Business attribution analysis method and device, equipment and storage medium

Abstract

The invention discloses a business attribution analysis method and device, equipment and a storage medium. The method comprises the following steps: determining an atomic unit of a to-be-analyzed service and an atomic process executed by the atomic unit on the service; determining a business phenomenon mode; calculating the support degree of the atomic unit to the business phenomenon mode, and adding a label to the atomic unit based on the business phenomenon mode and the support degree to obtain a labeled evidence of the atomic unit; and summarizing the support degrees of all the atomic units to the business phenomenon modes based on the tagging evidence of the atomic units, and extracting the business phenomenon modes of which the support degrees are greater than a preset threshold value as an attribution analysis result of business phenomenon occurrence. According to the business attribution analysis method and device, the business attribution problem is disassembled into the atomic units of business execution, the causal relationship between the business phenomenon and the influence factors is quantified based on the business phenomenon mode analysis of the atomic units, and the accuracy and credibility of business attribution analysis are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and more specifically, to a business attribution analysis method, apparatus, device, and storage medium. Background Art

[0002] With the rapid development of the digital economy, various industries are advancing digital transformation. This is particularly true in the field of intelligent transportation, where improving traffic management efficiency and service quality through digital means has become a key issue. For example, highways offer core services including construction, operation (security), maintenance, operations (collection), and services. These services are complexly coupled. For example, ensuring operation and collection depends on traffic demand (traffic volume), which in turn is influenced by numerous factors, including road construction and maintenance, operational security, service quality, road conditions, and weather.

[0003] Existing technologies often focus on local monitoring when analyzing traffic operations, such as real-time monitoring of congestion and accidents. However, they fail to adequately analyze the underlying causes of these phenomena, such as changes in traffic flow, road spatial characteristics, and environmental factors like traffic weather. This isolated monitoring approach results in an incomplete attribution analysis of business phenomena, which in turn leads to poor prediction, early warning, and response effectiveness. Furthermore, with the advancement of business dataization, the number of data dimensions and granularity are increasing, making attribution analysis based on multidimensional data increasingly difficult. In existing macro-analysis methods, the contributions of different individuals in a group to business phenomena may offset each other, thereby hiding the true causes of business problems. Summary of the Invention

[0004] The embodiments of the present application provide a business attribution analysis method, apparatus, device, and storage medium to at least solve the technical problem of insufficient accuracy and credibility of business phenomenon pattern analysis in related technologies.

[0005] According to one aspect of an embodiment of the present application, a business attribution analysis method is provided, including:

[0006] Determine the atomic units of the business to be analyzed and the atomic processes performed by the atomic units on the business. The atomic units are the basic execution units that constitute business activities during the business execution process, and the atomic processes are the basic business activities performed by the atomic units during the business execution process.

[0007] Determining a business phenomenon pattern, where the business phenomenon pattern is an action pattern that induces the business phenomenon during business execution and is summarized based on atomic process data of atomic units;

[0008] Calculating the support of the atomic unit for the business phenomenon pattern, adding labels to the atomic unit based on the business phenomenon pattern and the support, and obtaining labeled evidence of the atomic unit;

[0009] Based on the labeled evidence of the atomic units, the support of all atomic units for the business phenomenon pattern is summarized, and the business phenomenon pattern with support greater than a preset threshold is extracted as the attribution analysis result of the occurrence of the business phenomenon.

[0010] According to another aspect of an embodiment of the present application, a business attribution analysis device is provided, including:

[0011] The atomic unit definition module is used to determine the atomic unit of the business to be analyzed and the atomic process performed by the atomic unit on the business. The atomic unit is the basic execution unit that constitutes the business activity during the business execution process, and the atomic process is the basic business activity performed by the atomic unit during the business execution process;

[0012] A business phenomenon pattern definition module is used to determine a business phenomenon pattern, where the business phenomenon pattern is an action pattern that induces the occurrence of a business phenomenon during business execution and is summarized based on atomic process data of atomic units;

[0013] a calculation module, configured to calculate the support of the atomic unit for the business phenomenon pattern, and add labels to the atomic unit based on the business phenomenon pattern and the support, to obtain labeled evidence of the atomic unit;

[0014] The summary analysis module is used to summarize the support of all atomic units for business phenomenon patterns based on the labeled evidence of the atomic units, and extract business phenomenon patterns with support greater than a preset threshold as attribution analysis results of the occurrence of business phenomena.

