Label generation method and device for risk profile construction

Abstract

The present invention provides a label generation method and device for risk profile construction, which has the following characteristics: Steps S1-S3: dividing all features into multiple sub-feature sets, selecting features from the sub-feature sets based on importance values, correlation values, and feature screening amounts, and constructing corresponding screening feature sets; Step S4: for each sub-feature set, inputting the values ​​of all features corresponding to each target in the sub-feature set into a machine learning model, and combining the risk judgment label corresponding to the target to obtain the predicted value corresponding to each target in the sub-feature set; Step S5: for each sub-feature set, merging the features of the corresponding screening feature set and the predicted features to obtain the corresponding model feature set; Step S6: calculating the woe value of each feature corresponding to each target in the model feature set. In short, this method can generate multiple labels and corresponding label values ​​for each target to improve the accuracy of risk profiles.

Description

Technical Field

[0001] The present invention belongs to the field of risk profiling, and specifically relates to a label generation method and device for risk profiling construction. Background Art

[0002] Risk profiling is used in multiple fields, and the accuracy of risk profiling hinges on whether the risk label reflects the subject's risk status across a specific dimension. Existing risk label generation methods are primarily rule-driven. Rules are set based on the experience of domain experts and risk definitions, and logic or conditions are pre-defined based on these rules to generate risk labels. Rules can be based on a variety of criteria. For example, in the financial sector, criteria such as transaction frequency, spending amount, and abnormal account login behavior can be used as criteria. A rule might be set such as "If a user attempts to log in more than five times within 24 hours, label them as a 'frequently failed transaction account'." Rule-based label generation methods offer the advantages of strong interpretability and low implementation cost, but they are often linear and static. When risks exhibit nonlinear characteristics or multivariate dependencies, they struggle to capture complex risk characteristics, resulting in insufficient ability to characterize complex risk scenarios. Furthermore, rule-based methods rely too heavily on expert experience, resulting in insufficient timeliness and inability to mine complex risks.

[0003] In addition to rule-based approaches, some existing technologies use statistical methods or simple machine learning models to classify or score the risk of entities and generate risk labels. For example, methods such as logistic regression and decision trees can be used to perform preliminary assessments of individual risks. These methods can handle nonlinear relationships and can be applied to multidimensional data and large samples, thus having a certain degree of versatility. However, they also suffer from the following shortcomings: 1) Limited generalization of model labels: Current mainstream model label generation methods cannot fully capture complex risk characteristic patterns and are easily influenced by single-dimensional indicators, resulting in labels that are unable to effectively adapt to changing environments. This single-dimensional risk labeling system is unable to flexibly address risk differences across different entities and over time. 2) Lack of segmentation for different risk tiers: Traditional risk profiling methods often use single scores or classifications, lacking a fine-grained risk tier breakdown. Many scoring models lack a tiered labeling hierarchy, limiting the level of refinement in risk identification and failing to meet the requirements of refined risk control. 3) Lack of standardized support for model label generation methods: Currently, in the process of generating risk profiles, there is a lack of unified standards and methods for generating labels for different entities, resulting in poor reproducibility and interpretability of label generation. Most existing risk assessment methods are based on data-specific processing and cannot generalize label generation models across subjects, which affects their wide application. Summary of the Invention

[0004] The present invention is made to solve the above-mentioned problems, and its purpose is to provide a label generation method and device for risk profile construction.

[0005] The present invention provides a label generation method for risk profile construction, which is used to generate label results related to risk profiles corresponding to each target based on risk-related data. The risk-related data includes the values ​​of multiple different features corresponding to each target, and the risk judgment label corresponding to each target, which has such characteristics, including the following steps: Step S1, divide all features into multiple categories, and generate sub-feature sets corresponding to each category, the sub-feature sets including at least one feature and all values ​​corresponding to the feature; Step S2, for each sub-feature set, according to the number of features in the sub-feature set and the values ​​corresponding to the features, calculate the importance value and correlation value of each feature and the feature screening amount; Step S3, for each sub-feature set, according to the importance value, correlation value and feature screening amount, select features from the sub-feature set to construct the corresponding screening feature set Combine; Step S4, for each sub-feature set, input the values ​​of all features corresponding to each target in the sub-feature set into the machine learning model, and combine the risk judgment label corresponding to the target to obtain the predicted value corresponding to each target in the sub-feature set as the value of the predicted feature corresponding to the target in the sub-feature set; Step S5, for each sub-feature set, merge the features and prediction features of the corresponding screening feature set to obtain the corresponding model feature set; Step S6, for each model feature set, perform binning calculation to obtain the bin value of each feature corresponding to each target in the model feature set, and calculate the woe value of each feature corresponding to each target in the model feature set based on all the bin values ​​and risk judgment labels corresponding to the model feature set, wherein the label result of the target is each feature of the target in each model feature set and its corresponding woe value.

