Supplier risk level assessment method, device, equipment, medium and program product

Abstract

The invention provides a supplier risk level assessment method which can be applied to the technical field of artificial intelligence. The supplier risk level assessment method comprises the steps that in response to a supplier risk level assessment instruction, supplier risk data to be evaluated are collected according to target features, and the target features are determined based on historical assessment risk level results and Pearson's correlation coefficients; and inputting the supplier risk to-be-evaluated data into a risk prediction model to output a supplier risk level, the risk prediction model being obtained by pre-training based on an ensemble learning algorithm. The invention further provides a supplier risk level assessment device and equipment, a storage medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the field of integrated learning technology, and more specifically to a supplier risk level assessment method, device, equipment, medium and program product. Background Art

[0002] Large and medium-sized Internet companies and financial software R&D institutions in the industry often recruit outsourced personnel to help the company complete some R&D tasks. When it comes to outsourced personnel suppliers, the supplier's financial status, personnel supply capacity and other aspects are crucial. Companies regularly conduct risk assessments on outsourced suppliers to help identify potential outsourcing risks and take preventive measures in advance. If there are risk issues, the cooperation can be terminated in advance to avoid economic losses or reputational impacts due to improper behavior of suppliers.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a supplier risk level assessment method, apparatus, device, medium and program product.

[0005] According to a first aspect of the present disclosure, a method for assessing supplier risk level is provided, comprising:

[0006] In response to a supplier risk level assessment instruction, collecting supplier risk data to be assessed according to target characteristics, wherein the target characteristics are determined based on historical assessment risk level results and a Pearson correlation coefficient; and

[0007] The supplier risk data to be assessed is input into a risk prediction model to output the supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

[0008] According to an embodiment of the present disclosure, determining the target feature based on the historical risk level assessment results and the Pearson correlation coefficient includes:

[0009] Clean the collected supplier risk data and extract data feature sets;

[0010] The data feature set is screened based on historical assessment risk level results and the Pearson coefficient to determine the target feature.

[0011] According to an embodiment of the present disclosure, querying distributed inventory data according to user information and activity identifiers includes:

[0012] Determine the activity resource database according to the activity identifier;

[0013] Determine the corresponding database shard after performing a hash operation on the user information; and

[0014] Query database shard data.

[0015] According to an embodiment of the present disclosure, the data feature set is screened based on the historical risk level assessment results and the Pearson coefficient to determine the target feature, including:

[0016] Convert the data feature set into a data value set based on the one-hot encoding technology;

[0017] Calculate the Pearson coefficient between each data feature and risk level; and

[0018] The target feature is determined according to the Pearson coefficient.

[0019] According to an embodiment of the present disclosure, the risk prediction model obtained by pre-training based on an ensemble learning algorithm includes:

[0020] Generate model training data based on the target features;

[0021] The model training data is divided into layers according to the risk level type;

[0022] Select multiple ensemble learning algorithms for prediction training;

[0023] Call the natural rhythm optimization algorithm for iterative optimization to determine the optimal parameter combination of the model; and

[0024] The risk prediction model is determined based on the prediction accuracy of each algorithm.

[0025] According to an embodiment of the present disclosure, the ensemble learning algorithm includes an XGBoost algorithm, an AdaBoosting algorithm and a GDBT algorithm.

[0026] According to an embodiment of the present disclosure, inputting the supplier risk data to be assessed into a risk prediction model to output the supplier risk level includes:

[0027] Calculate the importance of each target feature;

[0028] Sort by feature importance and obtain the corresponding feature index; and

[0029] Output the names of suppliers with medium, high or above risks, risk prediction results and key target characteristics.

[0030] A second aspect of the present disclosure provides a supplier risk level assessment device, comprising:

[0031] A collection module, configured to respond to a supplier risk level assessment instruction and collect supplier risk data to be assessed according to a target feature, wherein the target feature is determined based on a historical assessment risk level result and a Pearson correlation coefficient; and

[0032] The risk assessment module is used to input the supplier risk data to be assessed into a risk prediction model to output the supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

[0033] According to an embodiment of the present disclosure, a target feature determination module is also included.

[0034] The target feature determination module is used to determine the target feature based on the historical assessment risk level results and the Pearson correlation coefficient.