[0015] According to another aspect of an embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to execute the business attribution analysis method through the computer program.

[0016] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned business attribution analysis method when running.

[0017] The technical solutions provided by the embodiments of the present application may have the following beneficial effects:

[0018] This application achieves refined attribution analysis of business phenomena by determining atomic units and their atomic processes, defining business phenomenon patterns, and calculating the support of atomic units for business phenomenon patterns. By breaking down the business into the basic execution unit (atomic unit) level, the causal relationship between business phenomena and influencing factors is quantified from a micro perspective, avoiding the hidden problems caused by the mutual offset of contributions from different atomic units in traditional macro and group analysis methods. This refined analysis method can more accurately capture the causes of business phenomena, and by calculating the support of atomic units for business phenomenon patterns and labeling atomic units based on the support, the accuracy and credibility of business attribution analysis are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 is a flow chart of a business attribution analysis method according to an embodiment of the present application;

[0021] Figure 2 This is a schematic diagram of highway business and data element analysis according to an embodiment of the present application;

[0022] Figure 3 This is a schematic diagram of a highway toll collection business model according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of a vehicle travel process in two comparison periods according to an embodiment of the present application;

[0024] Figure 5 Schematic diagram of a business attribution analysis device according to an embodiment of the present application;

[0025] Figure 6 It is a schematic structural diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Take highway business as an example, Figure 2 As shown in Figure 1, the core businesses of expressways include construction (construction), operation support (management), maintenance (maintenance), operating and charging (transportation), and services (services). The core data elements underlying these businesses can be categorized into three categories: "vehicles / passengers," "roads," and "environment." "Vehicles / passengers" represent traffic demand, "roads" represent traffic supply, and "environment" represents the variables influencing both demand and supply. As an open and complex system, the transportation system is inherently coupled between any two businesses, between multiple businesses, and between businesses and data elements. For example, operating and charging depends on both operation support and service businesses, as operational support and service levels directly or indirectly impact traffic demand, represented by "vehicles / passengers." With the collection and improvement of data elements related to traffic demand, traffic supply, and the traffic environment, objectively and quantifying the attribution relationships between these different businesses has become a key challenge in business development.

[0029] Based on this, an embodiment of the present application provides a business attribution method, which breaks down the business into atomic units of business execution, quantifies the attribution relationship between businesses from the perspective of atomic units, and obtains the attribution relationship between businesses after aggregation. For example, in the field of intelligent transportation, highway travel is composed of individual travelers, and changes in individual travel behavior have a direct or indirect causal relationship with various highway businesses. Therefore, when the travel OD remains unchanged, the change in the travel route is related to the difference and change in the security level and service level on the route before and after the change; the change in travel OD represents the change in travel demand, etc.

[0030] The following is combined with Figure 1 The business attribution analysis method of the embodiment of this application is described in detail. Figure 1 As shown, the method mainly includes the following steps:

[0031] S101 determines the atomic unit of the service to be analyzed and the atomic process performed by the atomic unit on the service.

[0032] Among them, the atomic unit is the basic execution unit that constitutes the business activity during the business execution process, and the atomic process is the basic business activity performed by the atomic unit during the business execution process.

[0033] In some embodiments, an atomic unit is the lowest or smallest execution unit that constitutes a business activity during the business execution process; an atomic process is the lowest or smallest business activity process executed by an atomic unit during the business execution process.

