Business data acquisition methods, devices, computer equipment, and storage media

By adjusting the initial weights and recombining them into the trained evaluation model, the evaluation data types with a greater impact were selected, which solved the problem of insufficient accuracy in business data collection and improved the efficiency of business data collection and evaluation.

CN115719185BActive Publication Date: 2026-11-10SHANGHAI PUDONG DEVELOPMENT BANK
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202211530621.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-01
Publication Date
2026-11-10
Estimated Expiration
2042-12-01

Smart Images

  • Figure CN115719185B_ABST
    Figure CN115719185B_ABST
Patent Text Reader

Abstract

This application relates to a business data acquisition method, apparatus, computer equipment, storage medium, and computer program product. The method includes: obtaining the initial weights set for each evaluation data type of the current object; determining an initial weight reassembly composed of the initial weights; adjusting the initial weights in the initial weight reassembly to obtain multiple reference weight reassemblies; determining the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reassembly relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reassembly relative to the corresponding initial evaluation score of the initial weight reassembly; and filtering all evaluation data types according to the degree of impact corresponding to each evaluation data type to determine the filtered target evaluation data type. This method can improve the efficiency of business data acquisition.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of big data technology, and in particular to a business data acquisition method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Artificial intelligence (AI) is a discipline that simulates human consciousness and thought processes. Machine learning, as the core technology of AI, has also made significant breakthroughs, endowing machines with powerful cognitive and predictive capabilities. Machine learning requires the use of large amounts of collected sample data for model training to facilitate later applications.

[0003] However, in some fields, such as the internet, digital finance, and e-commerce platforms, the data collection process may encounter problems such as insufficient accuracy of the collected data types or errors in data feature collection, leading to poor performance of the trained models in later applications. Taking banking business data as an example, this data reflects the bank's operational status, and different types of business data reflect this status to varying degrees. The quantity and variety of business data types also affect the efficiency of data collection. Therefore, there is an urgent need for a better method for business data collection. Summary of the Invention

[0004] Therefore, it is necessary to provide a business data acquisition method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve the accuracy of business data acquisition in response to the above-mentioned technical problems.

[0005] Firstly, this application provides a method for collecting business data. The method includes:

[0006] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0007] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0008] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0009] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0010] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0011] In one embodiment, the process of obtaining the trained evaluation model includes:

[0012] Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score;

[0013] Based on the differences between the predicted evaluation scores and the evaluation score labels of the sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

[0014] In one embodiment, the initial evaluation model includes multiple boosting models; the corresponding sample weights of the sample objects are input into the initial evaluation model, and the corresponding sample predicted evaluation scores are output, including:

[0015] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0016] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0017] In one embodiment, an initial weighting reorganization consisting of the initial weights is determined, and the initial weights in the initial weighting reorganization are adjusted to obtain multiple reference weighting reorganizations, including:

[0018] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0019] In one embodiment, the impact of weight changes on the business execution quality corresponding to the target business type is determined based on the adjustment magnitude of each reference weight reorganization relative to the initial weight group, and the change magnitude of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization. This includes:

[0020] For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0021] The change gradient value groups corresponding to each reference weight reorganization are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0022] In one embodiment, after recombining the reference weights and inputting them into the trained evaluation model, and outputting the corresponding reference evaluation scores, the method further includes:

[0023] Determine the maximum value among all reference evaluation scores, and then determine the target reference weight reorganization corresponding to the maximum value;

[0024] The initial weight reorganization is adjusted based on the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

[0025] Secondly, this application also provides a business data acquisition device. The device includes:

[0026] The first determination module is used to determine multiple evaluation data types for evaluating the execution quality of the target business type for the target business type to be executed by the current object, and to obtain the initial weights set by the current object for each evaluation data type.

[0027] The second determining module is used to determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations.

[0028] The output module is used to input the reorganized reference weights into the trained evaluation model and output the corresponding reference evaluation scores. The reference evaluation scores are used to characterize the degree of impact of the reorganized reference weights on the business execution quality corresponding to the target business type.

[0029] The third determining module is used to determine the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reorganization compared to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization compared to the initial evaluation score value of the initial weight reorganization.