[0006] The label generation method for risk profile construction provided by the present invention may also have the following features: wherein, in step S4, the predicted value is obtained by the LightGBM model, and its calculation expression is: In the formula is the predicted value corresponding to target k in sub-feature set j, is a sample consisting of the values ​​of all features corresponding to target k in sub-feature set j, The t-th tree pair sample of the LightGBM model The prediction results are: η is the learning rate, T is the total number of trees in the LightGBM model, and σ is the Sigmoid function.

[0007] The label generation method for risk profile construction provided by the present invention may also have the following feature: wherein, in step S2, the calculation expression of the correlation value is: In the formula is the correlation value of feature i in sub-feature set j, is the value of feature i corresponding to the kth target in sub-feature set j, is the mean of feature i in sub-feature set j, y k is the risk judgment label corresponding to the k-th target, is the mean of the risk judgment labels of all targets, and n is the total number of targets.

[0008] The label generation method for risk profile construction provided by the present invention may also have the following characteristics: wherein, in step S2, the corresponding importance value is calculated based on the information gain of the feature at the node of the tree of the LightGBM model, and the calculation expression of the importance value is: In the formula is the importance value of feature i in sub-feature set j, p is the total number of features in sub-feature set j, M is the total number of trees in the LightGBM model, N m is the set of nodes of the mth tree, is the indicator function, ΔL m,n is the information gain of feature i on node n of the mth tree. When node n is split using feature i, When node n does not use feature i for splitting,

[0009] The label generation method for risk profile construction provided by the present invention may also have the following characteristics: wherein, in step S2, the calculation expression of the feature screening amount is: Where select_len j is the feature screening amount corresponding to sub-feature set j, p is the total number of features in sub-feature set j, This is a floor operation.

[0010] The label generation method for risk profile construction provided by the present invention may also have the following features: wherein, in step S3, the calculation expression for screening the feature set is: X j,selected =X j,corr ∪X j,lgbm , where X j,selected is the filtered feature set corresponding to sub-feature set j, X j,corr Filter the feature set corresponding to the correlation of sub-feature set j, X j ,lgbmFor the importance screening feature set corresponding to the sub-feature set j, all features of the sub-feature set j are sorted from large to small according to the correlation value, and features of the number of feature screening amounts are selected from the beginning of the sequence to construct a correlation screening feature set. All features of the sub-feature set j are sorted from large to small according to the importance value, and features of the number of feature screening amounts are selected from the beginning of the sequence to construct an importance screening feature set.

[0011] In the label generation method for risk profile construction provided by the present invention, it can also have the following characteristics: wherein, step S6 includes the following sub-steps: step S6-1, generating multiple candidate splitting thresholds for the model feature set, and calculating the splitting values ​​corresponding to each candidate splitting threshold; step S6-2, taking the candidate splitting threshold corresponding to the largest splitting value as the optimal splitting point; step S6-3, repeating steps S6-1 to S6-2 until the number of optimal splitting points reaches a preset value; step S6-4, calculating the bin boundaries based on all optimal splitting points; step S6-5, calculating the bins value and bin0 value of each feature corresponding to each target in the model feature set as the bin value based on the bin boundaries; step S6-6, calculating the woe value of each feature corresponding to each target in the model feature set based on all bin values.

[0012] The label generation method for risk profile construction provided by the present invention may also have the following characteristics: wherein the calculation expression of the split value is: H(D j )=-Σ C∈C p(c)log2p(c), Where IG(D j ,θ) is the sub-feature set D j The corresponding candidate splitting threshold θ split value, |D| is the number of samples in D, is the sub-feature set D j The value of feature i corresponding to target k, v k For the target k, y k is the risk judgment label corresponding to target k, C is the sub-feature set D j The total number of features in the set D, p(c) is the number of features c in the sub-feature set D j The probability of , the calculation expression of the bin boundary is: In the formula for The set of bin boundary values, To sort all optimal split points in ascending order, the calculation expressions for bins value and bin0 value are: In the formula is the sub-feature set D j The value of feature i corresponding to target k in for The left boundary value of the u-th bin interval.