[0034] For simplicity of expression, we first conduct an abstract business model for the toll collection business of the highway. Figure 3 It is a modeling model for highway toll collection. For any highway toll collection entity X, the composition of its toll amount is determined by a number of related travel OD demands and the choice of paths between ODs. Here, the relevant meaning is that a certain path selected by the travel user who meets the OD demand passes through the toll collection area of ​​the highway toll collection entity X. Figure 3 In this example, assume that path 1 is entirely attributed to charging entity X, while path 2 is entirely attributed to charging entity X, and that these two paths are the only ones currently present at OD. This assumption is made for simplicity and does not affect the conclusions of this application. Furthermore, the return trip from D to O is not considered; this can be achieved by analyzing the trip from D to O independently using the same method. Furthermore, path 2 does not necessarily have to be a highway; it could also be a surface road.

[0035] Assume t i and t j These are two comparative evaluation cycles for the business being analyzed. i and Ω j They are t i and t j The data set of the business to be analyzed is collected during the cycle. The data set records all the atomic units that constitute the business to be analyzed and all the data items that execute the atomic processes related to the business to be analyzed.

[0036]

[0037] in, is the dataset of the kth atomic unit in the dataset, with 1≤k≤K i , where K i defines the number of atomic units in the dataset; in In the dataset, id(i,k) defines the unique ID of the atomic unit. Note that this ID is a globally unique identifier and can be compared with data in other datasets. cnt(i,k) defines t i The number of atomic processes that exist in the k-th atomic unit during the evaluation period. data(i,k) defines t iThe data of all atomic processes of the k-th atomic unit in the evaluation period:

[0038]

[0039] Defines the data collected by the zth atomic process and related to the business to be analyzed.

[0040] In the toll collection business scenario, in one embodiment, a traveling vehicle is regarded as the atomic unit of toll collection, and a travel process is regarded as the atomic process of toll collection. It should be noted that atomic units of other granularities can also be defined, such as toll stations, road section companies, etc., and atomic processes can be abstractly defined based on the business of toll stations and road section companies. If the entrance (exit) toll station is defined as an atomic unit, the traffic transfer of different types to each exit toll station (from each entrance toll station) can be defined as an atomic process, and the business phenomenon pattern analysis can be further performed based on the traffic differences and path differences between the toll stations. I will not go into details here.

[0041] For any given traveling vehicle, in order to conduct an attribution analysis of the business phenomenon of changes in its travel process in different cycles, it is necessary to set a comparative evaluation cycle (generally compared with the previous cycle) and collect the vehicle travel process of the two comparative cycles. Figure 4 It is the travel process of a vehicle in two comparison periods. Figure 4 A is the travel process in the comparison period, and there are 4 trips from O to D via path 1; Figure 4 B is the travel process in the current cycle. There is one trip from O to D via path1, one trip from O to D via path2, and three trips from O1 to D via path3.

[0042] S102 determines a business phenomenon pattern, where the business phenomenon pattern is an action pattern that induces the occurrence of business phenomena and is summarized based on atomic process data of atomic units during the execution of the business.

[0043] In one embodiment, business phenomenon pattern data is pre-associated with the business phenomenon, and the corresponding business phenomenon pattern is determined based on the business phenomenon to be attributed and analyzed. In one embodiment, one business phenomenon corresponds to at least two action patterns that induce the occurrence of the business phenomenon, that is, one business phenomenon corresponds to at least two business phenomenon patterns.

[0044] In an exemplary scenario, if the amount of fees charged by charging entity X decreases during a certain evaluation period, it is necessary to analyze the changes in each OD demand related to charging entity X to determine the actual amount of fees charged by charging entity X. Figure 3Analyze the OD shown. Based on the experience of business experts or the business phenomenon association patterns associated with this business phenomenon, there are three possible business phenomenon patterns related to this OD and the decrease in charges by charging entity X: (1) The traffic at O ​​decreases, indicating a decrease in travel demand at O; (2) The traffic at D decreases, indicating a decrease in travel demand to D; (3) The operational support level, service level, or environment of path 1 deteriorates over a period of time or continuously, or the operational support level, service level, or environment of path 2 improves over a period of time or continuously, resulting in traffic shifting to path 2.

