CN115719185A

CN115719185A - Business data acquisition method and device, computer equipment and storage medium

Info

Publication number: CN115719185A
Application number: CN202211530621.1A
Authority: CN
Inventors: 魏宇帆; 王尧
Original assignee: Shanghai Pudong Development Bank Co Ltd
Current assignee: Shanghai Pudong Development Bank Co Ltd
Priority date: 2022-12-01
Filing date: 2022-12-01
Publication date: 2023-02-28

Abstract

The application relates to a business data acquisition method, a business data acquisition device, a computer device, a storage medium and a computer program product. The method comprises the following steps: acquiring initial weights set by a current object for various evaluation data types; determining an initial weight set formed by all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets; determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group; and screening all the evaluation data types according to the influence degrees corresponding to all the evaluation data types, and determining the screened target evaluation data types. The method can improve the efficiency of business data acquisition.

Description

Business data acquisition method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of big data technologies, and in particular, to a method and an apparatus for acquiring service data, a computer device, and a storage medium.

Background

Artificial intelligence is a discipline that simulates human consciousness and thinking. Machine learning has also made a significant breakthrough as the core of artificial intelligence, and machines are endowed with powerful cognitive and predictive capabilities. Machine learning requires model training using a large amount of collected sample data to facilitate later-stage application.

However, in some fields, such as the internet, digital finance, e-commerce platforms, etc., in the process of data acquisition, the later application effect of the trained model may be poor due to factors such as insufficient accuracy of the acquired data type or acquisition errors of data characteristics. Taking the business data of the bank as an example, the business data of the bank can reflect the operation condition of the bank, and the different types of business data have different degrees of reaction to the operation condition of the bank. How many types of business data affect the collection efficiency of business data. Therefore, a method for collecting service data is urgently needed.

Disclosure of Invention

In view of the above, it is necessary to provide a business data collecting method, a business data collecting apparatus, a computer device, a computer readable storage medium, and a computer program product, which can improve the accuracy of business data collection.

In a first aspect, the present application provides a service data acquisition method. The method comprises the following steps:

aiming at a target service type required to be executed by a current object, determining a plurality of evaluation data types used for evaluating service execution quality corresponding to the target service type, and acquiring initial weights set by the current object aiming at the evaluation data types;

determining an initial weight set formed by all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

inputting each reference weight group into the trained evaluation model, and outputting a corresponding reference evaluation score, wherein the reference evaluation score is used for representing the influence degree of the reference weight group on the service execution quality corresponding to the target service type;

determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group;

and screening all the evaluation data types according to the influence degrees corresponding to all the evaluation data types, determining the screened target evaluation data types, and acquiring corresponding service data according to the target evaluation data types when the current object executes the target service corresponding to the target service types.

In one embodiment, the acquisition process of the trained evaluation model includes:

acquiring sample weight groups set by a plurality of sample objects aiming at each evaluation data type, inputting the sample weight groups corresponding to the sample objects into an initial evaluation model, and outputting corresponding sample prediction evaluation scores;

and training the initial evaluation model according to the difference between the corresponding sample prediction evaluation value of the sample object and the evaluation value label to obtain the trained evaluation model.

In one embodiment, the initial evaluation model comprises a plurality of boosting models; inputting the corresponding sample weight set of the sample object into the initial evaluation model, and outputting a corresponding sample prediction evaluation score, wherein the method comprises the following steps:

inputting the corresponding sample weight group of the sample object into each boosting model to obtain a plurality of sample pre-selection evaluation scores of the corresponding sample weight group of the sample object;

and outputting corresponding sample prediction evaluation scores according to a plurality of sample pre-selection evaluation scores of the sample weight groups corresponding to the sample objects.

In one embodiment, determining an initial weight set composed of initial weights, adjusting the initial weights in the initial weight set, and obtaining a plurality of reference weight sets includes:

acquiring an adjustment amplitude value, and performing multiple integral adjustments on the initial weight set according to the adjustment amplitude value to obtain multiple non-repetitive reference weight sets; and adjusting at least one initial weight in the initial weight set in each integral adjustment process, wherein the amplitude value is adjusted to be a positive value or a negative value.

