Method and device for determining object evaluation information

Abstract

The embodiment of the invention provides a method and device for determining object evaluation information, and the method comprises the steps: obtaining first data of a to-be-evaluated object in a first dimension and second data of the to-be-evaluated object in a second dimension, and enabling the first dimension and the second dimension to comprise a plurality of evaluation indexes; determining target feature information corresponding to the first data and the second data according to at least two predetermined target sub-boxes associated with a plurality of evaluation indexes; inputting the target feature information into a pre-trained object evaluation model, and determining a target evaluation result corresponding to the target object; wherein the target evaluation result comprises a first evaluation grade of the to-be-evaluated object in the first dimension and a second evaluation grade of the to-be-evaluated object in the second dimension. According to the technical scheme provided by the embodiment of the invention, the evaluation result corresponding to the to-be-evaluated object is determined by combining a plurality of evaluation indexes of two dimensions, and the determination accuracy and relevance of the evaluation result are improved.

Description

Technical Field

[0001] The embodiments of the present disclosure relate to the technical field of information analysis, and more particularly, to a method and apparatus for determining object evaluation information. Background Art

[0002] In the marketing and risk control fields of banks, there is an evaluation system for customers. The evaluation indicators for customers are inconsistent for different business fields. Different business fields have different evaluation indicators and different evaluation algorithms, which have a great impact on the overall evaluation results of customers. In addition, different business fields have independent evaluation strategies, which may lead to conflicts in the rating results of different businesses and inaccurate evaluation results.

[0003] When implementing the technical solution based on the above method, the inventor found the following problems:

[0004] The above method is unable to combine multiple evaluation indicators of different dimensions to comprehensively determine the evaluation level corresponding to the user. Secondly, the evaluation algorithm of data under different dimensions is highly biased, resulting in low accuracy and low correlation in determining the evaluation level. Summary of the invention

[0005] The embodiments of the present disclosure provide a method and device for determining an evaluation result of an object, so as to realize the effect of combining multiple evaluation indicators of two dimensions to determine the evaluation result corresponding to the object to be evaluated, thereby improving the accuracy of determining the target evaluation result.

[0006] In a first aspect, an embodiment of the present disclosure provides a method for determining object evaluation information, the method comprising:

[0007] Acquire first data of the object to be evaluated in a first dimension and second data of the object to be evaluated in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators;

[0008] Determining target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with the plurality of evaluation indicators;

[0009] Inputting the target feature information into a pre-trained object evaluation model to determine a target evaluation result corresponding to the target object;

[0010] The target evaluation result includes a first evaluation grade of the object to be evaluated under the first dimension and a second evaluation grade of the object to be evaluated under the second dimension.

[0011] In a second aspect, an embodiment of the present invention further provides a device for determining object evaluation information, the device comprising:

[0012] A data acquisition module, used to acquire first data of the object to be evaluated in a first dimension and second data of the object to be evaluated in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators;

[0013] A target feature information determination module, configured to determine target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with the plurality of evaluation indicators;

[0014] A target evaluation result determination module is used to input the target feature information into a pre-trained object evaluation model to determine the target evaluation result corresponding to the target object;

[0015] The target evaluation result includes a first evaluation grade of the object to be evaluated under the first dimension and a second evaluation grade of the object to be evaluated under the second dimension.

[0016] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0017] one or more processors;

[0018] a storage device for storing one or more programs,

[0019] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining object evaluation information as described in any one of the embodiments of the present invention.

[0020] In a fourth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute a method for determining object evaluation information as described in any one of the embodiments of the present invention.

[0021] The technical solution of the disclosed embodiment, after obtaining the first data of the object to be evaluated under the first dimension and the second data under the second dimension, can analyze and process the first data and the second data based on at least two predetermined target bins associated with multiple evaluation indicators to determine the target feature information corresponding to the first data and the second data. Further, the target feature information can be analyzed based on the object evaluation model to determine the target evaluation result corresponding to the target object, and multiple evaluation indicators of two dimensions are combined to determine the target evaluation result corresponding to the object to be evaluated, which solves the problem in the prior art that the relevant data under different dimensions of the object to be evaluated are evaluated separately to obtain the target evaluation results under different dimensions, and there are problems of low accuracy and high independence in determining the target evaluation results. Furthermore, during model training, the evaluation algorithm for data under different dimensions has a large bias, resulting in problems of low accuracy and low correlation. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0023] Figure 1 is a flow chart of a method for determining object evaluation information provided by an embodiment of the present disclosure;

[0024] Figure 2 is a schematic diagram of target feature information provided by an embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of an evaluation result provided by an embodiment of the present disclosure;

[0026] Figure 4 is a flow chart of a method for determining object evaluation information provided by an embodiment of the present disclosure;

[0027] Figure 5 is a schematic diagram of an information value provided by an embodiment of the present disclosure;

[0028] Figure 6 is a flow chart of a method for determining object evaluation information provided by an embodiment of the present disclosure;

[0029] Figure 7 A schematic diagram of the structure of a device for determining object evaluation information provided by an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0032] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0033] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0034] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0035] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0036] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0037] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0038] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0039] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0040] Before introducing the technical solutions provided by the embodiments of the present disclosure, an example description of the application scenarios may be given first.

[0041] The technical solution provided by the embodiments of the present disclosure can be applied in scenarios where a unified method is used to determine any user's evaluation level. Optionally, it can be applied in the banking field to determine the user's corresponding risk control dimension and marketing dimension evaluation level scenario.

[0042] Embodiment 1

[0043] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only the parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0044] Figure 1 It is a flow chart of a method for determining an object evaluation result provided by an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to situations where the evaluation level corresponding to the object to be evaluated under different dimensions is determined. The method can be executed by a device for determining object evaluation information. The device can be implemented in the form of software and / or hardware. The hardware can be an electronic device such as a server. The electronic device can execute the method for determining object evaluation information provided by the present technical solution.

[0045] like Figure 1 As shown, the method includes:

[0046] S110, obtaining first data of the object to be evaluated in a first dimension and second data in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators.

[0047] The object to be evaluated refers to the object whose data needs to be analyzed to determine its evaluation result. The object to be evaluated may be a user holding our bank's resources. The first dimension refers to the risk control dimension.

[0048] The first data refers to the data of the object to be evaluated under the risk control dimension. The second dimension refers to the marketing dimension.

[0049] The second data refers to the data of the object to be evaluated under the marketing dimension.