[0045] Furthermore, for period t i For a given atomic unit in the data set, if its unique ID is id(i,k), let the period t j There is an atomic unit in the data set. Let its unique ID be id(j,k′). If id(i,k)=id(j,k′), then data(i,k) and data(j,k′) serve as input for business phenomenon pattern mining.

[0046] In one embodiment, for a given business phenomenon pattern δ n , there is 1≤n≤N, where N is the number of business phenomenon patterns pre-set for the business to be analyzed.

[0047] S103 calculates the support of the atomic unit for the business phenomenon pattern, adds labels to the atomic unit based on the business phenomenon pattern and the support, and obtains labeled evidence of the atomic unit.

[0048] In one embodiment, the support degree of each atomic unit for each business phenomenon pattern is calculated based on the atomic process data set of the atomic unit within a preset comparison period and the preset function of each business phenomenon pattern.

[0049] In one embodiment, calculating the support degree of the atomic unit for the business phenomenon pattern includes: setting a judgment function corresponding to each business phenomenon pattern; reading the atomic process data set of each atomic unit within a preset comparison period based on the judgment function, and outputting the support degree of each atomic unit for the business phenomenon pattern.

[0050] Specifically, for any business phenomenon pattern δ n , we need to set a continuous judgment function F n (,), the judgment function is used to read the data set of each atomic unit within the comparison period and output the business phenomenon pattern δ n The support of . That is:

[0051] p(id(i,k),δ n )=F n(data(i,k),data(j,k′))

[0052] Among them, p(id(i,k),δ n ) is a result in [0,1].

[0053] In special cases, data(i,k) or data(j,k′) may be {}, indicating that the data is in period t i or t j Taking the toll collection service as an example, some vehicles have no data in the periodic dataset, which means that the vehicles are not online or are traveling at high speeds. Pattern analysis can also be performed on them.

[0054] Among them, the specific calculation method of the continuous judgment function needs to be determined according to the business scenario. Its core goal is to quantify the support degree of the atomic unit for the business phenomenon pattern by analyzing the atomic process data of the atomic unit in two comparison cycles.

[0055] For example, assuming that the business phenomenon pattern is "traffic is transferred from path 1 to path 2", the specific implementation is as follows: First, extract the feature: x1 (atomic unit in t i The number of times path1 is passed in the cycle); x2 (atomic unit in t j The number of times path1 is passed in the cycle); y1 (atomic unit in t i The number of times path2 is passed in the cycle); y2 (atomic unit in t j The number of times path2 is passed in a cycle).

[0056] Calculate feature differences: Δx = x2 - x1; Δy = y2 - y1.

[0057] Set the business phenomenon mode judgment function:

[0058]

[0059] Here, ∈ is a small positive number used to prevent the denominator from being zero.

[0060] Other methods may also be used to generate a determination function, which is not specifically limited in the embodiments of the present application, such as a determination function using a machine learning model, a determination function based on a probability model, etc.

[0061] Furthermore, when the support degree of the atomic unit for the business phenomenon pattern is greater than or equal to a preset first threshold, it is determined that the atomic unit and the business phenomenon pattern are correlated; the atomic unit with the correlation is labeled, and the business phenomenon pattern related to the atomic unit and the corresponding support degree are marked to obtain labeled evidence of the atomic unit. The specific threshold value is not specifically limited in this application, for example, the threshold is 0.5.

[0062] If p(id(i,k),δ n )>γ, where γ is the set threshold, which means that the atomic sample identified by id(i,k) has a significant effect on the business phenomenon pattern δ n There is a correlation, and the correlation measure is p(id(i,k),δ n ), based on the result analysis, the atomic unit id (i, k) is labeled and recorded as a triple, which includes the atomic unit, the business phenomenon pattern, and the support degree of the atomic unit for the business phenomenon pattern:

[0063] θ x = <id(i,l),δ n ,p(id(i,k),δ n )>;

[0064] Among them, θ x is a labeled evidence, x is the subscript. Let Θ i,j It is t i and t j Cycles compare the set of all labeled evidence.