In one embodiment, determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group includes:

aiming at the current reference weight group, acquiring a change gradient value corresponding to each weight type according to an adjustment amplitude value corresponding to the same weight type between the current reference weight group and the initial weight group and a change amplitude value corresponding to the current reference weight group, and forming a change gradient value group corresponding to the current reference weight group, wherein the change gradient value is used for representing the influence degree of the corresponding weight type in the current reference weight group on the service execution quality corresponding to the target service type when the corresponding weight type is changed;

and integrating the change gradient value groups corresponding to the reference weight groups to obtain a target change gradient value group, wherein each change gradient value in the target change gradient value group is used for representing the influence degree of the corresponding weight type on the service execution quality corresponding to the target service type.

In one embodiment, after inputting each reference weight set into the trained evaluation model and outputting a corresponding reference evaluation score, the method further includes:

determining a maximum value in all the reference evaluation scores, and determining a target reference weight set corresponding to the maximum value;

and adjusting the initial weight group according to the weight difference degree of the same weight type in the target reference weight group and the initial weight group, and evaluating the service execution quality according to the adjusted initial weight group.

In a second aspect, the application further provides a service data acquisition device. The device comprises:

the first determining module is used for determining multiple evaluation data types used for evaluating the service execution quality corresponding to the target service type aiming at the target service type required to be executed by the current object and acquiring the initial weight set by the current object aiming at each evaluation data type;

the second determining module is used for determining an initial weight set formed by the initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

the output module is used for inputting each reference weight group into the trained evaluation model and outputting a corresponding reference evaluation score, and the reference evaluation score is used for representing the influence degree of the reference weight group on the service execution quality corresponding to the target service type;

the third determining module is used for determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group;

and the fourth determining module is used for screening all the evaluation data types according to the influence degrees corresponding to all the evaluation data types, determining the screened target evaluation data types, and acquiring corresponding service data according to the target evaluation data types when the current object executes the target service corresponding to the target service types.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the following steps when executing the computer program:

In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of:

In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program which when executed by a processor performs the steps of:

determining an initial weight set composed of all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

According to the service data acquisition method, the service data acquisition device, the computer equipment, the storage medium and the computer program product, aiming at a target service type required to be executed by a current object, multiple evaluation data types used for evaluating service execution quality corresponding to the target service type are determined, and initial weights set by the current object aiming at the evaluation data types are obtained; determining an initial weight set composed of all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets; inputting each reference weight group into the trained evaluation model, and outputting a corresponding reference evaluation score, wherein the reference evaluation score is used for representing the influence degree of the reference weight group on the service execution quality corresponding to the target service type; determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group; and screening all the evaluation data types according to the influence degrees corresponding to all the evaluation data types, determining the screened target evaluation data types, and acquiring corresponding service data according to the target evaluation data types when the current object executes the target service corresponding to the target service types. By the method, the types of the service data can be reduced, so that the efficiency of acquiring the service data is improved.

Drawings

FIG. 1 is a flow chart illustrating a method for collecting service data according to an embodiment;

FIG. 2 is a schematic diagram of an initial evaluation model in one embodiment;

FIG. 3 is a flow chart illustrating a method for collecting service data according to an embodiment;

FIG. 4 is a block diagram of a service data acquisition device according to an embodiment;

FIG. 5 is a diagram of the internal structure of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.

At present, most banks analyze business data of the banks through manual experience so as to evaluate each business, however, the traditional evaluation mode consumes a lot of time, has certain limitations, cannot show the business level of any branch in a certain dimension, and also depends mainly on the technical proficiency of developers and the understanding of business knowledge. Therefore, the service data is evaluated and analyzed through traditional manual experience, so that the evaluation efficiency is low.

For the problems in the related art, as shown in fig. 1, an embodiment of the present invention provides a service data acquisition method, which can select an evaluation data type having a large influence on an evaluation result from multiple types of service data of a bank or other companies, thereby reducing the number of evaluation data types used for evaluating a corresponding target service type, and further improving the efficiency of evaluating the service execution quality of a target service type service. The embodiment is exemplified by applying the method to a terminal, wherein the terminal can be but is not limited to various personal computers, notebook computers, smart phones, tablet computers, internet of things devices and portable wearable devices, and the internet of things devices can be smart vehicle-mounted devices and the like.

It is understood that the method can also be applied to a server, and can also be applied to a system comprising a terminal and a server, and is implemented through interaction of the terminal and the server. In this embodiment, the method includes the steps of:

101. aiming at a target service type required to be executed by a current object, determining a plurality of evaluation data types used for evaluating service execution quality corresponding to the target service type, and acquiring initial weights set by the current object aiming at the evaluation data types;

the current object may be an organization or a company, such as a bank, that includes business data of multiple business types. The target traffic type refers to a traffic type that affects the overall operation level of the current object. The number of types of the target service type is at least one. The number of types of evaluation data types is at least two.

