined by:

[0130] First, extracting second data related to the attributes of the business object and third data related to the business behavior from the target data;

[0131] Next, the business behavior of the third data is matched with respect to the business object of the second data to form at least one business optimization strategy.

[0132] The attributes of business objects refer to static or background information related to customer characteristics or market characteristics, etc., which are used to help the business team understand the basic situation of customers or markets in order to formulate more targeted marketing strategies. The second data represents specific first data related to these business objects. For example, the customer's age group, gender, income level, credit limit, etc. Therefore, the second data can be understood as the first data with a "customer attribute label". By extracting the second data related to the attributes of the business object, these data provide background support for the next step of matching the corresponding business behavior.

[0133] Among them, business behavior refers to the interaction and communication between business agents and customers, such as call duration, communication content, promotion methods and other dynamic operational behaviors. Business behavior directly affects the customer conversion rate. The third data represents the target data related to business behavior. For example, call duration, call frequency, recommended product categories, used words, etc. The third data can be understood as the first data with a "business behavior label", which can guide business agents on how to best communicate with customers, demonstrate and other business behaviors.

[0134] In some embodiments, the business object attributes extracted from the second data, such as customers of a specific age group and credit limit, can be matched with corresponding third data, such as specific call duration and communication methods. By combining the second data with the third data, an accurate business optimization strategy can be formed, thereby significantly improving customer conversion rate and sales efficiency.

[0135] In actual operation, the business agent can lock in the target customer group based on the extracted second data and match the best business behavior indicated by the extracted third data. For example, for customers aged 26-35 with a credit limit of more than 100,000 yuan, a 2-5 minute call duration is used to introduce business products as an optimized business strategy.

[0136] Through the above method, this application first extracts the second data related to the attributes of the business object (such as customer age, credit limit, etc.) from the target data, and then extracts the third data related to the business behavior (such as call duration, communication skills, etc.). Then, for the business object represented by the second data, match the business behavior represented by the third data, so as to form a business optimization strategy, achieve accurate business optimization, and improve the performance of the business team and customer conversion rate.

[0137] It should be understood that, although each step in the flowcharts involved in the above-mentioned embodiments is displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0138] Based on the same inventive concept, the embodiment of the present application also provides a resource allocation device for implementing the data analysis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the embodiments of one or more resource allocation devices provided below can refer to the limitations on the data analysis method above, and the specific limitations will not be repeated here.

[0139] The resource allocation device provided in this embodiment may be integrated into a terminal device. Fig. 9 As shown, the resource allocation device may include: a data partitioning module 301 , a first analysis module 302 , a second analysis module 303 , a third analysis module 304 , and a strategy formulation module 305 .

[0140] The resource allocation device described above first divides the business data into first data under different dimensions through the data division module 301; then obtains a first interpretation value of the first data according to the business success rate and business data proportion corresponding to different business agents under the first data through the first analysis module 302; the first interpretation value represents the degree of association between the first data and the business data of the business agent; then obtains a first surprise value of the first data according to the difference between the business data proportions corresponding to different business agents under the first data through the second analysis module 303; then determines a second interpretation value corresponding to each dimension based on the first interpretation value of the first data through the third analysis module 304, and determines a second surprise value corresponding to each dimension based on the first surprise value of the first data; then determines at least one business optimization strategy corresponding to the business data according to the first interpretation value, the first surprise value, the second interpretation value and the second surprise value through the strategy formulation module 305.

[0141] In some embodiments, the first analysis module 302 is further specifically configured to:

[0142] Obtaining a first service success rate corresponding to the first service seat under the first data, and a second service success rate corresponding to the second service seat;

[0143] Acquire a proportion of the first business data corresponding to the first business seat and a proportion of the second business data corresponding to the second business seat;

[0144] A first interpretation value of the first data is obtained based on a first difference between the first business success rate and the second business success rate, a second difference between the first business data proportion and the second business data proportion, a first average value of the first business success rate and the second business success rate, and a second average value of the first business data proportion and the second business data proportion.

[0145] In some embodiments, the first analysis module 302 is further specifically configured to:

[0146] Obtaining a first product value of the first difference and the second average value, and a second product value of the second difference and the first average value;

[0147] An interpretation value of the first data is determined according to a sum of the first product value and the second product value.