[0065] In an exemplary scenario, three business phenomenon patterns of charging amount reduction are defined: 1) traffic of O decreases; 2) traffic of D decreases; 3) traffic of path 1 is transferred to path 2.

[0066] contrast Figure 4 Given the changes in the vehicle's travel process, we can conclude that O's travel times decreased from 4 to 2; D's arrival times increased from 4 to 5; path1 decreased from 4 to 1, and path2 increased from 0 to 1; O1 newly appeared, and its travel times increased from 0 to 3; path3 increased from 0 to 3.

[0067] By correlating the changes in vehicle trip frequency with business phenomenon patterns, we can draw the following conclusions:

[0068] 1) O's trip count dropped from 4 to 2, supporting the conclusion of business phenomenon model (I) that O's traffic volume decreased;

[0069] 2) The number of arrivals at D increases from 4 to 5, which does not support the conclusion of business phenomenon model (2) that the traffic at D decreases;

[0070] 3) The traffic on path 1 drops from 4 to 1, and the traffic on path 2 increases from 0 to 1, supporting the conclusion that the traffic on path 1 is transferred to path 2 in business phenomenon model (3).

[0071] The above analysis demonstrates that changes in bicycle trips can be used to establish support for business phenomenon patterns. Furthermore, the support index for business phenomenon patterns can be quantified based on the intensity of changes in bicycle trip counts, which will not be discussed further here. It is important to note that in the support calculation, recent historical data on changes in bicycle trip counts can be used as a reference. If a strong trend is observed, the support index for the business phenomenon pattern should be continuously strengthened.

[0072] S104 summarizes the support of all atomic units for the business phenomenon pattern based on the labeled evidence of the atomic units, and extracts the business phenomenon pattern with support greater than a preset threshold as the attribution analysis result of the occurrence of the business phenomenon.

[0073] In one embodiment, the support of all atomic units for business phenomenon patterns is summarized based on the labeled evidence of atomic units, including: obtaining the labeled evidence of all atomic units; inputting the labeled evidence of all atomic units into a preset summary calculation model, and outputting a summary data set containing all business phenomenon patterns and their corresponding support.

[0074] In one embodiment, the function of the summary calculation is set to F(), where the input Θ i,j On this basis, the support set of each business phenomenon model is obtained.

[0075]

[0076] Where (δ n ,p n ) n For δ n support.

[0077] In one embodiment, the summary calculation model can be constructed based on statistical theory methods; or, based on DS evidence reasoning theory methods; or, based on a preset neural network. The specific method is not specifically limited in this application.

[0078] For example, based on a statistical method, all support degrees of the business phenomenon pattern are accumulated to obtain a final support degree.

[0079] In one embodiment, it also includes: obtaining the support of all atomic units for the business phenomenon model based on the DS evidence reasoning theory method to obtain a data fusion result; performing cross-validation based on the data fusion result, and refining the business phenomenon model based on the cross-validation result.

[0080] Specifically, since this step contains the support results of all atomic units for the business phenomenon model, the support of all atomic units is regarded as evidence for the establishment of the business phenomenon model. Data fusion calculations and cross-validation between evidence can be performed based on algorithms such as evidence reasoning DS. In particular, the support of atomic units with different sources and large differences for the same business phenomenon model needs to be strengthened. For example, if other OD traffic passing through path 1 is also transferred, the support of business phenomenon model (3) can be further strengthened. Furthermore, if OD traffic from multiple different paths is transferred to path 2, business phenomenon model (3) can be refined, indicating that the operation guarantee level, service level, or environment of path 2 has been continuously optimized over a certain period of time.