[0030] The fourth determination module is used to filter all evaluation data types according to the degree of influence corresponding to each evaluation data type, determine the target evaluation data type after filtering, and collect the corresponding business data according to the target evaluation data type when the current object performs the target business type.

[0031] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0032] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0033] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0034] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0035] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0036] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0037] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0038] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0039] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0040] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0041] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0042] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0043] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0044] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0045] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0046] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0047] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0048] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0049] The aforementioned business data acquisition method, apparatus, computer equipment, storage medium, and computer program product, for the target business type to be executed by the current object, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type; determine an initial weight reassembly composed of each initial weight, adjust the initial weights in the initial weight reassembly to obtain multiple reference weight reassemblies; input each reference weight reassembly into the trained evaluation model, and output corresponding reference evaluation scores, which are used to characterize the degree of influence of the reference weight reassembly on the business execution quality corresponding to the target business type; determine the degree of influence of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reassembly compared to the initial weight group, and the change range of the corresponding reference score of each reference weight reassembly compared to the corresponding initial evaluation score of the initial weight reassembly; filter all evaluation data types according to the degree of influence corresponding to each evaluation data type, determine the filtered target evaluation data type, and collect the corresponding business data according to the target evaluation data type when the current object executes the target business type. This method can reduce the types of business data, thereby improving the efficiency of business data acquisition. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating a business data acquisition method in one embodiment;

[0051] Figure 2 This is a schematic diagram of the initial evaluation model in one embodiment;

[0052] Figure 3 This is a flowchart illustrating a business data acquisition method in one embodiment;

[0053] Figure 4 This is a structural block diagram of a service data acquisition device in one embodiment;

[0054] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0056] Currently, most banks rely on human experience to analyze business data and evaluate various business operations. However, this traditional evaluation method is time-consuming and has limitations. It cannot accurately reflect the business performance of a single branch across all dimensions, and its efficiency depends heavily on the technical proficiency and business knowledge of the developers. Therefore, evaluating and analyzing business data through traditional human experience results in low efficiency.

[0057] To address the problems existing in the aforementioned related technologies, such as Figure 1 As shown, this embodiment of the invention provides a business data collection method that can select evaluation data types that have a significant impact on evaluation results from various types of business data from banks or other companies. This reduces the number of evaluation data types used to evaluate the corresponding target business type, thereby improving the efficiency of evaluating the business execution quality of the target business type. This embodiment illustrates the method by applying it to a terminal, where the terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart vehicle-mounted devices, etc.

[0058] It is understood that this method can also be applied to servers, and to systems including terminals and servers, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0059] 101. For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0060] The current object can be an organization or company that generates business data across multiple business types, such as a bank. The target business type refers to the business type that affects the overall operational level of the current object. There must be at least one target business type. There must be at least two types of data being evaluated.

[0061] Each target business type has a unique business type number. For example, if there are three target business types: Target Business Type A, Target Business Type B, and Target Business Type C, the business type numbers for Target Business Type A, Target Business Type B, and Target Business Type C are IDX120001, IDX120002, and IDX120003, respectively.

[0062] Each target business type contains the same type of evaluation data. For example, the evaluation data types for the business execution quality corresponding to target business type IDX120001 are v1 and v2, respectively; the evaluation data types for the business execution quality corresponding to target business type IDX120002 should also be v1 and v2, respectively; and the evaluation data types for the business execution quality corresponding to target business type IDX120003 should also be v1 and v2, respectively. Here, data type v1 can represent the amount of interest-bearing business; data type v2 can represent the amount of business during the remaining term.

[0063] In one example, the current object can also be multiple types of images. The target business type refers to type A images and type B images. The various evaluation data types of business execution quality corresponding to type A images and type B images include image sharpness and image brightness.

[0064] For different target business types, the initial weights of the same evaluation data types corresponding to different target business types are the same. For example, if the evaluation data types for evaluating the business execution quality of target business class IDX120001 and target business class IDX120002 are v1 and v2 respectively, and the initial weight of v1 for IDX120001 is 70% and the initial weight of v2 for IDX120001 is 30%, then the initial weight of v1 for IDX120002 is 70% and the initial weight of v2 for IDX120002 is 30%.