[0148] In some embodiments, the second analysis module 303 is further specifically configured to:

[0149] The surprise value of the first data is determined according to a first difference between the first business data proportion and the second average value, and a second difference between the second business data proportion and the second average value.

[0150] In some embodiments, the second analysis module 303 is further specifically configured to:

[0151] Determine a first relative entropy between the first business data proportion and the second average value;

[0152] Determine a second relative entropy between the second business data proportion and the second average value;

[0153] The surprise value of the first data is determined according to the product of the first relative entropy and the second relative entropy.

[0154] In some embodiments, the policy formulation module 305 is further specifically configured to:

[0155] Adding first data whose first interpretation value is greater than a first threshold into the first set;

[0156] Adding the first data of the first set whose second interpretation value is greater than a second threshold into the second set;

[0157] According to the first surprise value of the first data and the second surprise value of the dimension, target data is selected from the second set, and at least one of the business optimization strategies is determined based on the target data.

[0158] In some embodiments, the policy formulation module 305 is further specifically configured to:

[0159] From the multiple dimensions of the second set, select M dimensions whose second surprise value is greater than a third threshold; from each dimension of the M dimensions, select N first data whose first surprise value is greater than a fourth threshold to obtain the target data.

[0160] In some embodiments, the policy formulation module 305 is further specifically configured to:

[0161] Extracting second data related to the attributes of the business object and third data related to the business behavior from the target data;

[0162] The business behavior of the third data is matched with respect to the business object of the second data to form at least one of the business optimization strategies.

[0163] By adopting the scheme of the embodiment of the present application, the business data is first divided into first data under different dimensions, and each dimension has at least two first data. Compared with business analysis of data of a single dimension, it has a more comprehensive data analysis basis, which is helpful to formulate more accurate and targeted business optimization strategies in the future. Then, according to the business success rate and business data proportion corresponding to different business seats under the first data, the first interpretation value of the first data is obtained. The first interpretation value represents the degree of association between the first data and the business data of the business seat, so that the performance differences generated by the first data of multiple dimensions can be analyzed based on the first interpretation value for interpretation; then, according to the difference between the business data proportions corresponding to different business seats under the first data, the first surprise value of the first data is obtained, so that the first data that has a significant impact on business performance can be identified without relying on manual judgment. Then, based on the first interpretation value and the first surprise value of the first data, the second interpretation value and the second surprise value corresponding to each dimension are obtained, so that the impact of the overall interpretation value and surprise value of the dimension on the business can be considered. Finally, based on the first explanation value, the first surprise value, the second explanation value and the second surprise value, at least one business optimization strategy corresponding to the business data is obtained, so that key dimensions that are highly explanatory and have a positive impact on business improvement, as well as key first data that are highly explanatory and have a positive impact on the business under the key dimensions can be selected, and the corresponding business optimization strategies can be efficiently and accurately generated based on the selected key dimensions and key first data.

[0164] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0165] In one embodiment, the computer device is a terminal device, for example, its internal structure diagram can be as follows: Fig.10As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a data analysis method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0166] Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0167] Based on the same inventive concept, an embodiment of the present application also provides a computer-readable storage medium, which may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0168] Since the computer program stored in the computer-readable storage medium can execute any one of the data analysis methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any one of the data analysis methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0169] Based on the same inventive concept, the embodiment of the present application further provides a computer program product or a computer program, which includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in various optional implementations in the above embodiments.

[0170] It should be noted that the resource consumption data, cost unit price data, blood relationship data and other data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of relevant countries and regions. A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0171] Any reference to the memory, database or other media used in the embodiments provided in the present application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (Re RAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0172] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0173] In the above-mentioned resource allocation device, computer-readable storage medium, computer equipment, and computer program product embodiments, the description of each embodiment has its own emphasis. For parts not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process and beneficial effects of the resource allocation device, computer-readable storage medium, computer program product, computer equipment and its corresponding units described above can refer to the description of the data analysis method in the above embodiment, and will not be repeated here.