Each target traffic type has a unique traffic type number. For example, the target service type is three service types, namely a target service type a, a target service type B, and a target service type C, and the service type numbers of the target service type a, the target service type B, and the target service type C are IDX120001, IDX120002, and IDX120003, respectively.

Each target service type comprises evaluation data of the same type, for example, the evaluation data types of the service execution quality corresponding to the evaluation target service type IDX120001 are v1 and v2 respectively; the evaluation data types of the service execution quality corresponding to the target service type IDX120002 should be v1 and v2, respectively; the evaluation data types of the service execution quality corresponding to the target service type IDX120003 should also be v1 and v2, respectively. Wherein, the data type v1 may be interest-bearing traffic; the data type v2 may be lifetime traffic.

In one example, the current object may also be a plurality of types of images, the target service type refers to an a type image and a B type image, and the plurality of evaluation data types of service execution quality corresponding to the a type image and the B type image include the definition of the image and the brightness of the image.

Aiming at different target service types, the initial weights of the same evaluation data types corresponding to the different target service types are the same. For example, the evaluation data types for evaluating the service performance quality corresponding to the target service class IDX120001 and the target service class IDX120002 are v1 and v2, respectively, where if the initial weight of v1 of IDX120001 is 70% and the initial weight of v2 of IDX120001 is 30%, the initial weight of v1 of IDX120002 is 70% and the initial weight of v2 of IDX120002 is 30%.

102. Determining an initial weight set formed by all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

the number of weights in the initial weight set is the same as the number of evaluation data types, for example, the evaluation data types are respectively four types v1, v2, v3, and v4, and the initial weights corresponding to v1, v2, v3, and v4 are respectively 10%, 20%, 30%, and 40%. The initial set of weights comprises four weights of 10%, 20%, 30%, 40%.

Adjusting the initial weights in the set of initial weights means that each of the initial weights in the set of initial weights can be adjusted upward or downward, wherein upward adjustment means increasing the value of the initial weight and downward adjustment means decreasing the value of the initial weight.

103. Inputting each reference weight group into the trained evaluation model, and outputting a corresponding reference evaluation score, wherein the reference evaluation score is used for representing the influence degree of the reference weight group on the service execution quality corresponding to the target service type;

the evaluation model is used for evaluating the weight division appropriateness of the input corresponding weight set, if the evaluation score of the corresponding weight set is higher, the weight division appropriateness of the corresponding weight set is higher, and if the evaluation score of the corresponding weight set is lower, the weight division appropriateness of the corresponding weight set is lower.

104. Determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjustment amplitude value of each reference weight group compared with the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared with the initial evaluation score value corresponding to the initial weight group;

wherein, the adjusted amplitude value refers to a variation value of each reference weight group relative to a corresponding initial weight in the initial weight group. For example, the initial weight set is: { x, y, z }; any reference weight set is { x + Δ x ₁ ,y+Δy ₁ ,z+Δz ₁ }; then Δ x ₁ Adjusting amplitude value, Δ y, for x ₁ Adjusting the amplitude value, Δ z, for y ₁ The amplitude value is adjusted for z.

The initial evaluation score is a score corresponding to the initial weight group, and the initial evaluation score is a score corresponding to the initial weight group output after the initial weight group is input to the trained evaluation model.

105. And screening all the evaluation data types according to the influence degrees corresponding to the evaluation data types, determining the screened target evaluation data types, and acquiring corresponding service data according to the target evaluation data types when the current object executes the target service corresponding to the target service types.

The target evaluation data type refers to data having a deep influence on the service execution quality. For example, the evaluation data types are v1, v2, v3, and v4; the weight change of each evaluation data type is respectively an evaluation data type v3, v1, v4 and v2 which are sequentially arranged according to the depth of the influence degree of the service execution quality corresponding to the target service type, and if the screening standard is to select the evaluation data type with the influence degree ranked at the top 3, the screened target evaluation data type is respectively v3, v1 and v4.

When the current object executes the target service corresponding to the target service type, after corresponding service data are collected according to the target evaluation data type, the corresponding target service can be evaluated according to the corresponding service data.