[0013] The label generation method for risk profile construction provided by the present invention may also have the following features: wherein, in step S6-6, the values ​​of the features corresponding to all targets are divided into multiple groups according to the bin0 values ​​of the features corresponding to the targets, and the values ​​of the features corresponding to all targets in the group have the same bin0 value, the risk judgment label is a binary classification label, and the calculation expression of the woe value is: In the formula For target k in sub-feature set D j The woe value of the corresponding feature i, N bad For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a bad category, N good For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a good category, N total For target k in sub-feature set D j The total number of values ​​of feature i corresponding to .

[0014] The present invention also provides a label generation device for risk profile construction, which is used to generate label results related to the risk profile corresponding to each target based on risk-related data, wherein the risk-related data includes the values ​​of multiple different features corresponding to each target, and the risk judgment label corresponding to each target, and has the following characteristics, including: a subset division module, which divides all features into multiple categories and generates sub-feature sets corresponding to each category, wherein the sub-feature sets include at least one feature and all values ​​corresponding to the feature; a screening module, which calculates the importance value and correlation value of each feature and the feature screening amount based on the number of features in the sub-feature set and the values ​​corresponding to the features; a screening set generation module, which selects features from the sub-feature set based on the importance value, correlation value and feature screening amount for each sub-feature set to construct a corresponding screening feature set; The prediction feature generation module inputs the values ​​of all features corresponding to each target in the sub-feature set into the machine learning model for each sub-feature set, and combines the risk judgment label corresponding to the target to obtain the prediction value corresponding to each target in the sub-feature set as the value of the prediction feature corresponding to the target in the sub-feature set; the collection module merges the features and prediction features of the corresponding screening feature set for each sub-feature set to obtain the corresponding model feature set; the woe calculation module performs binning calculation on each model feature set to obtain the bin value of each feature corresponding to each target in the model feature set, and calculates the woe value of each feature corresponding to each target in the model feature set based on all the bin values ​​and risk judgment labels corresponding to the model feature set, where the label result of the target is each feature of the target in each model feature set and its corresponding woe value.

[0015] Functions and effects of the invention

[0016] According to the label generation method and device for risk profile construction involved in the present invention, by dividing the features into sub-feature sets and calculating the corresponding importance values ​​and correlation values ​​and feature screening amounts, multiple features are screened from each sub-feature set, and then the prediction features are calculated to generate the corresponding model feature set. The woe value of each target is calculated for the features in each model feature set to obtain the label result. Therefore, the label generation method and device for risk profile construction of the present invention can generate multiple labels and corresponding label values ​​for each target that improve the accuracy of risk profiles. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a block diagram of a label generation device for risk profile construction in an embodiment of the present invention;

[0018] Figure 2 1 is a flow chart of the woe calculation module generating the woe value in an embodiment of the present invention;

[0019] Figure 3 is a flow chart of a label generation method for risk profile construction in an embodiment of the present invention;

[0020] Figure 4 is a radar schematic diagram of a high-risk user profile in an embodiment of the present invention;

[0021] Figure 5 is a radar schematic diagram of a low-risk user profile in an embodiment of the present invention;

[0022] Figure 6 Schematic diagram of visualization of a scorecard in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following embodiments and the accompanying drawings specifically illustrate the label generation method and device for risk profile construction of the present invention.

[0024] This embodiment provides a label generation device for risk profile construction, which generates risk profile-related label results for each target based on risk-related data. The risk-related data includes the values ​​of multiple different features corresponding to each target, as well as a risk assessment label corresponding to each target. The risk assessment label is a binary classification label.

[0025] In this embodiment, the label generation device for risk profile construction is applied to the financial field, and the risk-related data is financial transaction data D = [VXY]. Wherein, V = {v1, ..., v n} represents a set of trading accounts, each trading account is a target, for each v i ∈V has a binary label whether it is illegal, that is, the line judgment label y i ∈Y, n is the number of trading accounts, and the feature space constructed from the perspectives of statistical calculation, complex rules, expert knowledge, and intelligent algorithms is X, where X={X1,X2,…,X d}, X i ={χ i,k |k=1,2,…,n}, represents the feature χ i The feature set of i,k Indicates the feature χ corresponding to the kth trading account i In other embodiments, the label generation device for risk profile construction can be applied to other fields, such as food safety, where the target is a specified food and the risk-related data is production data related to the specified food, such as the manufacturer.

[0026] Figure 1It is a block diagram of a label generation device for risk profile construction in an embodiment of the present invention.

[0027] like Figure 1 As shown, the label generation device 100 for risk profile construction includes a subset division module 11, a screening module 12, a screening set generation module 13, a prediction feature generation module 14, a collection module 15, a woe calculation module 16 and a control module 17 for controlling the operation of the above modules.

[0028] The subset division module 11 is used to divide all features into multiple categories and generate a sub-feature set corresponding to each category. The sub-feature set includes at least one feature and all values ​​corresponding to the feature.