[0065] 102. Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0066] In the initial weight reorganization, the number of weights is the same as the number of evaluation data types. For example, if there are four evaluation data types, v1, v2, v3, and v4, and the initial weights corresponding to v1, v2, v3, and v4 are 10%, 20%, 30%, and 40%, respectively, then the initial weight reorganization includes four weights of 10%, 20%, 30%, and 40%.

[0067] Adjusting the initial weights in the initial weight reorganization means that each initial weight in the initial weight reorganization can be adjusted upwards or downwards. Upward adjustment means increasing the value of the initial weight, and downward adjustment means decreasing the value of the initial weight.

[0068] 103. Input the reorganized reference weights into the trained evaluation model and output the corresponding reference evaluation scores. The reference evaluation scores are used to characterize the degree of impact of the reorganized reference weights on the business execution quality corresponding to the target business type.

[0069] The evaluation model is used to evaluate the appropriateness of the weight division of the corresponding weight group in the input. The higher the evaluation score of the corresponding weight group, the higher the appropriateness of the weight division. The lower the evaluation score of the corresponding weight group, the lower the appropriateness of the weight division.

[0070] 104. Based on the adjustment range of each reference weight reorganization compared to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization compared to the corresponding initial evaluation score of the initial weight reorganization, determine the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type.

[0071] The adjustment magnitude refers to the change in each reference weight in each reference weight reorganization relative to the corresponding initial weight in the initial weight group. For example, if the initial weight reorganization is {x, y, z}, and any reference weight reorganization is {x + Δx1, y + Δy1, z + Δz1}, then Δx1 is the adjustment magnitude of x, Δy1 is the adjustment magnitude of y, and Δz1 is the adjustment magnitude of z.

[0072] The initial evaluation score is the score corresponding to the initial weight reassembly. The initial evaluation score is the score corresponding to the initial weight reassembly output after the initial weight reassembly is input into the trained evaluation model.

[0073] 105. Filter all evaluation data types according to the degree of influence corresponding to each evaluation data type, determine the target evaluation data type after filtering, and collect the corresponding business data according to the target evaluation data type when the current object performs the target business type.

[0074] The target evaluation data type refers to data that has a significant impact on the quality of business execution. For example, the evaluation data types are v1, v2, v3, and v4. The weight changes of each evaluation data type on the quality of business execution corresponding to the target business type are arranged in order of their degree of impact as evaluation data types v3, v1, v4, and v2. If the selection criterion is to select the top 3 evaluation data types in terms of their degree of impact, then the selected target evaluation data types are v3, v1, and v4.

[0075] When the current object executes the target business corresponding to the target business type, after collecting the corresponding business data according to the target evaluation data type, the corresponding target business can be evaluated based on the corresponding business data.

[0076] The method described in this invention filters various evaluation data types for the business execution quality of all target business types of a bank, thereby obtaining target evaluation data types that have a significant impact on all target business types. Based on these target evaluation data types, the corresponding target businesses of each target business type are evaluated, reducing the amount of business data and consequently the number of evaluation data types used to evaluate the corresponding target business types. This improves the efficiency of evaluating the business execution quality of target business types and also enhances the efficiency of business data processing.

[0077] In conjunction with the above embodiments, in one embodiment, the process of obtaining the trained evaluation model includes:

[0078] Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score;

[0079] Based on the differences between the predicted evaluation scores and the evaluation score labels of the sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

[0080] The sample objects can be institutions or companies that collect business data across multiple business types, such as banks. The sample business type refers to the type of business that affects the overall operational level of the sample object. Each weighted sample corresponds to a sample business type of a sample object, and each weighted sample corresponds to an evaluation score label. The evaluation score label for each weighted sample can be a score obtained by analyzing and evaluating the corresponding weighted sample.