[0050] It should be noted that both the first dimension and the second dimension include multiple evaluation indicators. Generally, for the object to be evaluated, the evaluation indicators are mainly reflected from four categories of the object to be evaluated, which are the attribute category of the object to be evaluated, the behavior category of the object to be evaluated, the value category of the object to be evaluated, and the demand category of the object to be evaluated. Accordingly, the evaluation indicators under the marketing dimension and the risk control dimension can be divided from the data generated under these four categories. Among them, the evaluation indicators under the risk control dimension may include whether the credit is granted, whether the repayment is in arrears, the average historical rating score indicator, the average bad debt rate indicator, the work status indicator, the income stability indicator, the credit limit indicator, and the total asset indicator. Optionally, the work status indicator may include employment, freelance, part-time, full-time, temporary, and contract workers. The income stability degree indicator may include the high income stability degree, the middle income stability degree, and the low income stability degree. The evaluation indicators under the marketing dimension may include the object basic information indicator, whether to hold a bank card indicator, the bank card status indicator, whether it is a general card indicator, the asset holding indicator, the average transaction value indicator, the average number of transactions indicator, and whether to hold a wealth management product indicator. Optionally, the basic information indicators of the object may include name and gender, etc. The bank card status indicators may include sleep, frozen and normal, etc. The average transaction value indicators may include daily transaction average value, monthly transaction average value and annual transaction average value, etc. The average transaction number indicators may include monthly average transaction number and annual average transaction number, etc.

[0051] It should be noted that after obtaining the first data of the object to be evaluated under the first dimension and the second data under the second dimension, the units of the data can be standardized. For example, the amounts involved in the data of multiple evaluation indicators are all unified in units of 10,000 yuan. It should also be noted that if there is data that needs special marking, it should be specially marked. For example, when the evaluation indicator of whether the credit is granted under the risk control dimension is no, the evaluation indicators such as the average historical rating score and the credit limit are set to -1, which means there is no value. When the evaluation indicator of whether the bank is held under the marketing dimension is no, the evaluation indicators related to bank cards (except credit cards, financial management, etc.) are set to -1.

[0052] Specifically, when the conditions for evaluating the object to be evaluated are met, all data associated with the object to be evaluated can be retrieved from the data system, and the data can be separated in turn according to a plurality of evaluation indicators corresponding to a predetermined first dimension and a plurality of evaluation indicators corresponding to a second dimension to obtain data corresponding to each evaluation indicator.

[0053] S120. Determine target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with a plurality of evaluation indicators.

[0054] Among them, the target binning can convert continuous numerical variables into discrete intervals or categories, thereby simplifying the data processing process and improving the performance of the model. It should be noted that there are at least two target bins associated with each evaluation indicator.

[0055] It should be noted that if Figure 2 As shown in the figure, the target feature information refers to the feature information of the target bin corresponding to the data. For example, when the average bad debt rate is 30%, it corresponds to the third bin, and the target feature information at this time is 3; when the average bad debt rate is 20%, it corresponds to the second bin, and the target feature information at this time is 2. When the working status is individual, the corresponding target feature information is individual; when the working status is enterprise, the corresponding target feature information is enterprise.

[0056] Specifically, for each evaluation indicator, each evaluation indicator is associated with multiple data. In order to analyze the data in the evaluation indicator, the data under the evaluation indicator is binned according to a pre-set binning method. By analyzing the data under each bin, the final bin associated with the evaluation indicator can be determined. The bin associated with the evaluation indicator is finally obtained as the target bin, and the target bin contains the data corresponding to the evaluation indicator. Based on S110, it can be known that the first data and the second data refer to data under different evaluation indicators. For each evaluation indicator, first determine what the target bin corresponding to the evaluation indicator is and the data in the target bin, and then obtain the characteristic value corresponding to the target bin. By splicing the characteristic values ​​of all indicator data, the target feature information can be obtained.

[0057] S130. Input the target feature information into a pre-trained object evaluation model to determine a target evaluation result corresponding to the target object; wherein the target evaluation result includes a first evaluation level of the object to be evaluated under a first dimension and a second evaluation level under a second dimension.

[0058] Among them, the object evaluation model is a pre-trained model. By inputting the target feature information into the model, the target evaluation result can be obtained. The target evaluation result refers to the evaluation system obtained by integrating the first dimension and the second dimension. The first evaluation level refers to the evaluation level corresponding to the risk control dimension of the object to be evaluated, and the second evaluation level refers to the evaluation level corresponding to the marketing dimension of the object to be evaluated. Figure 3As shown, the first dimension risk control dimension is divided into five levels IV: extremely low risk, relatively low risk, medium risk, medium-high risk and high risk; the second dimension marketing dimension is divided into eight levels AH: important value, important development, important maintenance, important retention, general value, general development, general maintenance and retention. At this time, 8*5 evaluation levels of 40 combinations of first dimension levels and second dimension levels can be obtained. The target evaluation result can be one of these 40 evaluation levels. After obtaining all possible target evaluation results, all possible target evaluation results are included in the target variable Y={evaluation level 1, evaluation level 2,..., evaluation level 40}. For example, when the target evaluation result is important value + extremely low risk, extremely low risk is the first evaluation level of the object to be evaluated under the first dimension, and important value is the second evaluation level of the object to be evaluated under the second dimension.

[0059] Specifically, first obtain a pre-trained object evaluation model, the input of the object evaluation model can be the target feature information, and the output of the object evaluation model can be the target evaluation result corresponding to the target object. Then, the target feature information is input into the pre-trained object evaluation model, and the target evaluation result corresponding to the target feature information can be obtained. Based on S110, it can be known that the first dimension and the second dimension refer to the risk control dimension and the marketing dimension respectively. For each dimension, the target evaluation result may include the evaluation level under the dimension.

[0060] The technical solution of the disclosed embodiment, after obtaining the first data of the object to be evaluated under the first dimension and the second data under the second dimension, can analyze and process the first data and the second data based on at least two predetermined target bins associated with multiple evaluation indicators to determine the target feature information corresponding to the first data and the second data. Further, the target feature information can be analyzed based on the object evaluation model to determine the target evaluation result corresponding to the target object, and multiple evaluation indicators of two dimensions are combined to determine the target evaluation result corresponding to the object to be evaluated, which solves the problem in the prior art that the relevant data under different dimensions of the object to be evaluated are evaluated separately to obtain the target evaluation results under different dimensions, and there are problems of low accuracy and high independence in determining the target evaluation results. Furthermore, during model training, the evaluation algorithm for data under different dimensions has a large bias, resulting in problems of low accuracy and low correlation.