According to the method provided by the embodiment of the invention, the target evaluation data type with a deep influence degree on all target service types of the bank can be obtained by screening the multiple evaluation data types of the service execution quality corresponding to all the target service types of the bank, so that the target service corresponding to the target service type is evaluated according to the target evaluation data type, the number of service data is reduced, the number of evaluation data types for evaluating the corresponding target service types is reduced, the evaluation efficiency of the service execution quality of the target service type service is improved, and the processing efficiency of the service data is improved.

With reference to the content of the foregoing embodiment, in an embodiment, the obtaining process of the trained evaluation model includes:

The sample object may be an organization or a company, such as a bank, including business data of a plurality of business types, and the sample business type refers to a business type affecting the overall operation level of the sample object. Each sample weight group corresponds to a sample business type of a sample object, and each sample weight group corresponds to an evaluation score label. The evaluation score label corresponding to each sample weight group may be a score obtained by analyzing and evaluating the corresponding sample weight group.

For example, the sample objects are organization a and organization B, the sample service type of the organization a includes A1 and A2, and the sample service type of the organization B includes B1 and B2; the evaluation data type of each sample object comprises v1, v2, v3 and v4; the sample weight of the evaluation data type v1 corresponding to the sample service type A1 of the organization A is 20%, the sample weight of the evaluation data type v2 is 10%, the sample weight of the evaluation data type v3 is 20%, the sample weight of the evaluation data type v4 is 50%, and the evaluation score label corresponding to the sample service type A1 of the organization A is 80%; the sample weight of the evaluation data type v1 corresponding to the sample business type A2 of the organization A is 15%, the sample weight of the evaluation data type v2 is 15%, the sample weight of the evaluation data type v3 is 20%, the sample weight of the evaluation data type v4 is 50%, and the evaluation score label corresponding to the sample business type A2 of the organization A is 85%; the sample weight of the evaluation data type v1 corresponding to the sample service type B1 of the organization B is 30%, the sample weight of the evaluation data type v2 is 10%, the sample weight of the evaluation data type v3 is 20%, the sample weight of the evaluation data type v4 is 40%, and the evaluation score label corresponding to the sample service type B1 of the organization B is 87; the sample weight of the evaluation data type v1 corresponding to the sample business type B2 of the organization B is 25%, the sample weight of the evaluation data type v2 is 5%, the sample weight of the evaluation data type v3 is 35%, the sample weight of the evaluation data type v4 is 35%, and the evaluation score label corresponding to the sample business type B2 of the organization B is 83.

Specifically, the sample weight sets are input into the initial evaluation model, the difference between the sample predicted evaluation score and the evaluation score label of each sample weight set is determined, and when the average value of the sum of squares of the differences between the sample predicted evaluation scores and the evaluation score labels of all the sample weight sets is not more than a preset value, the evaluation model training is completed. For example, the set of sample weights has 3 sets, where the set of sample predicted evaluation scores and evaluation score of the first set of sample weights are labeledLabels of the group samples are Q and Q1 respectively, labels of the group sample predicted evaluation score and the evaluation score of the second group sample weight are P and P1 respectively, labels of the group sample predicted evaluation score and the evaluation score of the third group sample weight are U and U1 respectively, when the group sample predicted evaluation score and the evaluation score of the third group sample weight are Q and Q1 respectively

And when the evaluation model is not more than the preset value, finishing the evaluation model training. Wherein, the size of the preset value can be adjusted.

According to the method provided by the embodiment of the invention, the initial evaluation model can be trained through the difference between the sample prediction evaluation value of each sample weight group and the evaluation value label, so that the accuracy of the evaluation model prediction is improved.

In combination with the above embodiments, in one embodiment, the initial evaluation model includes a plurality of boosting models; inputting the corresponding sample weight set of the sample object into the initial evaluation model, and outputting a corresponding sample prediction evaluation score, wherein the method comprises the following steps:

and outputting corresponding sample prediction evaluation scores according to the plurality of sample pre-selection evaluation scores of the corresponding sample weight groups of the sample objects.

Wherein, the initial evaluation model comprises at least two boosting models. Each boosting model is composed of a plurality of base models.

The boosting model refers to that misjudged sample data output by the last basic model is concerned more during each training, and the misjudged sample data can be given more weight when the next basic model is trained, so that the purpose of reinforcing the learning of misjudged samples is to lead the basic model in the boosting model to be more inclined to evaluate the score label more and more through continuous iteration.