[0081] Furthermore, under the business phenomenon model (1), two new models are refined:

[0082] (4) If O1 is very close to O, then O1 and O are in a competitive relationship, which means that there may be factors that affect the access to the highway from O, or that path 1 starting from O is deteriorating, indirectly supporting business phenomenon model (3).

[0083] (V) If the distance between O1 and O is very far, then O1 and O are in a non-competitive relationship, indicating that the vehicle has left its original activity area, thereby strengthening the support business phenomenon model (I).

[0084] Furthermore, the time label and other characteristics of the business model can be supplemented based on the analysis results. Figure 4 As shown in Figure B, if the itineraries are arranged by time, it can be explained that the changes in the itinerary occurred in the later stage of the evaluation. Based on the time tag, we can further associate it with specific factors such as traffic operation guarantees, changes in service capabilities, or changes in the traffic operation environment to further locate the problem.

[0085] By summarizing and analyzing the time tags of business phenomenon patterns across multiple atomic units and the recent changes in the support for these patterns, we can determine the temporal patterns of these patterns. This helps us determine whether they are random phenomena caused by localized random factors or long-term trends. If a strong trend is observed, the support index for the business phenomenon pattern should be strengthened.

[0086] Furthermore, based on the summary data, it is determined whether the support degree of the business phenomenon pattern is greater than or equal to the preset second threshold; if it is greater than or equal to the preset second threshold, it is determined that the business phenomenon pattern is established; the established business phenomenon patterns are sorted from high to low according to the size of the support degree to obtain the attribution analysis results.

[0087] In one embodiment, business phenomenon patterns with high support are sorted from high to low according to their support, and the sorted results are provided to business personnel. These results directly reflect the impact of different business phenomenon patterns on business changes, helping business personnel quickly identify key issues and trends.

[0088] If you want to conduct further sample analysis and offline investigations, you can select atomic unit data that is highly correlated with the business phenomenon pattern based on the quantitative results of the atomic unit and the business phenomenon pattern. This atomic unit data can reveal the specific manifestations and potential causes of the business phenomenon pattern.

[0089] Based on the detailed attribution analysis results, business personnel can formulate targeted optimization measures to improve business efficiency and service quality.

[0090] For example, business personnel discovered that the business phenomenon pattern of traffic shifting from path 1 to path 2 had the highest support rate and decided to further analyze this phenomenon. By examining the atomic unit data highly correlated with this pattern (such as the driving records of specific vehicles), they discovered that the operational assurance level and service quality of path 2 had improved recently, while path 1 had experienced repeated congestion. Based on this detailed information, business personnel can develop targeted optimization measures, such as improving traffic conditions on path 1 or further enhancing the service quality of path 2.

[0091] This application breaks down the business attribution problem into the atomic units of business execution, and quantifies the causal relationship between business phenomena and influencing factors based on the business phenomenon pattern analysis of atomic units, which significantly improves the accuracy and credibility of business phenomenon attribution analysis.

[0092] By adopting one or more of the above-mentioned embodiments, it is possible to achieve the following: 1) It is simple to quantify the attribution relationship from the perspective of atomic units; 2) Although different atomic units are random, a relatively objective attribution relationship can be obtained after group aggregation, and the credibility of the attribution analysis can be further enhanced based on cross-validation of samples and trend analysis in the time and space dimensions; 3) The correlation between atomic units and business phenomena can be quantified, that is, the typical atomic units of business phenomena can be obtained, and further targeted investigations can be carried out.

[0093] This application quantifies the causal relationship between businesses from the micro perspective of atomic units (the lowest level execution units) and atomic processes (the lowest level execution processes), avoiding the hidden problems caused by the offsetting contributions of different atomic units in traditional macro and group analysis methods. This refined analysis method can more accurately capture the causes of business phenomena, and by calculating the support of atomic units for business phenomenon patterns and labeling atomic units based on the support, it realizes the quantitative analysis of business phenomenon patterns. In addition, this method can also identify atomic units that have a significant impact on business phenomena and quantify the degree of their impact.