[0061] Embodiment 2

[0062] Figure 4This is a flow chart of a method for determining an object evaluation result provided by an embodiment of the present invention. Based on the above embodiment, a detailed description is given of determining at least two target bins corresponding to the evaluation index. The specific implementation method can refer to the technical solution of this embodiment. Among them, the technical terms that are the same or corresponding to the above embodiment are not repeated here.

[0063] like Figure 4 As shown, the method specifically comprises the following steps:

[0064] S210: Obtain first data of the object to be evaluated in a first dimension and second data in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators.

[0065] S220, obtaining a plurality of sample data, wherein the sample data includes at least one object identifier of an object to be used and data to be used of the object to be used in a first dimension and a second dimension. The object to be used refers to an object whose corresponding data is analyzed by a preset initial binning constraint condition to determine which bins correspond to each evaluation indicator.

[0066] The object identifier refers to a globally unique value associated with the object to be used and used to unambiguously identify the object to be used. The data to be used refers to the data corresponding to all evaluation indicators of all objects to be used in the first dimension and the second dimension.

[0067] Specifically, multiple sample data of multiple objects to be used are obtained. For each object to be used, the sample data includes a globally unique value for unambiguously identifying the object to be used, and also includes sample data corresponding to each evaluation indicator under the first dimension and sample data corresponding to each evaluation indicator under the second dimension.

[0068] S230. For multiple evaluation indicators, determine at least one to-be-processed bin corresponding to the evaluation indicator and the indicator data in the to-be-processed bin according to the initial bin constraint corresponding to the evaluation indicator and the to-be-used data corresponding to at least one to-be-used object under the evaluation indicator.

[0069] Among them, the initial binning constraint refers to the pre-set constraint for binning according to the principle of initial binning corresponding to each evaluation indicator. In the initial binning stage, the binning constraint will be set according to the characteristics of the characteristic indicators. For example, for discrete data, the number of target bins is usually equal to the number of data categories of the data. For example, the data categories of the evaluation indicator of stability are three discrete values ​​of high, medium and low, including three categories, so the target bins associated with stability are three. For example, for continuous data, the continuous data is simply divided into intervals according to equal width, equal distance and other methods, and one interval range is a target bin. Take the initial binning constraint corresponding to the holding asset indicator as an example to illustrate. The initial binning constraint condition can be: the range of the held assets can be 0-100,000 (not included), 100,000-500,000 (not included), 500,000-1,000,000 (not included), 1,000,000-3,000,000 (not included), 3,000,000-5,000,000 (not included), 5,000,000-10,000,000 (not included), 10,000,000-20,000,000 (not included), 20,000,000-50,000 (not included) and more than 50,000,000. The data with held assets within the range can be used as nine target bins. The initial binning constraint condition corresponding to the average transaction value indicator is used as an example for explanation. The initial binning constraint condition can be: first, the data with the daily average transaction value of 0-10,000 (not included), 10,000-50,000 (not included), 50,000-100,000 (not included) and more than 100,000 are used as four target bins. Then, for each of the four target bins corresponding to the average daily transaction value, the data with the average monthly transaction value of 0-10,000 (not included), 10,000-50,000 (not included), 50,000-100,000 (not included), and more than 100,000 yuan are continued to be used as four target bins. Finally, for each of the four target bins corresponding to the average monthly transaction value, the data with the average annual transaction value of 0-10,000 (not included), 10,000-50,000 (not included), 50,000-100,000 (not included), and more than 100,000 yuan are used as four target bins again; the initial bin constraint corresponding to the average historical rating score indicator is used as an example to illustrate. The initial bin constraint can be: the data with the average historical rating score between 0-300 points (not included), 300-600 points (not included), and more than 600 points are used as three target bins; the initial bin constraint corresponding to the average bad debt rate indicator is used as an example to illustrate. The initial binning constraints may be: the average bad debt rate is between 0-10% (inclusive), 10-30% (inclusive), 30-50% (inclusive) and data above 50% as four target bins; the initial binning constraints corresponding to the average number of transactions indicator are used as an example for explanation.The initial binning constraints may be: taking the data with average transaction times between 0-5 (exclusive), 5-10 (exclusive), 10-20 (exclusive), 20-50 (exclusive), 50-100 (exclusive), 100-300 (exclusive), 300-500 (exclusive), 500-1000 (exclusive) and above 1000 as nine target bins.

[0070] The pending bins refer to all bins obtained under all evaluation indicators according to the initial bin constraints corresponding to the evaluation indicators. For each pending bin, the indicator data refers to the data to be used under a pending bin.

[0071] Specifically, for each evaluation indicator, each evaluation indicator has its corresponding initial binning constraint. After obtaining the data to be used corresponding to the object to be used under each evaluation indicator, the data of the object to be used is binned according to the initial binning constraint corresponding to each indicator. For each evaluation indicator, after binning according to the initial binning constraint, the bins to be processed corresponding to each evaluation indicator can be determined. After binning the data under each evaluation indicator of the object to be used, the data under each evaluation indicator is saved in the bins to be processed corresponding to each evaluation indicator.

[0072] S240: For multiple evaluation indicators, determine gain information and information value of the evaluation indicator according to indicator data in at least one to-be-processed sub-box corresponding to the evaluation indicator.

[0073] In this embodiment, the gain information refers to the influence of the evaluation indicator. If the gain information of an evaluation indicator is larger, it not only means that the evaluation indicator is more important in the corresponding dimension, but also means that the evaluation indicator has a greater influence among all evaluation indicators. For example, assuming that the gain information of whether to grant credit is 5 and the gain information of the average bad debt rate is 4, then the influence of the whether to grant credit evaluation indicator is greater than that of the average bad debt rate evaluation indicator. Based on the gain information results, evaluation indicators with higher influence and purity can be screened.

[0074] Information value is a way to measure the predictive ability of an evaluation indicator. The prediction evaluation indicators are ranked according to their importance. The larger the value of the information value, the stronger the predictive ability of the evaluation indicator. In this embodiment, the greater the difference in information value, the stronger the ability of the bin to distinguish samples with different evaluation results.

[0075] In an embodiment of the present invention, step S240 may include: determining at least one evaluation result corresponding to the evaluation indicator based on the data to be used corresponding to the evaluation indicator; wherein the evaluation result includes an evaluation level under the first dimension and an evaluation level under the second dimension; determining the first information entropy of the evaluation indicator based on the first sample number corresponding to each evaluation result in at least one evaluation result, the total number of multiple sample data, and the first information entropy calculation function; determining the second information entropy corresponding to the evaluation indicator based on at least one evaluation result corresponding to each to-be-processed bin under the evaluation indicator, the second sample number corresponding to at least one evaluation result, the number of bin samples in the to-be-processed bin, and the second information entropy calculation function; determining the gain information of the evaluation indicator based on the first information entropy and the second information entropy corresponding to the evaluation indicator.