[0081] For example, consider sample objects A and B. Institution A's sample business types include A1 and A2, and Institution B's sample business types include B1 and B2. Each sample object's evaluation data types include v1, v2, v3, and v4. For Institution A's sample business type A1, the sample weight for evaluation data type v1 is 20%, the sample weight for evaluation data type v2 is 10%, the sample weight for evaluation data type v3 is 20%, and the sample weight for evaluation data type v4 is 50%. The evaluation score label for Institution A's sample business type A1 is 80. Similarly, for Institution A's sample business type A2, the sample weight for evaluation data type v1 is 15%, the sample weight for evaluation data type v2 is 15%, and the sample weight for evaluation data type v3 is 20%. The sample weight for data type v4 is 50%, and the evaluation score label corresponding to business type A2 of institution A is 85; the sample weight for evaluation data type v1 of institution B is 30%, the sample weight for evaluation data type v2 is 10%, the sample weight for evaluation data type v3 is 20%, and the sample weight for evaluation data type v4 is 40%, and the evaluation score label corresponding to business type B1 of institution B is 87; the sample weight for evaluation data type v1 of institution B is 25%, the sample weight for evaluation data type v2 is 5%, the sample weight for evaluation data type v3 is 35%, and the sample weight for evaluation data type v4 is 35%, and the evaluation score label corresponding to business type B2 of institution B is 83.

[0082] Specifically, the sample weights are input into the initial evaluation model, and the difference between the predicted evaluation score and the evaluation score label for each sample weight group is determined. The evaluation model training is complete when the average of the sum of squares of the differences between the predicted evaluation scores and the evaluation score labels for all sample weight groups is not greater than a preset value. For example, there are three sample weight groups: the predicted evaluation score and evaluation score label for the first group are Q and Q1, the predicted evaluation score and evaluation score label for the second group are P and P1, and the predicted evaluation score and evaluation score label for the third group are U and U1. The evaluation model training is complete when the value is not greater than a preset value. The size of the preset value can be adjusted.

[0083] The method provided in this invention can train the initial evaluation model by using the difference between the sample predicted evaluation score and the evaluation score label of each sample weight group, thereby improving the accuracy of the evaluation model prediction.

[0084] In conjunction with the above embodiments, in one embodiment, the initial evaluation model includes multiple boosting models; the corresponding sample weights of the sample objects are reorganized and input into the initial evaluation model, and the corresponding sample predicted evaluation scores are output, including:

[0085] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0086] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0087] The initial evaluation model contains at least two boosting models. Each boosting model consists of multiple base models.

[0088] Boosting models focus more on misclassified samples from the previous base model during each training iteration. When training the next base model, these misclassified samples are given greater weight. The purpose of this is to strengthen the learning of misclassified samples, so that the base models in the boosting model can iterate continuously, making the prediction results output by the boosting model increasingly more aligned with the evaluation score labels.

[0089] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding predicted evaluation score is output, including:

[0090] For each sample weight group, the average of the multiple pre-selected evaluation scores is taken, and this average is used as the predicted evaluation score for the corresponding sample weight group; or

[0091] For each sample weight group, multiple pre-selected evaluation scores are voted on, and the pre-selected evaluation score with the highest vote is used as the sample predicted evaluation score for the corresponding sample weight group.

[0092] Specifically, the training data for each boosting model is the same, and the training data for the base model of each boosting model is obtained from the training data of the previous base model. By transforming the training data of the previous base model, the training data for the current base model can be obtained.

[0093] Figure 2 This is a schematic diagram of the initial evaluation model in one embodiment, including three boosting models. During training, each boosting model outputs a pre-selected evaluation score for a sample. Figure 2 In this context, training data refers to the reweighting of samples.

[0094] The method provided in this invention can improve the accuracy of the output results of the trained evaluation model by training multiple boosting models, thereby improving the accuracy of the prediction results of the evaluation model.

[0095] In conjunction with the above embodiments, in one embodiment, an initial weighting reorganization consisting of each initial weight is determined, and the initial weights in the initial weighting reorganization are adjusted to obtain multiple reference weighting reorganizations, including:

[0096] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0097] In this initial weighting, each initial weight corresponds to a set of adjustment values. For example, if the initial weighting is {x, y, z}, then the initial weight x corresponds to a set of adjustment values, the initial weight y corresponds to a set of adjustment values, and the initial weight z corresponds to a set of adjustment values.

[0098] Taking an initial weighting reorganization that includes three initial weights as an example, the initial weighting reorganization is adjusted multiple times according to the adjustment magnitude to obtain multiple non-overlapping reference weighting reorganizations, including:

[0099] The initial weighted reorganization is: {x,y,z};

[0100] The adjusted reference weights are reorganized as follows:

[0101] {x+i1×d,y+i2×d,z+i3×d},i1∈Z,i2∈Z,i3∈Z,d>0; where i1×d is the adjustment range of the initial weight x; i2×d is the adjustment range of the initial weight y; and i3×d is the adjustment range of the initial weight z; each overall adjustment process adjusts at least one initial weight in the initial weight reorganization, and the adjustment range is positive or negative.