[0094] This method is not only applicable to intelligent transportation, such as highway toll collection, but can also be extended to other business scenarios requiring attribution analysis. By defining different atomic units and business phenomenon patterns, this method can be flexibly applied to a variety of industries and fields, demonstrating broad applicability. It significantly improves the accuracy and reliability of business phenomenon pattern analysis, providing strong support for the digital transformation of business management.

[0095] According to another aspect of the embodiment of the present application, a business attribution analysis device for implementing the above-mentioned business attribution analysis method is also provided. Figure 5 As shown, the device includes:

[0096] Atomic unit definition module 501 is used to determine the atomic unit of the business to be analyzed and the atomic process performed by the atomic unit on the business. The atomic unit is the basic execution unit that constitutes the business activity during the business execution process, and the atomic process is the basic business activity performed by the atomic unit during the business execution process;

[0097] The business phenomenon pattern definition module 502 is used to determine the business phenomenon pattern. The business phenomenon pattern is the action pattern that induces the occurrence of the business phenomenon during the business execution process, which is summarized based on the atomic process data of the atomic unit;

[0098] A calculation module 503 is used to calculate the support of the atomic unit for the business phenomenon pattern, add labels to the atomic unit based on the business phenomenon pattern and the support, and obtain labeled evidence of the atomic unit;

[0099] The summary analysis module 504 is used to summarize the support of all atomic units for business phenomenon patterns based on the labeled evidence of atomic units, and extract business phenomenon patterns with support greater than a preset threshold as attribution analysis results of the business phenomenon occurrence.

[0100] It should be noted that the business attribution analysis device provided in the above embodiment, when executing the business attribution analysis method, only uses the division of the above-mentioned functional modules as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the business attribution analysis device provided in the above embodiment and the business attribution analysis method embodiment are based on the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0101] According to another aspect of an embodiment of the present application, an electronic device corresponding to the business attribution analysis method provided in the aforementioned embodiment is also provided to execute the aforementioned business attribution analysis method.

[0102] Please refer to Figure 6 , which shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 6 As shown, the electronic device includes: a processor 600, a memory 601, a bus 602 and a communication interface 603, and the processor 600, the communication interface 603 and the memory 601 are connected through the bus 602; the memory 601 stores a computer program that can be run on the processor 600, and when the processor 600 runs the computer program, it executes the business attribution analysis method provided by any of the aforementioned embodiments of the present application.

[0103] The memory 601 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage. The system network element communicates with at least one other network element via at least one communication interface 603 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0104] Bus 602 can be an ISA bus, a PCI bus, or an EISA bus. Buses can be divided into address buses, data buses, and control buses. Memory 601 is used to store programs, and processor 600 executes the programs upon receiving execution instructions. The business attribution analysis method disclosed in any of the aforementioned embodiments of this application can be applied to or implemented by processor 600.

[0105] The processor 600 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor 600 or by software instructions. The above processor 600 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 601 , and the processor 600 reads the information in the memory 601 and completes the steps of the above method in combination with its hardware.

[0106] The electronic device provided in the embodiment of the present application and the business attribution analysis method provided in the embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented by them.

[0107] According to another aspect of the embodiments of the present application, a computer-readable storage medium corresponding to the business attribution analysis method provided in the aforementioned embodiments is also provided, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the business attribution analysis method provided in any of the aforementioned embodiments.

[0108] It should be noted that examples of computer-readable storage media may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0109] The computer-readable storage medium provided in the above-mentioned embodiments of the present application and the business attribution analysis method provided in the embodiments of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by the application programs stored therein.

[0110] The technical features of the above embodiments can 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.