[0076] The evaluation result corresponding to the evaluation indicator can be one or more. For example, if the evaluation indicator is income stability, and the to-be-used data corresponding to the evaluation indicator is high income stability, then the evaluation result corresponding to the evaluation indicator is multiple, which may include Figure 3 Columns I and II in .

[0077] It should be noted that the evaluation results include the evaluation level under the first dimension and the evaluation level under the second dimension. For example, for an evaluation result of important value + very low risk, very low risk is the evaluation level under the first dimension, and important value is the evaluation level under the second dimension.

[0078] In this embodiment, the binning condition is not considered first. The first sample size refers to the sample size corresponding to each evaluation result. For example, it can be the sample size corresponding to the vth evaluation result, which can be expressed as D v The total number of sample data refers to the total number of samples, which can be represented by D. The first information entropy calculation function refers to the function used to calculate all information entropies corresponding to the rating results without considering the binning condition. The sum of the information entropies of all rating results calculated by the first information entropy calculation function is called the first information entropy. The first information entropy calculation function can be:

[0079]

[0080] It should be noted that for each rating result, the information entropy corresponding to the evaluation result can be: The first information entropy can be obtained by summing up the information entropy corresponding to each evaluation result.

[0081] In this embodiment, considering the binning condition, if the bin to be processed is A i , the second sample size refers to the bin A to be processed iThe number of samples of the vth rating result can be expressed as D Aiv The number of samples in the pending bins represents the number of samples in the pending bins A. i The total number of samples can be expressed as D Ai The second information entropy calculation function refers to a function used to calculate the information entropy corresponding to the rating results of all the bins to be processed under the binning condition. The sum of the information entropy of the rating results of all the bins to be processed calculated by the second information entropy calculation function is called the second information entropy. The second information entropy calculation function can be:

[0082]

[0083] It should be noted that for each pending bin, for example, the pending bin may be A i , the pending A i The information entropy corresponding to all the evaluation results in the bin to be processed can be: The second information entropy can be obtained by summing the information entropy corresponding to the evaluation result of each bin to be processed.

[0084] Among them, the gain information of the evaluation indicator can be determined according to the first information entropy and the second information entropy corresponding to the evaluation indicator. The calculation formula of the gain information of the evaluation indicator can be:

[0085] Gain(D)=Ent(D 不考虑分箱 )-Ent(D 考虑分箱 )

[0086] Among them, Gain(D) represents the gain information of the evaluation index.

[0087] Specifically, after obtaining the corresponding data to be used according to the evaluation index, determine the evaluation level under the first dimension and the evaluation level under the second dimension corresponding to the evaluation index. Then, calculate the information entropy corresponding to the evaluation result without considering the binning, that is, the first information entropy. Further, calculate the information entropy under all bins corresponding to the evaluation result when considering the binning, that is, the second information entropy. Finally, subtract the second information entropy from the first information entropy to determine the gain information of the evaluation index.

[0088] In an embodiment of the present invention, step S240 may further include: for at least one sub-box to be processed, determining at least two evaluation results corresponding to the indicator data in the sub-box to be processed; for any two evaluation results, determining the sub-box value corresponding to the sub-box to be processed according to the third sample quantity and the fourth sample quantity corresponding to the two evaluation results in the sub-box to be processed, and the fifth sample quantity and the sixth sample quantity corresponding to the two evaluation results under the evaluation indicator; determining the sub-box value corresponding to the sub-box to be processed according to the third sample quantity, the fourth sample quantity, the fifth sample quantity and the sixth sample quantity corresponding to all any two evaluation results in the sub-box to be processed; determining the information value of the evaluation indicator according to the sub-box value of at least one sub-box to be processed associated with the evaluation indicator.

[0089] It should be noted that for all pending bins of each characteristic indicator, all evaluation results under the pending bins are obtained. For a pending bin under a certain characteristic indicator, the third sample number refers to the number of samples with evaluation result i in the pending bin. For example, the third sample number can be the number of samples with evaluation result i in the kth pending bin for each characteristic indicator, which can be expressed as num ki The fourth sample number refers to the number of samples with evaluation result i in a certain bin to be processed. For example, the fourth sample number can be the number of samples with evaluation result j in the kth bin to be processed for each feature indicator, which can be expressed as num kj To express; for a certain characteristic index, the fifth sample size refers to the total number of samples of evaluation result i in the characteristic index, which can be expressed by sum i The sixth sample size refers to the total number of samples with evaluation result j in the feature indicator, which can be expressed as sum j The bin value refers to the sample ratio difference between the evaluation results of each pair in a bin to be processed under all evaluation results. Optionally, the bin value corresponding to the bin to be processed can be determined according to the third sample number, the fourth sample number, the fifth sample number and the sixth sample number. The information value of the evaluation indicator can be determined for the bin values ​​of all bins to be processed associated with the current evaluation indicator.

[0090] In an embodiment of the present invention, for any two evaluation results, the first rating proportion is determined based on the third sample number and the fifth sample number of the two evaluation results, and the second rating proportion is determined based on the fourth sample number and the sixth sample number; the rating values ​​corresponding to the two evaluation results are determined based on the first rating proportion, the second rating proportion and the sample proportion function; the bin value of the bin to be processed is determined based on the rating values ​​corresponding to all any two evaluation results in the bin to be processed.

[0091] The first rating ratio refers to the value of the third sample number of the two evaluation results to the fifth sample number. For example, the first rating ratio can be the ratio of the number of samples with evaluation result i in the pending bin to the total number of samples of the evaluation results corresponding to the pending bin, which can be expressed as P ki The second rating ratio refers to the ratio of the fourth sample number of the two evaluation results to the sixth sample number. For example, the second rating ratio can be the ratio of the number of samples with evaluation result j in the pending bin to the total number of evaluation result samples corresponding to the pending bin, which can be expressed as P kj To express.

[0092] The sample proportion function is a function that determines the rating value corresponding to any two evaluation results. The rating value can be obtained by determining any two evaluation results according to the sample proportion function. If the bin to be processed is k at this time, the sample proportion function can be:

[0093]

[0094] Among them, WOE is the rating value.

[0095] In an embodiment of the present invention, in the sub-box to be processed, the rating difference information is determined based on the first rating proportion and the second rating proportion of the two evaluation results, and the values ​​to be superimposed of the two evaluation results are determined based on the rating difference information and the rating values ​​corresponding to the two evaluation results; the sub-box value of the sub-box to be processed is obtained by accumulating all the values ​​to be superimposed in the sub-box to be processed.