Outputting a corresponding sample predicted rating score based on a plurality of sample pre-selected rating scores for a corresponding set of sample weights for the sample object, comprising:

averaging a plurality of sample preselected evaluation scores of each sample weight set, and taking the obtained average as a sample predicted evaluation score of the corresponding sample weight set; or

And voting a plurality of sample preselected evaluation scores of each sample weight group, and taking the preselected evaluation score with the highest voting score as the sample predicted evaluation score of the corresponding sample weight group.

Specifically, the training data of each boosting model is the same, and the training data of the base model of each boosting model is obtained according to the training data of the previous base model. The training data of the current base model can be obtained by converting the training data of the previous base model.

Fig. 2 is a schematic diagram of an initial evaluation model in an embodiment, which includes 3 boosting models, each of which outputs a sample pre-selected evaluation score in a training process, and in fig. 2, training data refers to a sample weight set.

According to the method provided by the embodiment of the invention, the accuracy of the output result of the trained evaluation model can be improved by training the plurality of boosting models, so that the accuracy of the prediction result of the evaluation model is improved.

With reference to the content of the foregoing embodiments, in an embodiment, determining an initial weight set composed of initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets, includes:

acquiring an adjustment amplitude value, and performing multiple times of overall adjustment on the initial weight set according to the adjustment amplitude value to obtain multiple non-repetitive reference weight sets; and adjusting at least one initial weight in the initial weight set in each integral adjustment process, wherein the amplitude value is adjusted to be a positive value or a negative value.

Each initial weight in the initial weight set corresponds to a group of adjusted amplitude values. For example, if the initial weight set is { x, y, z }, the initial weight x corresponds to a set of adjusted amplitude values, the initial weight y corresponds to a set of adjusted amplitude values, and the initial weight z corresponds to a set of adjusted amplitude values.

Taking the example that the initial weight set comprises three initial weights, performing multiple integral adjustments on the initial weight set according to the adjustment amplitude value to obtain multiple non-repeated reference weight sets, including:

the initial weight set is: { x, y, z };

the reference weight set obtained after adjustment is as follows:

{x+i ₁ ×d,y+i ₂ ×d,z+i ₃ ×d}，i ₁ ∈Z,i ₂ ∈Z,i ₃ ∈Z,d>0; wherein i ₁ Xd is the adjusted amplitude value of the initial weight x; i.e. i ₂ Xd is the adjusted amplitude value of the initial weight y; i.e. i ₃ Xd is the adjusted amplitude value of the initial weight z; and adjusting at least one initial weight in the initial weight set in each integral adjusting process, wherein the amplitude value is adjusted to be a positive value or a negative value.

According to the method provided by the embodiment of the invention, the initial weight group is integrally adjusted through adjusting the amplitude value, a plurality of reference weight groups can be obtained, at least one initial weight in the initial weight group is adjusted in each integral adjustment process, and the reference weight group obtained each time is different from the reference weight group obtained before, so that repeated reference weight groups are avoided being obtained, and the efficiency of screening and evaluating data types is improved.

With reference to the content of the foregoing embodiment, in an embodiment, determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjusted amplitude value of each reference weight group compared to the initial weight group and the change amplitude value of the reference score value corresponding to each reference weight group compared to the initial evaluation score value corresponding to the initial weight group includes:

Taking the case that the current reference weight set includes three reference weight types, for the current reference weight set, obtaining a variation gradient value corresponding to each weight type according to an adjusted amplitude value corresponding to the same weight type between the current reference weight set and the initial weight set and a variation amplitude value corresponding to the current reference weight set, includes:

the current reference weight set is { x2, y2, z2}, and the corresponding reference score value of the current reference weight set is k2; the initial weight set is { x, y, z }, the initial evaluation score of the initial weight set is k, wherein | z-z2| =0, | x-x2| = z>0，|y-y2|>0. The sum of the adjusted amplitude values corresponding to the same weight type between the current reference weight group and the initial weight group is x-x2, y-y2 and 0 respectively; the change amplitude value corresponding to the current reference weight set is k-k2. Since the value of the variation amplitude of the weight type corresponding to z2 is 0, the value of the variation gradient of the weight type corresponding to z2 is 0; x2 is a variable gradient value of the weight type

y2 is a variable gradient value of the weight type

The set of gradient values corresponding to the current set of reference weights is set as

Wherein, if

It is indicated that the degree of influence of the weight type corresponding to x2 on the service execution quality corresponding to the target service type is greater than the degree of influence of the weight type corresponding to y2 on the service execution quality corresponding to the target service type. If it is

It is indicated that the degree of influence of the weight type corresponding to x2 on the service execution quality corresponding to the target service type is smaller than the degree of influence of the weight type corresponding to y2 on the service execution quality corresponding to the target service type. If it is

It means that the influence degree of the weight type corresponding to x2 on the service execution quality corresponding to the target service type is equal to the influence degree of the weight type corresponding to y2 on the service execution quality corresponding to the target service type.