[0102] The method provided in this invention adjusts the initial weighting reorganization by adjusting the amplitude value, thereby obtaining multiple reference weighting reorganizations. In each overall adjustment process, at least one initial weight in the initial weighting reorganization is adjusted, so that each reference weighting reorganization obtained is different from the previous one, thereby avoiding duplicate reference weighting reorganizations and improving the efficiency of screening evaluation data types.

[0103] In conjunction with the above embodiments, in one embodiment, based on the adjustment magnitude of each reference weight reorganization relative to the initial weight group, and the change magnitude of the corresponding reference score value of each reference weight reorganization relative to the corresponding initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined, including:

[0104] For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0105] The change gradient value groups corresponding to each reorganization of reference weights are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0106] Taking the current reference weight reorganization, which includes three types of reference weights, as an example, for the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization, and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, including:

[0107] The current reference weight reorganization is {x2, y2, z2}, and the corresponding reference score is k2. The initial weight reorganization is {x, y, z}, and the initial evaluation score of the initial weight group is k, where |z-z2|=0, |x-x2|>0, and |y-y2|>0. Therefore, the adjustment magnitudes for the same weight types between the current and initial weight reorganizations are x-x2, y-y2, and 0, respectively; the change magnitude for the current reference weight reorganization is k-k2. Since the change magnitude for the weight type corresponding to z2 is 0, the change gradient for the weight type corresponding to z2 is 0; the change gradient for the weight type corresponding to x2 is... The gradient value of the weight type change corresponding to y2 is The set of change gradient values ​​corresponding to the current reference weight reorganization is as follows Among them, if This indicates that the weight type corresponding to x2 has a greater impact on the execution quality of the target business type than the weight type corresponding to y2. If This indicates that the influence of the weight type corresponding to x2 on the execution quality of the target business type is less than the influence of the weight type corresponding to y2 on the execution quality of the target business type. If This means that the influence of the weight type corresponding to x2 on the execution quality of the target business type is equal to the influence of the weight type corresponding to y2 on the execution quality of the target business type.

[0108] The gradient value sets corresponding to the reorganization of each reference weight are integrated to obtain the target gradient value set, including:

[0109] The average gradient values ​​for the same weight type are calculated to obtain the average gradient value for each weight type. Based on the average gradient value for each weight type, the target gradient value set is determined. For example, there are 5 reference weight reassemblies, each including three weight types. The gradient values ​​for the first weight type are {r1, r2, r3, r4, r5}, the gradient values ​​for the second weight type are {h1, h2, h3, h4, h5}, and the gradient values ​​for the third weight type are {g1, g2, g3, g4, g5}. Then the target gradient value set is {(r1+r2+r3+r4+r5) / 5, (h1+h2+h3+h4+h5) / 5, (g1+g2+g3+g4+g5) / 5}.

[0110] Specifically, after determining the target change gradient value set, all evaluation data types are filtered based on the magnitude of the values ​​in the target change gradient value set to determine the filtered target evaluation data types. For example, if each reference weight reorganization includes 5 weight types, 3 weight types need to be selected as target evaluation data types. If the corresponding target change gradient value set for each reference weight reorganization is {2, 7, 15, 6, 7, 11}, then the weight types corresponding to 15, 7, and 11 are selected as target evaluation data types, and the corresponding business data is collected according to the target evaluation data types.

[0111] The method provided in this invention can filter out the evaluation data types that have a greater impact on the business execution quality of the target business type from all evaluation data types based on the degree of influence of the weight changes of each evaluation data type on the business execution quality of the target business type. This reduces the amount of business data collected and improves the efficiency of data collection.