[0096] The rating difference information is the first rating percentage minus the second rating percentage of the two evaluation results. For example, the rating difference information can be: ki -p kj The sum of the rating difference information and the rating values ​​corresponding to the two evaluation results is the value to be superimposed. For example, the superimposed value can be represented by IV, and the value to be superimposed can be: IV = (p ki -p kj )*WOE. The sum of all the values ​​to be added in the pending bin is the bin value of the pending bin. For example, the bin value of the pending bin can be expressed as IV k To express, the bin value of the bin to be processed can be:

[0097] IV k =∑(p ki -p kj )*WOE(1≤i≤39,i+1≤j≤1).

[0098] S250: Determine at least two target bins corresponding to the evaluation indicators based on the gain information and information values ​​of the multiple evaluation indicators.

[0099] Among them, for each evaluation indicator, the target bin corresponding to the evaluation indicator refers to the bin corresponding to when the information value of the evaluation indicator is within the preset value range.

[0100] In an embodiment of the present invention, when the information value of an evaluation indicator is not within a preset value range, and, based on the gain information, it is determined that the evaluation indicator is within the target range among all evaluation indicators, at least one to-be-processed sub-box of the evaluation indicator is used as the target sub-box of the evaluation indicator; when the information value of an evaluation indicator is not within a preset value range, and, based on the gain information, it is determined that the evaluation indicator is not within the target range among all evaluation indicators, the sub-box value of at least one to-be-processed sub-box corresponding to the evaluation indicator is obtained; at least one to-be-processed sub-box is merged according to the sub-box value to update at least one to-be-processed sub-box corresponding to the evaluation indicator, and the gain information and information value of at least one to-be-processed sub-box after the update are determined. It should be noted that, if Figure 5 As shown in , the information value of the evaluation indicator is within the preset value range, which means that the value of the information value of the evaluation indicator can be between 0.02 and 0.5. Figure 5 As shown in the figure, if the information value of the evaluation indicator is less than 0.02, it means that the prediction effect of the binning is almost non-existent; if the information value of the evaluation indicator is between 0.02 and 0.1, it means that the prediction effect of the binning is weak; if the information value of the evaluation indicator is between 0.1 and 0.3, it means that the prediction effect of the binning is medium; if the information value of the evaluation indicator is between 0.3 and 0.5, it means that the prediction effect of the binning is strong; if the information value of the evaluation indicator is greater than 0.5, it means that the prediction effect of the binning is difficult to perform and needs to be confirmed. If the information value of the evaluation indicator is outside the preset value range, it means that the value of the information value of the evaluation indicator is less than 0.02 or greater than 0.5.

[0101] It should be noted that for each evaluation indicator, a value of gain information is preset, and the range of the value of the gain information of the evaluation indicator lower than the preset value is the target range of the evaluation indicator. After sorting the gain information of the evaluation indicator from top to bottom, if it is within the target range, the evaluation indicator has little impact on the overall situation and is an evaluation indicator with low impact; if it is outside the target range, the evaluation indicator has a greater impact on the overall situation and is an evaluation indicator with high impact.

[0102] It should be noted that when the bins corresponding to the evaluation index at this time are not very reasonable, it is necessary to merge the pending bins corresponding to the evaluation index. At this time, the bin value of the pending bin corresponding to the evaluation index is obtained, and the bins with similar bin values ​​can be merged. Specifically, when the information value of the evaluation index is not within the preset value range, and the evaluation index is determined to be within the target range among all evaluation indicators based on the gain information, it means that although the prediction ability is low at this time, because the influence of the evaluation index among all evaluation indicators is smaller at this time, at least one pending bin of the evaluation index can be used as the target bin of the evaluation index. When the information value of the evaluation index is not within the preset value range, and the evaluation index is determined to be not within the target range among all evaluation indicators based on the gain information, it means that the prediction ability is low at this time, and the greater the influence of the evaluation index among all evaluation indicators, then for the pending bins whose gain information of the evaluation index is not within the preset value range, it is necessary to merge. The bin value of the bin to be processed is obtained, and bins with similar bin values ​​are merged, and the bin to be processed is updated after merging, and the gain information and information value of the bin to be processed after the update are calculated according to the step S240.

[0103] S260: Determine target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with a plurality of evaluation indicators.

[0104] S270. Input the target feature information into a pre-trained object evaluation model to determine a target evaluation result corresponding to the target object; wherein the target evaluation result includes a first evaluation level of the object to be evaluated under a first dimension and a second evaluation level under a second dimension.

[0105] The technical solution of the disclosed embodiment, after obtaining multiple sample data, can determine the bins to be processed corresponding to the evaluation indicators and the indicator data in the bins to be processed based on the initial bin constraints corresponding to the evaluation indicators and the data to be used corresponding to the user under the evaluation indicators. Furthermore, for each evaluation indicator, the gain information and information value of the evaluation indicator can be determined based on the indicator data corresponding to the evaluation indicator. Finally, based on the gain information and information value of multiple evaluation indicators, the target bins corresponding to the evaluation indicators can be determined. The advantage of determining the target bins corresponding to the evaluation indicators based on the above method is that the target bins can improve the accuracy of determining the target bins by combining the gain information and information value of the evaluation indicators.

[0106] Embodiment 3

[0107] Figure 6The present invention provides a flow chart of a method for determining object evaluation information provided in an embodiment of the present invention. Based on the above embodiment, a detailed description of determining an object evaluation model is given. The specific implementation method can refer to the technical solution of this embodiment. The technical terms that are the same or corresponding to the above embodiment are not repeated here.

[0108] like Figure 6 As shown, the method specifically comprises the following steps:

[0109] S310: For multiple sample data, determine the bins to which the data to be used belongs under each evaluation indicator according to the data to be used in the sample data, and determine the characteristic values ​​corresponding to the bins.

[0110] The multiple sample data may be understood as the object identifier of any object to be used and the data to be used of the object to be used in the first dimension and the second dimension.

[0111] Specifically, after obtaining the data to be used in the sample data, for determining the data to be used of any object to be used, the method provided in S220-S250 is used to determine the bin to which the data to be used belongs under each evaluation indicator, and determine the eigenvalue corresponding to the bin.

[0112] S320: Determine the sample characteristics of the sample data according to the characteristic value corresponding to each evaluation indicator in the data to be used under the corresponding bin.