[0029] In this embodiment, the subset division module 11 divides the information into m categories, that is, the sub-feature set X is obtained. 1 ,X 2 ,…,X m , and satisfies

[0030] The screening module 12 is used to calculate the importance value and relevance value of each feature and the feature screening amount for each sub-feature set according to the number of features in the sub-feature set and the values ​​corresponding to the features.

[0031] The calculation expression of the correlation value is:

[0032]

[0033] In the formula is the correlation value of feature i in sub-feature set j, is the value of feature i corresponding to the kth target in sub-feature set j, is the mean of feature i in sub-feature set j, y k is the risk judgment label corresponding to the k-th target, is the mean of the risk judgment labels of all targets, and n is the total number of targets.

[0034] Among them, the screening module 12 calculates the corresponding importance value based on the information gain of the feature on the node of the tree of the LightGBM model. The calculation expression of the importance value is:

[0035]

[0036] In the formula is the importance value of feature i in sub-feature set j, p is the total number of features in sub-feature set j, M is the total number of trees in the LightGBM model, N m is the set of nodes of the mth tree, is the indicator function, ΔLm,n The information gain is the loss reduction when feature i is used as the splitting loss at node n of the mth tree. When node n is split using feature i, When node n does not use feature i for splitting,

[0037] The loss function corresponding to the LightGBM model in this embodiment is the binary cross entropy loss L, and its calculation expression is:

[0038]

[0039] Where N is the total number of samples, y i is the risk judgment label of the i-th sample, is the predicted probability of the LightGBM model for the i-th sample.

[0040] Among them, the calculation expression of the feature screening amount is:

[0041]

[0042] Where select_len j is the feature screening amount corresponding to sub-feature set j, p is the total number of features in sub-feature set j, This is a floor operation.

[0043] The screening set generation module 13 is used to select features from each sub-feature set according to the importance value, the relevance value and the feature screening amount, and construct a corresponding screening feature set.

[0044] Among them, the calculation expression for filtering feature sets is:

[0045] X j,selected =X j,corr ∪X j,lgbm ,

[0046] Where X j,selected is the filtered feature set corresponding to sub-feature set j, X j,corr Filter the feature set corresponding to the correlation of sub-feature set j, X j,lgbm Filter the feature set for the importance corresponding to the sub-feature set j.

[0047] Sort all the features of sub-feature set j from large to small according to the correlation value, select the number of features of feature screening from the beginning of the sequence, and construct the correlation screening feature set. Correlation screening feature set X j,corr The calculation expression is:

[0048] X j,corr =argsort(|correlationsj |)[:select_len j ],

[0049]

[0050] In the formula, argsort(|correlatons j |) is to sort the correlation values ​​of the sub-feature set j in descending order according to the absolute value, that is, in this embodiment, the correlation values ​​are sorted according to their absolute values, [:select_len j ] is the feature screening amount corresponding to the sub-feature set j.

[0051] Sort all the features of sub-feature set j from large to small according to their importance values, select the number of features to be screened from the beginning of the sequence, and construct an importance screening feature set. Importance screening feature set X j,lgbm The calculation expression is:

[0052] X j,lgbm =argsort(|lgbm_importance j |)[:select_len j ],

[0053]

[0054] In the formula, argsort(|lgbm_importance j |) is to sort the importance values ​​of sub-feature set j in descending order according to absolute value.

[0055] The prediction feature generation module 14 is configured to input the values ​​of all features corresponding to each target in each sub-feature set into a machine learning model, combine the values ​​with the risk assessment label corresponding to the target, and obtain the predicted value corresponding to each target in the sub-feature set as the value of the prediction feature corresponding to the target in the sub-feature set. In this embodiment, the risk assessment label serves as the target variable, i.e., the label, of the machine learning model.

[0056] The machine learning model is the LightGBM model, and the prediction feature generation module 14 obtains the prediction value through the LightGBM model. The calculation expression is:

[0057]

[0058] In the formula is the predicted value corresponding to target k in sub-feature set j, is a sample consisting of the values ​​of all features corresponding to target k in sub-feature set j, The t-th tree pair sample of the LightGBM model The prediction results are divided based on the decision tree structure to map the input features to the value of a leaf node. η is the learning rate used to scale the output of each tree. T is the total number of trees in the LightGBM model. σ is the Sigmoid function used to convert the output into a probability value:

[0059] The collection module 15 is used to combine the features of the corresponding screening feature set and the prediction features of each sub-feature set to obtain a corresponding model feature set.