[0111] The above embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A business attribution analysis method, characterized in that: include: Determine the atomic units of the business to be analyzed and the atomic processes performed by the atomic units on the business. The atomic units are the basic execution units that constitute business activities during the business execution process, and the atomic processes are the basic business activities performed by the atomic units during the business execution process. Determining a business phenomenon pattern, where the business phenomenon pattern is an action pattern that induces the business phenomenon during business execution and is summarized based on atomic process data of atomic units; Calculating the support of the atomic unit for the business phenomenon pattern, adding labels to the atomic unit based on the business phenomenon pattern and the support, and obtaining labeled evidence of the atomic unit; Based on the labeled evidence of the atomic units, the support of all atomic units for the business phenomenon pattern is summarized, and the business phenomenon pattern with support greater than a preset threshold is extracted as the attribution analysis result of the occurrence of the business phenomenon.

2. The method according to claim 1, characterized in that Calculating the support degree of the atomic unit for the business phenomenon model includes: Set the decision function corresponding to each business phenomenon pattern; The atomic process data set of each atomic unit within a preset comparison period is read based on the determination function, and the support degree of each atomic unit to the business phenomenon pattern is output.

3. The method according to claim 2, characterized in that Adding labels to the atomic units based on the business phenomenon pattern and support to obtain labeled evidence of the atomic units includes: When the support degree of the atomic unit for the business phenomenon pattern is greater than or equal to a preset first threshold, determining that there is a correlation between the atomic unit and the business phenomenon pattern; The atomic units with correlation are labeled, and the business phenomenon patterns related to the atomic units and the corresponding support are marked to obtain the labeled evidence of the atomic units.

4. The method according to claim 1, wherein Based on the labeled evidence of the atomic units, the support of all atomic units for the business phenomenon pattern is summarized, including: Obtain labeled evidence for all atomic units; The labeled evidence of all the atomic units is input into a preset summary calculation model, and a summary data set containing all business phenomenon patterns and their corresponding support is output.

5. The method according to claim 4, characterized in that Before inputting the labeled evidence of all the atomic units into the preset summary calculation model, it also includes: Constructing the summary calculation model based on statistical theory methods; or, Construct the summary calculation model based on the DS evidence reasoning theory method; or, The summary calculation model is constructed based on a preset neural network.

6. The method according to claim 1, wherein Extract business phenomenon patterns with support greater than a preset threshold as attribution analysis results of the business phenomenon, including: Determine whether the support degree of the business phenomenon pattern is greater than or equal to a preset second threshold based on the summary data; When the value is greater than or equal to the preset second threshold, determining that the business phenomenon mode is established; The established business phenomenon patterns are sorted from high to low according to the degree of support to obtain the attribution analysis result.

7. The method according to claim 1, characterized in that Also includes: Obtain the support of all atomic units for business phenomenon patterns based on the DS evidence reasoning theory method to obtain data fusion results; Cross-validation is performed based on the data fusion result, and the business phenomenon model is refined based on the cross-validation result.

8. A business attribution analysis device, characterized in that: include: The atomic unit definition module is used to determine the atomic unit of the business to be analyzed and the atomic process performed by the atomic unit on the business. The atomic unit is the basic execution unit that constitutes the business activity during the business execution process, and the atomic process is the basic business activity performed by the atomic unit during the business execution process; A business phenomenon pattern definition module is used to determine a business phenomenon pattern, where the business phenomenon pattern is an action pattern that induces the occurrence of a business phenomenon during business execution and is summarized based on atomic process data of atomic units; a calculation module, configured to calculate the support of the atomic unit for the business phenomenon pattern, and add labels to the atomic unit based on the business phenomenon pattern and the support, to obtain labeled evidence of the atomic unit; The summary analysis module is used to summarize the support of all atomic units for business phenomenon patterns based on the labeled evidence of the atomic units, and extract business phenomenon patterns with support greater than a preset threshold as attribution analysis results of the occurrence of business phenomena.

9. An electronic device, characterized in that: The system comprises a processor and a memory storing program instructions, wherein the processor is configured to execute a business attribution analysis method according to any one of claims 1 to 7 when executing the program instructions.

10. A computer-readable medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions are executed by a processor to implement a business attribution analysis method as described in any one of claims 1 to 7.

Images