[0113] Among them, the sample characteristics of the sample data can be obtained by replacing the bins to which the sample data belongs under each evaluation indicator with the corresponding characteristic values ​​under the bins.

[0114] Specifically, Figure 2 As shown, after determining the eigenvalue corresponding to each evaluation indicator in the data to be used under the corresponding bin, the sample features of the sample data under each evaluation indicator are determined according to the eigenvalue. For example, if the credit limit of the evaluation indicator of the data to be used is 200,000, then the bin to which it belongs is bin 2, and the eigenvalue corresponding to bin 2 is 2. Then, the bin to which each sample data belongs under each evaluation indicator is replaced with the eigenvalue corresponding to the bin to which it belongs, and the sample features of the sample data can be determined. The sample features of the sample data can be used as training samples for the object evaluation model in the future.

[0115] In order to improve the accuracy of model training, multiple sample data of different objects to be used under each evaluation indicator can be obtained.

[0116] S330: Based on the sample features and the actual evaluation results of the predetermined sample data, the sample data is updated to obtain an object evaluation model through training based on the updated sample data.

[0117] The actual evaluation result of the sample data is predetermined. In this case, the sample data includes sample features of the sample data and the actual evaluation result of the sample data.

[0118] It should be noted that the model parameters in the object evaluation model do not meet the expected requirements. Therefore, there is a certain difference between the evaluation results output based on the model parameters at this time and the actual evaluation results. Therefore, the corresponding error loss value can be determined based on the evaluation results corresponding to each sample feature and the actual evaluation results.

[0119] In this embodiment, the evaluation model of the object to be trained can be a LightGBM network model. It should be noted that the evaluation results corresponding to each sample feature can be obtained, that is, the specific model type is not specifically limited.

[0120] It should be noted that the training parameters can be set to default values ​​before training the evaluation model of the object to be trained. When training the evaluation model of the object to be trained, the training parameters in the model can be corrected based on the output results of the evaluation model of the object to be trained, that is, the object evaluation model can be obtained by correcting the loss function in the evaluation model of the object to be trained. Each evaluation result has a corresponding loss value, which is based on the evaluation results corresponding to each sample feature and the actual evaluation results.

[0121] Specifically, after inputting each sample feature into the evaluation model of the object to be trained, the evaluation model of the object to be trained can obtain an evaluation result corresponding to each sample feature. According to the evaluation result and the actual evaluation result, the loss value corresponding to the sample feature can be determined, and the model parameters in the evaluation model of the object to be trained can be corrected by using the back-propagation method.

[0122] Specifically, the training error of the loss function, that is, the loss parameter, can be used as a condition for detecting whether the loss function has reached convergence, such as whether the training error is less than the preset error or whether the error change trend tends to be stable, or whether the current number of iterations is equal to the preset number. If the detection reaches the convergence condition, such as the training error of the loss function is less than the preset error or the error change tends to be stable, it indicates that the training of the evaluation model of the object to be trained is completed, and the iterative training can be stopped at this time. If it is detected that the convergence condition is not met at present, the features of each sample can be further obtained to train the evaluation model of the object to be trained until the training error of the loss function is within the preset range.

[0123] When the training error of the loss function reaches convergence, the object evaluation model training is considered complete. At this time, in order to verify the accuracy of the object evaluation model, a test set is obtained, which contains the features of each sample and the actual evaluation results. After the sample features are input into the object evaluation model, the object evaluation model can obtain the evaluation results. After all the test sets have obtained the evaluation results, the probability that the evaluation results of the object evaluation model are correct can be obtained. When the accuracy reaches a certain score, the object evaluation model training is completed and can be used as the target object evaluation model.

[0124] The technical solution of the disclosed embodiment is, for multiple sample data, based on the data to be used in the sample data, to which the data to be used belongs under each evaluation indicator. Further, the characteristic value corresponding to the bin is determined, and the sample characteristics of the sample data are determined according to the characteristic value corresponding to each evaluation indicator in the data to be used under the corresponding bin. Finally, based on the sample characteristics and the actual evaluation results of the predetermined sample data, the sample data is updated to obtain an object evaluation model based on the training of the updated sample data, which solves the problem in the prior art that the relevant data under different dimensions of the object to be evaluated are evaluated separately to obtain the target evaluation results under different dimensions, and there is a low accuracy and high independence in determining the target evaluation results, and solves the problem that when the model is trained, the evaluation algorithm for data under different dimensions has a large bias, resulting in low accuracy and low correlation. The accuracy and correlation of the evaluation results are improved.

[0125] Embodiment 4

[0126] Figure 7 It is a structural diagram of a device for determining object evaluation information provided by an embodiment of the present disclosure. As shown in the figure, the device includes: a data acquisition module 410, a target feature information determination module 420 and a target evaluation result determination module 430.

[0127] A data acquisition module 410 is used to acquire first data of an object to be evaluated in a first dimension and second data in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators; a target feature information determination module 420 is used to determine the target feature information corresponding to the first data and the second data based on at least two target bins predetermined and associated with the multiple evaluation indicators; a target evaluation result determination module 430 is used to input the target feature information into a pre-trained object evaluation model to determine the target evaluation result corresponding to the target object; wherein the target evaluation result includes a first evaluation grade of the object to be evaluated in the first dimension and a second evaluation grade in the second dimension.

[0128] The technical solution of the disclosed embodiment, after obtaining the first data of the object to be evaluated under the first dimension and the second data under the second dimension, can analyze and process the first data and the second data based on at least two predetermined target bins associated with multiple evaluation indicators to determine the target feature information corresponding to the first data and the second data. Further, the target feature information can be analyzed based on the object evaluation model to determine the target evaluation result corresponding to the target object, and multiple evaluation indicators of two dimensions are combined to determine the target evaluation result corresponding to the object to be evaluated, which solves the problem in the prior art that the relevant data under different dimensions of the object to be evaluated are evaluated separately to obtain the target evaluation results under different dimensions, and there are problems of low accuracy and high independence in determining the target evaluation results. Furthermore, during model training, the evaluation algorithm for data under different dimensions has a large bias, resulting in problems of low accuracy and low correlation.

[0129] On the basis of the above technical solutions, the device also includes:

[0130] A sample data acquisition module, used to acquire a plurality of sample data, wherein the sample data includes at least one object identifier of an object to be used and data to be used of the object to be used in the first dimension and the second dimension;

[0131] An evaluation indicator bin determination module is used to determine, for the multiple evaluation indicators, at least one to-be-processed bin corresponding to the evaluation indicator and the indicator data in the to-be-processed bin according to the initial bin constraint corresponding to the evaluation indicator and the to-be-used data corresponding to the at least one to-be-used object under the evaluation indicator;

[0132] An evaluation index result determination module, used to determine the gain information and information value of the evaluation index according to the index data in at least one to-be-processed sub-box corresponding to the evaluation index for the multiple evaluation indexes;

[0133] The target bin determination module is used to determine at least two target bins corresponding to the evaluation indicators according to the gain information and information value of the multiple evaluation indicators.