[0060] The calculation expression of the model feature set in this embodiment is:

[0061]

[0062] In the formula is the model feature set corresponding to the sub-feature set j, and || is the concatenation operation. For example, to filter the feature set X j,selected There are 4 categories of features in , then the model feature set There are 5 categories of features, that is, the screening feature set X j,selected Features of the four categories and predicted features.

[0063] The Woe calculation module 16 is used to perform binning calculations on each model feature set to obtain the bin values ​​of each feature corresponding to each target in the model feature set, and calculate the Woe value of each feature corresponding to each target in the model feature set based on all the bin values ​​corresponding to the model feature set and the risk judgment label. The label result of the target is the various features of the target in each model feature set and their corresponding Woe values.

[0064] Figure 2 It is a schematic diagram of the process of generating a woe value by a woe calculation module in an embodiment of the present invention.

[0065] like Figure 2 As shown, the process of generating the woe value by the woe calculation module 16 includes the following steps:

[0066] Step S6-1: Generate multiple candidate splitting thresholds for the model feature set, and calculate the splitting value corresponding to each candidate splitting threshold. In this embodiment, the candidate splitting threshold is obtained by the average value of the sorted adjacent samples.

[0067] The calculation expression of the split value is:

[0068]

[0069] H(D j )=-Σc∈C p(c)log2p(c),

[0070]

[0071] Where IG(D j ,θ) is the sub-feature set D j The corresponding splitting value of the candidate splitting threshold θ, |D| is the number of samples in D, and a sample includes the values ​​of all features corresponding to a target in the set D, is the sub-feature set D j The value of feature i corresponding to target k, v k For the target k, y k is the risk judgment label corresponding to target k, C is the sub-feature set D j The total number of features in the set D, p(c) is the number of features c in the sub-feature set D j probability.

[0072] Step S6-2: The candidate splitting threshold corresponding to the maximum splitting value is used as the optimal splitting point.

[0073] Step S6-3: Repeat steps S6-1 to S6-2 until the number of optimal splitting points reaches a preset value. In this embodiment, at most four optimal splitting points are generated by iterating two layers.

[0074] Step S6-4: Calculate the bin boundaries based on all the optimal splitting points.

[0075] The calculation expression of the bin boundary is:

[0076]

[0077] In the formula for The set of bin boundary values, In order to sort all optimal splitting points in ascending order, in this embodiment, negative infinity -∞ and positive infinity ∞ are used to ensure that the final bin boundaries are such that all values ​​can be covered.

[0078] In step S6-5, based on the binning boundaries, the bins value and bin0 value of each feature corresponding to each target in the model feature set are calculated as the binning value.

[0079] Among them, the calculation expressions of bins value and bin0 value are:

[0080]

[0081] In the formula is the sub-feature set D j The value of feature i corresponding to target k in , for The left boundary value of the u-th bin interval.

[0082] Step S6-6, based on all bin values, calculate the woe value of each feature corresponding to each target in the model feature set.

[0083] Among them, according to the bin0 value of the feature corresponding to the target, the values ​​of the features corresponding to all targets are divided into multiple groups The values ​​of the features corresponding to all targets in the group have the same bin0 value, that is, if The target corresponding feature

[0084] The calculation expression of woe value is:

[0085]

[0086] In the formula For target k in sub-feature set D j The woe value of the corresponding feature i, N bad For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a bad category, N good For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a good category, N total For target k in sub-feature set D j The total number of values ​​of the corresponding feature j in . In this embodiment, the risk judgment label is 0 or 1, 0 represents a good category, and 1 represents a bad category. In other embodiments, other forms can be used to replace 0 and 1 to distinguish between good and bad categories.

[0087] The control module 17 stores a control program for controlling the operation of each module.

[0088] The following describes the process of performing a label generation method for risk profile construction using the label generation device 100 for risk profile construction in conjunction with the accompanying drawings.

[0089] Figure 3 It is a flowchart of a label generation method for risk profile construction in an embodiment of the present invention.

[0090] like Figure 3 As shown in Figure 2, the label generation method for risk profile construction includes the following steps:

[0091] In step S1 , all features are divided into multiple categories using the subset division module 11 , and a sub-feature set corresponding to each category is generated. The sub-feature set includes at least one feature and all values ​​corresponding to the feature.

[0092] In step S2, the screening module 12 is used to calculate the importance value and the relevance value of each feature and the feature screening amount for each sub-feature set according to the number of features in the sub-feature set and the values ​​corresponding to the features.

[0093] In step S3, the screening set generation module 13 selects features from the sub-feature sets according to the importance value, the relevance value and the feature screening amount, and constructs a corresponding screening feature set.