[0112] In conjunction with the above embodiments, in one embodiment, after recombining the reference weights and inputting them into the trained evaluation model, and outputting the corresponding reference evaluation scores, the method further includes:

[0113] Determine the maximum value among all reference evaluation scores, and then determine the target reference weight reorganization corresponding to the maximum value;

[0114] The initial weight reorganization is adjusted based on the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

[0115] Specifically, after obtaining the trained evaluation model, each reference weight reorganization can be input into the trained evaluation model to obtain the corresponding reference evaluation score for each reference weight reorganization. From all the reference evaluation scores, the reference weight reorganization corresponding to the highest reference evaluation score is determined as the target reference weight reorganization. The initial weight reorganization is adjusted according to the degree of difference between the target reference weight reorganization and the initial weight reorganization for the same weight type, and the business execution quality is evaluated according to the adjusted initial weight reorganization. For example, if there are 3 sets of reference weight reorganizations, with the first reference weight group having a reference evaluation score of 80, the second reference weight group having a reference evaluation score of 85, and the third reference weight group having a reference evaluation score of 87, then the third reference weight reorganization is the target reference weight reorganization.

[0116] The initial weight reorganization is adjusted according to the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, including:

[0117] Taking the target reference weight reorganization, which includes three weight types, as an example, the target reference weight reorganization is (x m y m , z m If the initial weights are (x, y, z), then the initial weights will be adjusted to (x, y, z). m y m , z m The adjusted initial weighting is obtained, and the business execution quality is evaluated for the business data of various evaluation data types of the target business type of the current object according to the adjusted initial weighting.

[0118] The method provided in this invention determines the maximum value among all reference evaluation scores, selects the optimal reference weight reorganization as the target reference weight reorganization from each reference weight reorganization, adjusts the weights of each weight type in the initial weight group through the target reference weight reorganization, and evaluates the business execution quality according to the adjusted initial weight reorganization, which can improve the accuracy of the evaluation results.

[0119] In one embodiment, such as Figure 3 As shown, a business data collection method, applied to a bank's business optimization guidance system, includes:

[0120] 301. For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0121] 302. Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0122] 303. Input the reorganized reference weights into the trained evaluation model and output the corresponding reference evaluation scores. The reference evaluation scores are used to characterize the degree of impact of the reorganized reference weights on the business execution quality corresponding to the target business type.

[0123] 304. Determine the maximum value among all reference evaluation scores, and then reorganize the target reference weights corresponding to the maximum value;

[0124] 305. Adjust the initial weight reorganization according to the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and evaluate the business execution quality according to the adjusted initial weight reorganization.

[0125] The method provided in this invention adjusts the initial weights of each evaluation data type, thereby increasing the weights of evaluation data types that have a greater impact on the business execution quality corresponding to the target business type, thus making the evaluation scores obtained based on the evaluation model more accurate.

[0126] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0127] Based on the same inventive concept, this application also provides a business data acquisition device for implementing the business data acquisition method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more business data acquisition device embodiments provided below can be found in the limitations of the business data acquisition method described above, and will not be repeated here.

[0128] In one embodiment, such as Figure 4As shown, a business data acquisition device is provided, including: a first determining module 401, a second determining module 402, an output module 403, a third determining module 404, and a fourth determining module 405, wherein:

[0129] The first determining module 401 is used to determine multiple evaluation data types for evaluating the execution quality of the target business type for the target business type to be executed by the current object, and to obtain the initial weights set by the current object for each evaluation data type.

[0130] The second determining module 402 is used to determine the initial weight reorganization composed of each initial weight, adjust the initial weight in the initial weight reorganization, and obtain multiple reference weight reorganizations.

[0131] The output module 403 is used to input the reorganized reference weights into the trained evaluation model and output the corresponding reference evaluation scores. The reference evaluation scores are used to characterize the degree of impact of the reorganized reference weights on the business execution quality corresponding to the target business type.

[0132] The third determining module 404 is used to determine the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reorganization compared to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization compared to the initial evaluation score value of the initial weight reorganization.

[0133] The fourth determination module 405 is used to filter all evaluation data types according to the degree of influence corresponding to each evaluation data type, determine the target evaluation data type after filtering, and collect the corresponding business data according to the target evaluation data type when the current object performs the target business type corresponding to the target business.

[0134] In one embodiment, the output module 403 includes:

[0135] The first acquisition submodule is used to acquire the sample weight reorganization set for each evaluation data type of multiple sample objects, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score.

[0136] The training submodule is used to train the initial evaluation model based on the difference between the corresponding sample prediction evaluation scores and evaluation score labels of the sample objects, and obtain the trained evaluation model.