[0134] On the basis of the above technical solutions, the evaluation index result determination module further includes: a gain information determination submodule and an information value determination submodule, including:

[0135] An evaluation result determination unit, configured to determine at least one evaluation result corresponding to the evaluation indicator according to the to-be-used data corresponding to the evaluation indicator; wherein the evaluation result includes an evaluation grade under the first dimension and an evaluation grade under the second dimension;

[0136] A first information entropy determination unit, configured to determine a first information entropy of the evaluation indicator according to a first sample quantity corresponding to each evaluation result in the at least one evaluation result, a total quantity of the plurality of sample data, and a first information entropy calculation function;

[0137] A second information entropy determination unit, configured to determine the second information entropy corresponding to the evaluation indicator according to at least one evaluation result corresponding to each of the to-be-processed bins under the evaluation indicator, the second number of samples corresponding to the at least one evaluation result, the number of bin samples in the to-be-processed bins, and a second information entropy calculation function;

[0138] A gain information determining unit, configured to determine the gain information of the evaluation indicator according to the first information entropy and the second information entropy corresponding to the evaluation indicator;

[0139] An evaluation result determination unit, configured to determine, for the at least one sub-box to be processed, at least two evaluation results corresponding to the indicator data in the sub-box to be processed;

[0140] A sample quantity determination unit, for determining, for any two evaluation results, a third sample quantity and a fourth sample quantity corresponding to the two evaluation results in the to-be-processed bins, and a fifth sample quantity and a sixth sample quantity corresponding to the two evaluation results under the evaluation index;

[0141] A bin value determination unit, configured to determine the bin value corresponding to the bin to be processed according to the third sample quantity, the fourth sample quantity, the fifth sample quantity and the sixth sample quantity corresponding to any two evaluation results in the bin to be processed;

[0142] The information value determination unit is used to determine the information value of the evaluation indicator according to the bin value of at least one to-be-processed bin associated with the evaluation indicator.

[0143] Based on the above technical solutions, the box value determination unit includes:

[0144] For any two evaluation results, determine the first rating proportion based on the third sample size and the fifth sample size of the two evaluation results, and determine the second rating proportion based on the fourth sample size and the sixth sample size; determine the rating values ​​corresponding to the two evaluation results based on the first rating proportion, the second rating proportion and the sample proportion function; determine the bin value of the bin to be processed based on the rating values ​​corresponding to all any two evaluation results in the bin to be processed.

[0145] On the basis of the above technical solutions, the bin value of the bin to be processed is determined according to the rating values ​​corresponding to any two evaluation results in the bin to be processed, including:

[0146] In the sub-box to be processed, the rating difference information is determined based on the first rating proportion and the second rating proportion of the two evaluation results, and the values ​​to be superimposed of the two evaluation results are determined based on the rating difference information and the rating values ​​corresponding to the two evaluation results; the sub-box value of the sub-box to be processed is obtained by accumulating all the values ​​to be superimposed in the sub-box to be processed.

[0147] Based on the above technical solutions, the target bin determination module includes:

[0148] When the information value of the evaluation indicator is not within the preset value range, and it is determined based on the gain information that the evaluation indicator is within the target range among all the evaluation indicators, at least one to-be-processed sub-box of the evaluation indicator is used as the target sub-box of the evaluation indicator; when the information value of the evaluation indicator is not within the preset value range, and it is determined based on the gain information that the evaluation indicator is not within the target range among all the evaluation indicators, the sub-box value of at least one to-be-processed sub-box corresponding to the evaluation indicator is obtained; the at least one to-be-processed sub-box is merged according to the sub-box value to update the at least one to-be-processed sub-box corresponding to the evaluation indicator, and the gain information and information value of the at least one to-be-processed sub-box after the update are determined.

[0149] On the basis of the above technical solutions, the device also includes:

[0150] A feature value determination module, for determining, for the plurality of sample data, the bins to which the to-be-used data belongs under each evaluation indicator according to the to-be-used data in the sample data, and determining feature values ​​corresponding to the bins;

[0151] A sample feature determination module, used to determine the sample features of the sample data according to the feature value corresponding to each evaluation indicator in the data to be used under the corresponding bin;

[0152] The object evaluation model determination module is used to update the sample data based on the sample characteristics and the actual evaluation results of the sample data determined in advance, so as to obtain the object evaluation model through training based on the updated sample data.

[0153] Based on the above technical solutions, the first dimension is the risk control dimension, the second dimension is the marketing dimension, and the target evaluation result is the first evaluation level of the object to be evaluated corresponding to the risk control dimension, and the second evaluation level of the object to be evaluated corresponding to the marketing dimension.

[0154] The device for determining object evaluation information provided by the embodiments of the present disclosure can execute the method for determining object evaluation information provided by any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0155] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present disclosure.

[0156] Embodiment 5

[0157] Figure 8 Schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 8 , which shows an electronic device (eg, Figure 8 The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0158] like Figure 8 As shown, the electronic device 500 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An edit / output (I / O) interface 505 is also connected to the bus 504.

[0159] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0160] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 509, or installed from a storage device 508, or installed from a ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0161] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0162] The electronic device provided in the embodiment of the present disclosure and the method for determining object evaluation information provided in the above embodiment belong to the same inventive concept. Technical details not fully described in this embodiment can be found in the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0163] Embodiment 6

[0164] An embodiment of the present disclosure provides a computer storage medium having a computer program stored thereon. When the program is executed by a processor, the method for determining object evaluation information provided in the above embodiment is implemented.

[0165] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0166] In some embodiments, the server can communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0167] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0168] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device:

[0169] Acquire first data of the object to be evaluated in a first dimension and second data of the object to be evaluated in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators;

[0170] Determining target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with the plurality of evaluation indicators;

[0171] Inputting the target feature information into a pre-trained object evaluation model to determine a target evaluation result corresponding to the target object;

[0172] The target evaluation result includes a first evaluation grade of the object to be evaluated under the first dimension and a second evaluation grade of the object to be evaluated under the second dimension.