[0094] In step S4, the prediction feature generation module 14 is used to input the values ​​of all features corresponding to each target in the sub-feature set into the machine learning model for each sub-feature set, and combined with the risk judgment label corresponding to the target, the prediction value corresponding to each target in the sub-feature set is obtained as the value of the prediction feature corresponding to the target in the sub-feature set.

[0095] In step S5, the set module 15 is used to combine the features of the corresponding screening feature set and the prediction features of each sub-feature set to obtain a corresponding model feature set.

[0096] In step S6, the Woe calculation module 16 is used to perform binning calculations on each model feature set to obtain the bin values ​​of each feature corresponding to each target in the model feature set, and the Woe values ​​of each feature corresponding to each target in the model feature set are calculated based on all the bin values ​​and risk judgment labels corresponding to the model feature set.

[0097] In this embodiment, a radar chart of the user portrait of the target can be constructed based on the label results of the target.

[0098] Figure 4 Schematic diagram of a radar of a high-risk user profile in an embodiment of the present invention.

[0099] Figure 5 Schematic diagram of a radar of a low-risk user profile in an embodiment of the present invention.

[0100] like Figure 4 and Figure 5As shown in the figure, the labeling result includes eight features and their corresponding values: m_woe_1, m_woe_2, m_woe_3, m_woe_4, m_woe_5, m_woe_6, m_woe_7, and m_woe_8. This shows that abnormal subjects (high-risk users) occupy a large area, while normal subjects (low-risk users) occupy a small area. This intuitively demonstrates that the labeling result accurately depicts the multi-dimensional risk hierarchy of the subject.

[0101] In this embodiment, the label generation method used for risk profile construction generates the label result, namely the model label X m_woe , with general statistics optimization label X anno Conduct a comparative experiment on the scorecard model, which generally optimizes the label X anno For a financial institution, internal expert rule-driven labeling is used to build a scoring model SCM based on logistic regression. m_woe and general statistics optimization tab X anno The training data set is used for training, and the trained models are used for binary classification prediction. The specific indicator results are shown in the following table:

[0102] KS AUC <![CDATA[f10]]> <![CDATA[Recall1]]> <![CDATA[f11 <!-- 9 -->]]> <![CDATA[SCM+X anno ]]> 0.7454 0.9533 0.9657 0.5707 0.6782 <![CDATA[SCM+X m_woe ]]> 0.7744 0.9614 0.9692 0.6272 0.7210

[0103] The first column in the table above shows the models trained with different training sets. The second to sixth columns show the KS index, AUC index, f10 index, Recall1 index and f11 index of the models. For example, the cell in the second row and second column shows the general statistical optimization label X. anno The KS index of the scoring model SCM trained on the constructed training set is 0.7454. m_woe The classification ability of the scoring model SCM is further improved under the data representation of m_woe The anomaly detection effect of the scoring model SCM is enhanced.

[0104] Based on the scoring model (SCM), this example extracts feature coefficients and intercepts, sets a baseline score and a PDO for each doubling of odds, calculates factor weights and offsets, and creates a scorecard containing features, coefficients, sample scores, and predicted probabilities. The scores are divided into several equally spaced scoring intervals, and the number and ratio of positive and negative samples within each scoring interval are calculated, providing a direct reflection of the distribution of positive and negative samples within each scoring interval.

[0105] Figure 6 Schematic diagram of visualization of a scorecard in an embodiment of the present invention.

[0106] like Figure 6As shown in the figure, the horizontal axis is the sample score interval value score_bins, the vertical axis of the histogram is the number of samples in the score interval Count, and the vertical axis of the line graph is the bad sample ratio Bad Rate in the score interval. m_woe The scorecard intuitively shows how each module distinguishes between positive and negative, achieving a strong correlation between the abnormal subject and the risk score.

[0107] In this example, some key model labels are selected, grouped using bin values, and the number and proportion of bad samples are calculated within the group. The results are shown in the following table:

[0108]

[0109] The first column in the table above is the different model labels χ m_woe1 , χ lm_woe2 and χ m_woe3 The second to sixth columns are the woe value, bin range, number of samples in the bad category, total number of samples, and the proportion of samples in the bad category. It can be seen that in the given example key model label grouping, the proportion of bad samples is more than 90%, showing that the model label X m_woe Accurate expression of bad samples in specific dimensions.