Integrating the change gradient value set corresponding to each reference weight set to obtain a target change gradient value set, wherein the steps of:

averaging the change gradient values of the same weight type to obtain a change gradient average value of each weight type, and determining a target change gradient value group according to the change gradient average value of each weight type. For example, there are 5 sets of reference weight sets, each of the reference weight sets includes three weight types, where the variation gradient values of the first weight type are { r1, r2, r3, r4, r5}, the variation gradient values of the second weight type are { h1, h2, h3, h4, h5}, the variation gradient values of the third weight type are { g1, g2, g3, g4, g5}, and the target variation gradient value set is { (r 1+ r2+ r3+ r4+ r 5)/5, (h 1+ h2+ h3+ h4+ h 5)/5, (g 1+ g2+ g3+ g4+ g 5)/5 }.

Specifically, after the target change gradient value group is determined, all evaluation data types are screened according to the size of the value in the target change gradient value group, and the screened target evaluation data type is determined. For example, each reference weight group includes 5 weight types, 3 weight types need to be screened out from the reference weight groups as target evaluation data types, a target change gradient value group corresponding to each reference weight group is {2,7, 15,6,7, 11}, then the weight types corresponding to 15, 7, and 11 are used as target evaluation data types, and corresponding service data are collected according to the target evaluation data types.

According to the method provided by the embodiment of the invention, the evaluation data type with deeper influence degree on the service execution quality corresponding to the target service type can be screened from all the evaluation data types according to the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type, so that the data quantity of the collected service data is reduced, and the data collection efficiency is improved.

With reference to the content of the foregoing embodiment, in an embodiment, after inputting each reference weight set to the trained evaluation model and outputting a corresponding reference evaluation score, the method further includes:

Specifically, after the trained evaluation model is obtained, each reference weight group may be input to the trained evaluation model to obtain a reference evaluation score corresponding to each reference weight group, a reference weight group corresponding to a maximum reference evaluation score is determined as a target reference weight group from all reference evaluation scores, the initial weight group is adjusted according to a weight difference degree of the same weight type in the target reference weight group and the initial weight group, and service execution quality evaluation is performed according to the adjusted initial weight group. For example, there are 3 sets of reference weights, the first set of reference weights has a reference evaluation score of 80, the second set of reference weights has a reference evaluation score of 85, and the third set of reference weights has a reference evaluation score of 87, so that the third set of reference weights is the target set of reference weights.

Adjusting the initial weight group according to the weight difference degree of the same weight type in the target reference weight group and the initial weight group, wherein the weight difference degree comprises the following steps:

taking the example that the target reference weight set includes 3 weight types, the target reference weight set is (x) _m ，y _m ，z _m ) If the initial weight set is (x, y, z), the initial weight set is adjusted to (x, y, z) _m ，y _m ，z _m ) And obtaining the adjusted initial weight set, and carrying out service execution quality evaluation on the service data of multiple evaluation data types of the target service type of the current object according to the adjusted initial weight set.

According to the method provided by the embodiment of the invention, the maximum value is determined in all the reference evaluation scores, the optimal reference weight set can be selected from all the reference weight sets to serve as the target reference weight set, the weights of all weight types of the initial weight set are adjusted through the target reference weight set, the service execution quality evaluation is carried out according to the adjusted initial weight set, and the accuracy of the evaluation result can be improved.

In one embodiment, as shown in fig. 3, a business data collecting method applied to a business optimization guidance system of a bank includes:

301. aiming at a target service type required to be executed by a current object, determining a plurality of evaluation data types used for evaluating service execution quality corresponding to the target service type, and acquiring initial weights set by the current object aiming at the evaluation data types;

302. determining an initial weight set formed by all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

303. inputting each reference weight group into the trained evaluation model, and outputting a corresponding reference evaluation score, wherein the reference evaluation score is used for representing the influence degree of the reference weight group on the service execution quality corresponding to the target service type;

304. determining a maximum value in all the reference evaluation scores, and determining a target reference weight set corresponding to the maximum value;

305. and adjusting the initial weight group according to the weight difference degree of the same weight type in the target reference weight group and the initial weight group, and performing service execution quality evaluation according to the adjusted initial weight group.