[0174] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0175] The above is a detailed introduction to a data analysis method, device, computer equipment, computer-readable storage medium and computer program product provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A data analysis method, characterized in that: The method comprises: Divide the business data into first data under different dimensions; According to the business success rates and business data proportions corresponding to different business seats under the first data, a first interpretation value of the first data is obtained; the first interpretation value represents the degree of association between the first data and the business data of the business seat; obtaining a first surprise value of the first data according to a difference between business data proportions corresponding to different business seats under the first data; Determine a second interpretation value corresponding to each dimension based on the first interpretation value of the first data, and determine a second surprise value corresponding to each dimension based on the first surprise value of the first data; At least one business optimization strategy corresponding to the business data is determined according to the first explanation value, the first surprise value, the second explanation value, and the second surprise value.

2. The method according to claim 1, characterized in that The obtaining, according to the business success rates and business data proportions corresponding to different business seats under the first data, a first interpretation value of the first data includes: Obtaining a first service success rate corresponding to the first service seat under the first data, and a second service success rate corresponding to the second service seat; Acquire a proportion of the first business data corresponding to the first business seat and a proportion of the second business data corresponding to the second business seat; A first interpretation value of the first data is obtained based on a first difference between the first business success rate and the second business success rate, a second difference between the first business data proportion and the second business data proportion, a first average value of the first business success rate and the second business success rate, and a second average value of the first business data proportion and the second business data proportion.

3. The method according to claim 2, characterized in that The obtaining, according to the business success rates and business data proportions corresponding to different business seats under the first data, a first interpretation value of the first data also includes: Obtaining a first product value of the first difference and the second average value, and a second product value of the second difference and the first average value; An interpretation value of the first data is determined according to a sum of the first product value and the second product value.

4. The method according to claim 3, characterized in that The obtaining, according to the difference between the proportions of business data corresponding to different business seats under the first data, a first surprise value of the first data includes: The surprise value of the first data is determined according to a first difference between the first business data proportion and the second average value, and a second difference between the second business data proportion and the second average value.

5. The method according to claim 4, characterized in that The determining the surprise value of the first data according to a first difference between the first business data proportion and the second average value, and a second difference between the second business data proportion and the second average value, comprises: Determine a first relative entropy between the first business data proportion and the second average value; Determine a second relative entropy between the second business data proportion and the second average value; The surprise value of the first data is determined according to the product of the first relative entropy and the second relative entropy.

6. The method according to claim 1, characterized in that The determining, according to the first explanation value, the first surprise value, the second explanation value, and the second surprise value, at least one service optimization strategy corresponding to the service data comprises: Adding first data whose first interpretation value is greater than a first threshold into the first set; Adding the first data of the first set whose second interpretation value is greater than a second threshold into the second set; According to the first surprise value of the first data and the second surprise value of the dimension, target data is selected from the second set, and at least one of the business optimization strategies is determined based on the target data.

7. The method according to claim 6, characterized in that The selecting target data from the second set according to the first surprise value of the first data and the second surprise value of the dimension includes: From the multiple dimensions of the second set, select M dimensions whose second surprise value is greater than a third threshold; from each dimension of the M dimensions, select N first data whose first surprise value is greater than a fourth threshold to obtain the target data.

8. The method according to claim 6 or 7, characterized in that The determining at least one of the service optimization strategies based on the target data includes: Extracting second data related to the attributes of the business object and third data related to the business behavior from the target data; The business behavior of the third data is matched with respect to the business object of the second data to form at least one of the business optimization strategies.

9. A data analysis device, characterized in that: The device comprises: A data division module, used to divide the business data into first data under different dimensions; A first analysis module, configured to obtain a first interpretation value of the first data according to the business success rates and business data proportions corresponding to different business agents under the first data; the first interpretation value represents the degree of association between the first data and the business data of the business agent; A second analysis module is used to obtain a first surprise value of the first data according to a difference between business data proportions corresponding to different business seats under the first data; a third analysis module, configured to determine a second interpretation value corresponding to each dimension based on the first interpretation value of the first data, and to determine a second surprise value corresponding to each dimension based on the first surprise value of the first data; A strategy formulation module is used to determine at least one business optimization strategy corresponding to the business data according to the first interpretation value, the first surprise value, the second interpretation value and the second surprise value.

10. A computer device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps of the data analysis method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps of the data analysis method according to any one of claims 1 to 8.

12. A computer software, characterized in that: The computer software comprises a computer program, and the computer program is used by a processor to execute the steps of the data analysis method according to any one of claims 1 to 8.

Images