[0110] Functions and Effects of the Embodiments

[0111] According to the label generation method and device for risk profiling involved in this embodiment, by dividing the features into sub-feature sets and calculating the corresponding importance values ​​and correlation values ​​as well as the feature screening amount, multiple features are screened from each sub-feature set, and then the prediction features are calculated to generate the corresponding model feature set. The Woe value of each target is calculated for the features in each model feature set to obtain the label result. In short, this method can effectively characterize nonlinear and multivariate relationships, better express the multidimensional risk status of the subject, achieve fine-grained risk hierarchy division, improve the accuracy of risk profiling, and enhance the ability to mine complex risks.

[0112] Those skilled in the art will appreciate that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A label generation method for risk profile construction, for generating risk profile-related label results corresponding to each target based on risk-related data, wherein the risk-related data includes the values ​​of multiple different features corresponding to each target and the risk judgment label corresponding to each target, characterized in that: The following steps are involved: Step S1: Divide all the features into multiple categories, and generate a sub-feature set corresponding to each category, wherein the sub-feature set includes at least one feature and all the values ​​corresponding to the feature; Step S2: for each of the sub-feature sets, calculating the importance value and the relevance value of each feature and the feature screening amount according to the number of features in the sub-feature set and the values ​​corresponding to the features; Step S3, for each of the sub-feature sets, selecting the feature from the sub-feature set according to the importance value, the relevance value, and the feature screening amount, to construct a corresponding screening feature set; Step S4: For each of the sub-feature sets, the values ​​of all the features corresponding to each of the targets in the sub-feature set are input into a machine learning model, and combined with the risk judgment label corresponding to the target, a predicted value corresponding to each of the targets in the sub-feature set is obtained as the value of the predicted feature corresponding to the target in the sub-feature set; Step S5: for each of the sub-feature sets, merging the features of the corresponding screening feature set and the prediction features to obtain a corresponding model feature set; Step S6: perform binning calculation on each of the model feature sets to obtain the bin values ​​of each of the features corresponding to each of the targets in the model feature set, and calculate the woe value of each of the features corresponding to each of the targets in the model feature set based on all the bin values ​​corresponding to the model feature set and the risk judgment label. The label result of the target is each feature of the target in each model feature set and its corresponding woe value. The label generation method for risk profile construction is applied in the financial field. The risk-related data is financial transaction data D = [VXY], where V = {v1, ..., v n } represents a set of trading accounts, each trading account is a target, for each v i ∈V has a binary label whether it is illegal, that is, the line judgment label y i ∈Y, n is the number of trading accounts, and the feature space is X, where X={X1,X2,…,X d }, X i ={χ i,k |k=1,2,…,n}, represents feature x i The feature set of x i,k Indicates the feature x corresponding to the kth trading account i The value of .

2. The label generation method for risk profile construction according to claim 1, characterized in that: in, In step S4, the predicted value is obtained by the LightGBM model, and its calculation expression is: In the formula is the predicted value corresponding to target k in sub-feature set j, is a sample consisting of the values ​​of all features corresponding to target k in sub-feature set j, The t-th tree pair sample of the LightGBM model The prediction results are: η is the learning rate, T is the total number of trees in the LightGBM model, and σ is the Sigmoid function.

3. The label generation method for risk profile construction according to claim 1, characterized in that: in, In step S2, the calculation expression of the correlation value is: In the formula is the correlation value of feature i in sub-feature set j, is the value of feature i corresponding to the kth target in sub-feature set j, is the mean of feature i in sub-feature set j, y k is the risk judgment label corresponding to the k-th target, is the mean of the risk judgment labels of all targets, and n is the total number of targets.

4. The label generation method for risk profile construction according to claim 1, characterized in that: in, In step S2, the corresponding importance value is calculated based on the information gain of the feature at the node of the tree of the LightGBM model. The calculation expression of the importance value is: In the formula is the importance value of feature i in sub-feature set j, p is the total number of features in sub-feature set j, M is the total number of trees in the LightGBM model, N m is the set of nodes of the mth tree, is the indicator function, ΔL m,n is the information gain of feature i at node n of the mth tree, When node n is split using feature i, When node n does not use feature i for splitting, 5. The label generation method for risk profile construction according to claim 1, characterized in that: in, In step S2, the calculation expression of the characteristic screening amount is: Where select_len j is the feature screening amount corresponding to sub-feature set j, p is the total number of features in sub-feature set j, This is a floor operation.

6. The label generation method for risk profile construction according to claim 1, characterized in that: in, In step S3, the calculation expression of the screening feature set is: X j , selected =X j,corr ∪X j,lgbm , Where X j,selected is the filtered feature set corresponding to sub-feature set j, X j,corr Filter the feature set corresponding to the correlation of sub-feature set j, X j,lgbm Filter the feature set for the importance of sub-feature set j, Sort all the features of sub-feature set j from large to small according to the correlation value, select the features of the feature screening quantity from the beginning of the sequence, and construct the correlation screening feature set, All the features of the sub-feature set j are sorted from large to small according to the importance value, and the features of the feature screening quantity are selected from the beginning of the sequence to construct the importance screening feature set.