[0137] In one embodiment, obtaining a submodule includes:

[0138] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0139] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0140] In one embodiment, the second determining module 402 includes:

[0141] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0142] In one embodiment, the third determining module 404 includes:

[0143] The second acquisition submodule is used to acquire the change gradient value corresponding to each weight type based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, and to form a change gradient value group corresponding to the current reference weight reorganization. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0144] The integration submodule is used to integrate the change gradient value groups corresponding to the reorganization of each reference weight to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0145] In one embodiment, the output module 403 further includes:

[0146] The determination submodule is used to determine the maximum value among all reference evaluation scores and to determine the target reference weight reorganization corresponding to the maximum value.

[0147] The adjustment submodule is used to adjust the initial weight reorganization according to the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and to evaluate the business execution quality according to the adjusted initial weight reorganization.

[0148] Each module in the aforementioned business data acquisition device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0149] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a business data acquisition method. The display unit 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 an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0150] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0151] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0152] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0153] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0154] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0155] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0156] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0157] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0158] Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score;

[0159] Based on the differences between the predicted evaluation scores and the evaluation score labels of the sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

[0160] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0161] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0162] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0163] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0164] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0165] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0166] For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0167] The change gradient value groups corresponding to each reorganization of reference weights are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0168] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0169] Determine the maximum value among all reference evaluation scores, and then determine the target reference weight reorganization corresponding to the maximum value;

[0170] The initial weight reorganization is adjusted based on the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

[0171] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0172] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0173] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0174] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0175] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0176] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0177] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0178] Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score;

[0179] Based on the differences between the predicted evaluation scores and the evaluation score labels of the sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

[0180] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0181] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0182] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0183] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0184] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0185] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0186] For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0187] The change gradient value groups corresponding to each reorganization of reference weights are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0188] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0189] Determine the maximum value among all reference evaluation scores, and then determine the target reference weight reorganization corresponding to the maximum value;

[0190] The initial weight reorganization is adjusted based on the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

[0191] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0192] For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type;

[0193] Determine the initial weight reorganization composed of each initial weight, adjust the initial weights in the initial weight reorganization, and obtain multiple reference weight reorganizations;

[0194] Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type.

[0195] Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type is determined.

[0196] All evaluation data types are filtered according to their degree of influence to determine the target evaluation data type. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type.

[0197] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0198] Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score;

[0199] Based on the differences between the predicted evaluation scores and the evaluation score labels of the sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

[0200] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0201] The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects;

[0202] Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

[0203] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0204] Obtain the adjustment range value, and perform multiple overall adjustments on the initial weight reorganization according to the adjustment range value to obtain multiple non-repeating reference weight reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weight reorganization is adjusted, and the adjustment range value is positive or negative.

[0205] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0206] For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type changes in the current reference weight reorganization.

[0207] The change gradient value groups corresponding to each reorganization of reference weights are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

[0208] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0209] Determine the maximum value among all reference evaluation scores, and then determine the target reference weight reorganization corresponding to the maximum value;

[0210] The initial weight reorganization is adjusted based on the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

[0211] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) 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 the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0212] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0213] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0214] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for collecting business data, characterized in that, The method includes: For the target business type that the current object needs to execute, determine multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and obtain the initial weights set by the current object for each evaluation data type; Determine an initial weighting reorganization consisting of each initial weight, adjust the initial weights in the initial weighting reorganization, and obtain multiple reference weighting reorganizations; Each reference weight is reorganized and input into the trained evaluation model, and the corresponding reference evaluation score is output. The reference evaluation score is used to characterize the degree of impact of the reorganization of reference weights on the business execution quality corresponding to the target business type. Based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, the degree of impact of the weight change of each evaluation data type on the business execution quality corresponding to the target business type is determined. All evaluation data types are filtered according to the degree of influence corresponding to each evaluation data type to determine the target evaluation data type after filtering. When the current object performs the target business corresponding to the target business type, the corresponding business data is collected according to the target evaluation data type. The step of determining the impact of weight changes on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reorganization relative to the initial weight group, and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score of the initial weight reorganization, includes: For the current reference weight reorganization, based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, the change gradient value corresponding to each weight type is obtained, and a change gradient value group corresponding to the current reference weight reorganization is formed. The change gradient value is used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type in the current reference weight reorganization changes. The change gradient value groups corresponding to each reference weight reorganization are integrated to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

2. The method according to claim 1, characterized in that, The process of obtaining the trained evaluation model includes: Obtain the sample weight reorganization of multiple sample objects for each evaluation data type, input the corresponding sample weight reorganization of the sample objects into the initial evaluation model, and output the corresponding sample prediction evaluation score; Based on the differences between the predicted evaluation scores and evaluation score labels of the corresponding sample objects, the initial evaluation model is trained to obtain the trained evaluation model.