According to the method provided by the embodiment of the invention, the initial weight of each evaluation data type is adjusted, so that the weight of the evaluation data type with a deeper influence degree on the service execution quality corresponding to the target service type can be improved, and the accuracy of the evaluation score obtained based on the evaluation model is higher.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the application also provides a service data acquisition device for realizing the service data acquisition method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so the specific limitations in one or more embodiments of the service data acquisition device provided below can refer to the limitations on the service data acquisition method in the foregoing, and details are not described here.

In one embodiment, as shown in fig. 4, there is provided a service data acquisition apparatus, 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:

a first determining module 401, configured to determine, for a target service type that needs to be executed by a current object, multiple evaluation data types used for evaluating service execution quality corresponding to the target service type, and obtain an initial weight set by the current object for each evaluation data type;

a second determining module 402, configured to determine an initial weight set formed by the initial weights, and adjust the initial weights in the initial weight set to obtain multiple reference weight sets;

an output module 403, configured to input each reference weight group to the trained evaluation model, and output a corresponding reference evaluation score, where the reference evaluation score is used to represent an influence degree of the reference weight group on service execution quality corresponding to the target service type;

a third determining module 404, configured to determine, according to the adjusted amplitude value of each reference weight group compared to the initial weight group and the changed amplitude value of the reference score value corresponding to each reference weight group compared to the initial score value corresponding to the initial weight group, an influence degree of a weight change of each evaluation data type on service execution quality corresponding to the target service type;

a fourth determining module 405, configured to screen all evaluation data types according to the influence degrees corresponding to the evaluation data types, determine a target evaluation data type after screening, and acquire corresponding service data according to the target evaluation data type when the current object executes a target service corresponding to the target service type.

In one embodiment, the output module 403 includes:

the first obtaining sub-module is used for obtaining sample weight groups set by a plurality of sample objects aiming at each evaluation data type, inputting the sample weight groups corresponding to the sample objects into the initial evaluation model and outputting corresponding sample prediction evaluation scores;

and the training submodule is used for training the initial evaluation model according to the difference between the corresponding sample prediction evaluation value of the sample object and the evaluation value label to obtain the trained evaluation model.

In one embodiment, the fetch submodule includes:

inputting the sample weight group corresponding to the sample object into each boosting model to obtain a plurality of sample pre-selection evaluation scores of the sample weight group corresponding to the sample object;

In one embodiment, the second determining module 402 includes:

acquiring an adjustment amplitude value, and performing multiple times of overall adjustment on the initial weight set according to the adjustment amplitude value to obtain multiple non-repetitive reference weight sets; and adjusting at least one initial weight in the initial weight set in each integral adjustment process, wherein the adjustment amplitude value is a positive value or a negative value.

In one embodiment, the third determining module 404 includes:

the second obtaining submodule is used for obtaining a change gradient value corresponding to each weight type according to the adjustment amplitude value corresponding to the same weight type between the current reference weight group and the initial weight group and the change amplitude value corresponding to the current reference weight group aiming at the current reference weight group, and forming a change gradient value group corresponding to the current reference weight group, wherein the change gradient value is used for representing the influence degree of the corresponding weight type in the current reference weight group on the service execution quality corresponding to the target service type when the corresponding weight type is changed;

and the integration submodule is used for integrating the change gradient value set corresponding to each reference weight set to obtain a target change gradient value set, and each change gradient value in the target change gradient value set is used for representing the influence degree of the corresponding weight type on the service execution quality corresponding to the target service type.

In one embodiment, the output module 403 further includes:

the determining submodule is used for determining the maximum value in all the reference evaluation scores and determining a target reference weight set corresponding to the maximum value;

and the adjusting submodule is used for adjusting the initial weight group according to the weight difference degree of the target reference weight group and the same weight type in the initial weight group, and carrying out service execution quality evaluation according to the adjusted initial weight group.

The modules in the service data acquisition device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 5. The computer apparatus includes a processor, a memory, an input/output interface, a communication interface, a display unit, and an input device. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The input/output interface of the computer device is used for exchanging information between the processor and an external device. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a business data collection method. The display unit of the computer device is used for forming a visual visible picture, and can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.