7. The label generation method for risk profile construction according to claim 1, Its characteristics are: Wherein, the step S6 includes the following sub-steps: Step S6-1, generating multiple candidate splitting thresholds for the model feature set, and calculating the splitting value corresponding to each candidate splitting threshold; Step S6-2, taking the candidate splitting threshold corresponding to the largest splitting value as the optimal splitting point; Step S6-3, repeating steps S6-1 to S6-2 until the number of the optimal splitting points reaches a preset value; Step S6-4, calculating the bin boundaries based on all the optimal splitting points; Step S6-5, calculating, based on the binning boundaries, the bins value and bin0 value of each of the features corresponding to each of the targets in the model feature set as the binning values; Step S6-6: Calculate the woe value of each feature corresponding to each target in the model feature set based on all the bin values.

8. The label generation method for risk profile construction according to claim 7, characterized in that: in, The calculation expression of the split value is: H(D j )=-∑ c∈C p(c)log2p(c), Where IG(D j ,θ) is the sub-feature set D j The corresponding candidate splitting threshold θ split value, |D| is the number of samples in D, is the sub-feature set D j The value of feature i corresponding to target k, v k For the target k, y k is the risk judgment label corresponding to target k, C is the sub-feature set D j The total number of features in the set D, p(c) is the number of features c in the sub-feature set D j The probability of The calculation expression of the bin boundary is: In the formula for The set of bin boundary values, To sort all optimal splitting points in ascending order, The calculation expressions of the bins value and the bin0 value are: In the formula is the sub-feature set D j The value of feature i corresponding to target k in for The left boundary value of the u-th bin interval.

9. The label generation method for risk profile construction according to claim 7, characterized in that: in, In step S6-6, the values ​​of the features corresponding to all the targets are divided into multiple groups according to the bin0 values ​​of the features corresponding to the targets. The values ​​of the features corresponding to all the targets in the group have the same bin0 value, The risk judgment label is a binary classification label. The calculation expression of the woe value is: In the formula For target k in sub-feature set D j The woe value of the corresponding feature i, N bad For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a bad category, N good For target k in sub-feature set D j The total number of risk judgment labels corresponding to all values ​​in the group where the corresponding feature i value is a good category, N total For target l in sub-feature set D j The total number of values ​​of feature i corresponding to .

10. A label generation device for risk profile construction, used to generate label results related to risk profiles for each target based on risk-related data, wherein the risk-related data includes the values ​​of multiple different features corresponding to each target and the risk judgment label corresponding to each target, characterized in that: include: a subset division module, dividing all the features into multiple categories, generating a sub-feature set corresponding to each category, wherein the sub-feature set includes at least one feature and all the values ​​corresponding to the feature; A screening module, for each of the sub-feature sets, calculates an importance value and a relevance value of each of the features and a feature screening amount based on the number of features in the sub-feature set and the values ​​corresponding to the features; a screening set generating module, for each of the sub-feature sets, selecting the features from the sub-feature sets according to the importance value, the relevance value and the feature screening amount, to construct a corresponding screening feature set; A prediction feature generation module, for each of the sub-feature sets, inputs the values ​​of all the features corresponding to each of the targets in the sub-feature set into a machine learning model, combines the risk judgment label corresponding to the target, and obtains the prediction value corresponding to each of the targets in the sub-feature set as the value of the prediction feature corresponding to the target in the sub-feature set; A collection module, for each of the sub-feature sets, combines the features of the corresponding screening feature set and the prediction features to obtain a corresponding model feature set; The woe calculation module performs binning calculation on each of the model feature sets to obtain the bin values ​​of each of the features corresponding to each of the targets in the model feature set, and calculates the woe values ​​of each of the features corresponding to each of the targets in the model feature set based on all the bin values ​​corresponding to the model feature set and the risk judgment label. The label result of the target is each feature of the target in each model feature set and its corresponding woe value. The label generation device for risk profile construction is applied in the financial field, and the risk-related data is financial transaction data D=[VXY], where V={v1,...,v n } represents a set of trading accounts, each trading account is a target, for each v i ∈V has a binary label whether it is illegal, that is, the line judgment label y i ∈Y, n is the number of trading accounts, and the feature space is X, where X={X1,X2,…,X d }, X i ={χ i,k |k=1,2,…,n}, represents feature x i The feature set of x i,k Indicates the feature x corresponding to the kth trading account i The value of .

Images