3. The method according to claim 2, characterized in that, The initial evaluation model includes multiple boosting models; the step of inputting the corresponding sample weights of the sample objects into the initial evaluation model and outputting the corresponding sample predicted evaluation scores includes: The corresponding sample weights of the sample objects are input into each boosting model to obtain multiple sample pre-selection evaluation scores for the corresponding sample weight groups of the sample objects; Based on the pre-selected evaluation scores of multiple samples in the corresponding sample weight group of the sample object, the corresponding sample predicted evaluation score is output.

4. The method according to claim 1, characterized in that, The process of determining an initial weighting reorganization composed of initial weights, adjusting the initial weights in the initial weighting reorganization to obtain multiple reference weighting reorganizations includes: Obtain the adjustment range value, and perform multiple overall adjustments on the initial weighting reorganization according to the adjustment range value to obtain multiple non-repeating reference weighting reorganizations; wherein, in each overall adjustment process, at least one initial weight in the initial weighting reorganization is adjusted, and the adjustment range value is positive or negative.

5. The method according to claim 1, characterized in that, After recombining the reference weights and inputting them into the trained evaluation model to output the corresponding reference evaluation scores, the process also includes: Determine the maximum value among all reference evaluation scores, and determine the target reference weight reorganization corresponding to the maximum value; The initial weight reorganization is adjusted according to the degree of weight difference of the same weight type in the target reference weight reorganization and the initial weight reorganization, and the business execution quality is evaluated according to the adjusted initial weight reorganization.

6. A business data acquisition device, characterized in that, The device includes: The first determining module is used to determine, for the target business type to be executed by the current object, multiple evaluation data types for evaluating the business execution quality corresponding to the target business type, and to obtain the initial weights set by the current object for each evaluation data type; The second determining module is used to determine the initial weight reorganization composed of each initial weight, adjust the initial weight in the initial weight reorganization, and obtain multiple reference weight reorganizations. The output module is used to input the reorganized reference weights into the trained evaluation model and output the corresponding reference evaluation scores. The reference evaluation scores are used to characterize the degree of impact of the reorganized reference weights on the business execution quality corresponding to the target business type. The third determining module is used to determine the degree of impact of the weight changes of each evaluation data type on the business execution quality corresponding to the target business type based on the adjustment range of each reference weight reorganization relative to the initial weight group and the change range of the corresponding reference score value of each reference weight reorganization relative to the initial evaluation score value of the initial weight reorganization. The fourth determination module is used to filter all evaluation data types according to the degree of influence corresponding to each evaluation data type, determine the target evaluation data type after filtering, and collect the corresponding business data according to the target evaluation data type when the current object performs the target business type. The third determining module includes a second obtaining submodule, configured to, for the current reference weight reorganization, obtain change gradient values ​​corresponding to each weight type based on the adjustment magnitude value corresponding to the same weight type between the current reference weight reorganization and the initial weight reorganization and the change magnitude value corresponding to the current reference weight reorganization, and form a change gradient value group corresponding to the current reference weight reorganization. The change gradient values ​​are used to characterize the degree of impact on the business execution quality corresponding to the target business type when the corresponding weight type in the current reference weight reorganization changes. The integration submodule is used to integrate the change gradient value groups corresponding to each reference weight reorganization to obtain the target change gradient value group. Each change gradient value in the target change gradient value group is used to characterize the degree of influence of the corresponding weight type on the business execution quality corresponding to the target business type.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Human body falling behavior evaluation method and device, computer equipment and storage medium

    CN114494976A

  • Service quality evaluation method and device

    CN114693034A

  • Real estate assessment method and device

    CN115099860A