[0173] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0174] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0175] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0176] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0177] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0178] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0179] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0180] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

Claims

1. A method for determining object evaluation information, characterized in that include: Acquire first data of the object to be evaluated in a first dimension and second data of the object to be evaluated in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators; Determining target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with the plurality of evaluation indicators; Inputting the target feature information into a pre-trained object evaluation model to determine a target evaluation result corresponding to the target object; The target evaluation result includes a first evaluation grade of the object to be evaluated under the first dimension and a second evaluation grade of the object to be evaluated under the second dimension.

2. The method according to claim 1, characterized in that The method further comprises: Acquire a plurality of sample data, wherein the sample data includes at least one object identifier of an object to be used and data to be used of the object to be used in the first dimension and the second dimension; For the multiple evaluation indicators, according to the initial binning constraints corresponding to the evaluation indicators and the to-be-used data corresponding to the at least one to-be-used object under the evaluation indicators, determine at least one to-be-processed bin corresponding to the evaluation indicators and the indicator data in the to-be-processed bin; For the multiple evaluation indicators, determine the gain information and information value of the evaluation indicator according to the indicator data in at least one to-be-processed sub-box corresponding to the evaluation indicator; At least two target bins corresponding to the evaluation indicators are determined according to the gain information and the information value of the multiple evaluation indicators.

3. The method according to claim 2, characterized in that The step of determining the gain information of the evaluation indicator according to the indicator data in at least one to-be-processed sub-box corresponding to the evaluation indicator comprises: Determine at least one evaluation result corresponding to the evaluation indicator according to the to-be-used data corresponding to the evaluation indicator; wherein the evaluation result includes an evaluation grade under the first dimension and an evaluation grade under the second dimension; Determine the first information entropy of the evaluation indicator according to the first sample quantity corresponding to each evaluation result in the at least one evaluation result, the total quantity of the plurality of sample data, and the first information entropy calculation function; Determine the second information entropy corresponding to the evaluation indicator according to at least one evaluation result corresponding to each of the to-be-processed bins under the evaluation indicator, the second sample quantity corresponding to the at least one evaluation result, the number of bin samples in the to-be-processed bins, and a second information entropy calculation function; Gain information of the evaluation indicator is determined according to the first information entropy and the second information entropy corresponding to the evaluation indicator.

4. The method according to claim 2, characterized in that: The step of determining the information value of the evaluation indicator according to the indicator data in at least one to-be-processed sub-box corresponding to the evaluation indicator comprises: For the at least one sub-box to be processed, determining at least two evaluation results corresponding to the indicator data in the sub-box to be processed; For any two evaluation results, the third sample quantity and the fourth sample quantity corresponding to the two evaluation results in the to-be-processed bins, and the fifth sample quantity and the sixth sample quantity corresponding to the two evaluation results under the evaluation index; Determine the bin value corresponding to the bin to be processed according to the third sample quantity, the fourth sample quantity, the fifth sample quantity and the sixth sample quantity corresponding to any two evaluation results in the bin to be processed; The information value of the evaluation indicator is determined according to the bin value of at least one to-be-processed bin associated with the evaluation indicator.

5. The method according to claim 4, characterized in that The determining the bin value corresponding to the bin to be processed according to the third sample quantity, the fourth sample quantity, the fifth sample quantity and the sixth sample quantity corresponding to all any two evaluation results in the bin to be processed comprises: For any two evaluation results, the first rating proportion is determined based on the third sample quantity and the fifth sample quantity of the two evaluation results, and the second rating proportion is determined based on the fourth sample quantity and the sixth sample quantity; Determine the rating values ​​corresponding to the two evaluation results according to the first rating proportion, the second rating proportion and the sample proportion function; The bin value of the bin to be processed is determined according to the rating values ​​corresponding to all any two evaluation results in the bin to be processed.

6. The method according to claim 5, characterized in that Determining the bin value of the bin to be processed according to the rating values ​​corresponding to all any two evaluation results in the bin to be processed includes: In the to-be-processed sub-box, rating difference information is determined according to the first rating proportion and the second rating proportion of the two evaluation results, and the to-be-superimposed values ​​of the two evaluation results are determined according to the rating difference information and the rating values ​​corresponding to the two evaluation results; The sub-box value of the sub-box to be processed is obtained by accumulating all the values ​​to be superimposed in the sub-box to be processed.

7. The method according to claim 2, characterized in that The step of determining at least two target bins corresponding to the evaluation indicators based on the gain information and information values ​​of the multiple evaluation indicators comprises: When the information value of the evaluation indicator is not within the preset value range, and when it is determined based on the gain information that the evaluation indicator is within the target range among all the evaluation indicators, at least one to-be-processed bin of the evaluation indicator is used as a target bin of the evaluation indicator; When the information value of the evaluation indicator is not within the preset value range, and when it is determined based on the gain information that the evaluation indicator is not within the target range among all the evaluation indicators, obtaining a bin value of at least one to-be-processed bin corresponding to the evaluation indicator; The at least one to-be-processed sub-bin is merged according to the sub-bin value to update the at least one to-be-processed sub-bin corresponding to the evaluation indicator, and the gain information and information value of the at least one to-be-processed sub-bin after the update are determined.

8. The method according to claim 2, characterized in that: After obtaining the target bin, the method further includes: For the plurality of sample data, according to the to-be-used data in the sample data, determining the bin to which the to-be-used data belongs under each evaluation indicator, and determining the characteristic value corresponding to the bin; Determine the sample characteristics of the sample data according to the characteristic value corresponding to each evaluation indicator in the data to be used under the corresponding bin; Based on the sample features and a predetermined actual evaluation result of the sample data, the sample data is updated to obtain the object evaluation model through training based on the updated sample data.

9. The method according to claim 1, characterized in that: The first dimension is the risk control dimension, the second dimension is the marketing dimension, and the target evaluation result is the first evaluation level of the object to be evaluated corresponding to the risk control dimension, and the second evaluation level of the object to be evaluated corresponding to the marketing dimension.

10. A device for determining object evaluation information, characterized in that include: A data acquisition module, used to acquire first data of the object to be evaluated in a first dimension and second data of the object to be evaluated in a second dimension, wherein the first dimension and the second dimension include multiple evaluation indicators; A target feature information determination module, configured to determine target feature information corresponding to the first data and the second data according to at least two predetermined target bins associated with the plurality of evaluation indicators; A target evaluation result determination module is used to input the target feature information into a pre-trained object evaluation model to determine the target evaluation result corresponding to the target object; The target evaluation result includes a first evaluation grade of the object to be evaluated under the first dimension and a second evaluation grade of the object to be evaluated under the second dimension.

Images