It will be appreciated by those skilled in the art that the configuration shown in fig. 5 is a block diagram of only a portion of the configuration associated with the present application, and is not intended to limit the computing device to which the present application may be applied, and that a particular computing device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

In one embodiment, the processor when executing the computer program further performs the steps of:

acquiring sample weight sets set by a plurality of sample objects aiming at each evaluation data type, inputting the sample weight sets corresponding to the sample objects into an initial evaluation model, and outputting corresponding sample prediction evaluation scores;

In one embodiment, the processor, when executing the computer program, further performs the steps of:

and adjusting the initial weight group according to the weight difference degree of the same weight type in the target reference weight group and the initial weight group, and performing service execution quality evaluation according to the adjusted initial weight group.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when executed by a processor, performs the steps of:

In one embodiment, the computer program when executed by the processor further performs the steps of:

In one embodiment, a computer program product is provided, comprising a computer program which when executed by a processor performs the steps of:

and screening all the evaluation data types according to the influence degrees corresponding to the evaluation data types, determining the screened target evaluation data types, and acquiring corresponding service data according to the target evaluation data types when the current object executes the target service corresponding to the target service types.

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 for analysis, stored data, displayed data, etc.) referred to in the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the relevant laws and regulations and standards of the relevant countries and regions.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, which may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. 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), among others. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A method for collecting service data, the method comprising:

aiming at a target service type required to be executed by a current object, determining a plurality of evaluation data types for evaluating service execution quality corresponding to the target service type, and acquiring initial weights set by the current object aiming at the evaluation data types;

2. The method of claim 1, wherein the obtaining of the trained evaluation model comprises:

3. The method of claim 2, wherein the initial evaluation model comprises a plurality of boosting models; inputting the corresponding sample weight set of the sample object into an initial evaluation model, and outputting a corresponding sample prediction evaluation score, wherein the method comprises the following steps:

inputting the sample weight group corresponding to the sample object into each boosting model to obtain a plurality of sample pre-selected evaluation scores of the sample weight group corresponding to the sample object;

4. The method of claim 1, wherein determining an initial weight set composed of initial weights, adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets comprises:

acquiring an adjustment amplitude value, and performing multiple times of overall adjustment on the initial weight set according to the adjustment amplitude value to obtain multiple non-repetitive reference weight sets; and adjusting at least one initial weight in the initial weight set in each integral adjusting process, wherein the adjusting amplitude value is a positive value or a negative value.

5. The method according to claim 1, wherein the determining the influence degree of the weight change of each evaluation data type on the service execution quality corresponding to the target service type according to the adjusted amplitude value of each reference weight group compared to the initial weight group and the changed amplitude value of the reference score value corresponding to each reference weight group compared to the initial score value corresponding to the initial weight group comprises:

aiming at a current reference weight group, acquiring a change gradient value corresponding to each weight type according to an adjustment amplitude value corresponding to the same weight type between the current reference weight group and the initial weight group and a change amplitude value corresponding to the current reference weight group, and forming a change gradient value group corresponding to the current reference weight group, wherein the change gradient value is used for representing the influence degree of the corresponding weight type in the current reference weight group on the service execution quality corresponding to the target service type when the corresponding weight type is changed;

6. The method of claim 1, wherein after inputting each set of reference weights to the trained evaluation model and outputting the corresponding reference evaluation score, further comprising:

and adjusting the initial weight group according to the weight difference degree of the target reference weight group and the same weight type in the initial weight group, and performing service execution quality evaluation according to the adjusted initial weight group.

7. A service data collection apparatus, the apparatus comprising:

the system comprises a first determining module, a second determining module and a third determining module, wherein the first determining module is used for determining multiple evaluation data types used for evaluating the service execution quality corresponding to a target service type aiming at the target service type required to be executed by a current object, and acquiring initial weights set by the current object aiming at the evaluation data types;

the second determining module is used for determining an initial weight set formed by all initial weights, and adjusting the initial weights in the initial weight set to obtain a plurality of reference weight sets;

a third determining module, configured to determine, according to an adjustment amplitude value of each reference weight group compared to the initial weight group and a change amplitude value of a reference score value corresponding to each reference weight group compared to the initial score value corresponding to the initial weight group, an influence degree of a weight change of each evaluation data type on service execution quality corresponding to the target service type;

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.

9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.