[0035] According to an embodiment of the present disclosure, the target feature determination module includes a data extraction submodule and a first determination submodule.

[0036] The data extraction submodule is used to clean the collected supplier risk data and extract data feature sets;

[0037] The first determination submodule is used to screen the data feature set based on historical risk level assessment results and the Pearson coefficient to determine the target feature.

[0038] According to an embodiment of the present disclosure, the first determining submodule includes a data conversion unit, a calculation unit and a determination unit.

[0039] A data conversion unit, used to convert a data feature set into a data value set based on a one-hot encoding technique;

[0040] A calculation unit, used to calculate the Pearson coefficient between each data feature and the risk level; and

[0041] A determination unit is used to determine the target feature according to the Pearson coefficient.

[0042] According to an embodiment of the present disclosure, a risk prediction model training module is also included.

[0043] The risk prediction model training module includes: a generation submodule, a splitting submodule, a training submodule, a parameter optimization submodule and a second determination submodule.

[0044] A generating submodule, used for generating model training data based on the target features;

[0045] A splitting submodule, used for splitting the model training data into layers according to risk level types;

[0046] The training submodule is used to select multiple ensemble learning algorithms for prediction training;

[0047] The parameter optimization submodule is used to call the natural rhythm optimization algorithm for iterative optimization to determine the optimal parameter combination of the model; and

[0048] The second determination submodule is used to determine the risk prediction model according to the prediction accuracy of each algorithm.

[0049] According to an embodiment of the present disclosure, the risk assessment module includes a calculation submodule, a sorting submodule and an output submodule.

[0050] The calculation submodule is used to calculate the importance of each target feature;

[0051] A sorting submodule, used to sort features according to their importance and obtain corresponding feature indexes; and

[0052] The output submodule is used to output the names of suppliers with medium, high and above risks, risk prediction results and key target characteristics.

[0053] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0054] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the above computer program or instructions are executed by a processor.

[0055] The fifth aspect of the present disclosure further provides a computer program product, including a computer program or instructions, which implement the steps of the above method when the above computer program or instructions are executed by a processor.

[0056] The supplier risk level assessment method provided by the embodiment of the present disclosure uses the Pearson coefficient to screen the indicators in the data set, remove invalid indicators, and improve the effectiveness of the model training data. The risk prediction model pre-trained by the ensemble learning algorithm is used to perform risk prediction on the collected supplier information to obtain the risk level assessment results of each supplier. Since the risk prediction model uses the ensemble learning algorithm to determine the optimal algorithm and parameters, it is more accurate than a single machine learning prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The above contents and other purposes, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0058] Figure 1The application scenario diagram of the supplier risk level assessment method, apparatus, device, medium and program product according to the embodiment of the present disclosure is schematically shown;

[0059] Figure 2 A flowchart of a method for assessing supplier risk level according to an embodiment of the present disclosure is schematically shown;

[0060] Figure 3 One of the flow charts for determining target features based on historical risk level assessment results and Pearson correlation coefficient according to an embodiment of the present disclosure is schematically shown;

[0061] Figure 4 Schematically shows a second flow chart of determining target features based on historical risk level assessment results and Pearson correlation coefficient according to an embodiment of the present disclosure;

[0062] Figure 5 A flowchart schematically shows a method of inputting the supplier risk data to be assessed into a risk prediction model to output a supplier risk level according to an embodiment of the present disclosure;

[0063] Figure 6 A flowchart schematically shows a method for training a risk prediction model according to an embodiment of the present disclosure;

[0064] Figure 7 A schematic diagram of a structure of a device for assessing supplier risk level according to an embodiment of the present disclosure is shown; and

[0065] Figure 8 A block diagram of an electronic device suitable for implementing a supplier risk level assessment method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0066] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0067] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "include", "comprising", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0068] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0069] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0070] First, the terms that appear in the embodiments of the present disclosure are explained:

[0071] Risk Assessment: The process of identifying, analyzing and evaluating potential risks in order to determine their likelihood and impact so that appropriate preventive measures can be taken.

[0072] Quantitative and qualitative indicators: Quantitative indicators are data that can be quantified, such as financial data, personnel supply capacity, etc.; qualitative indicators are descriptive data that are difficult to quantify, such as supplier reputation, service quality, etc.

[0073] Machine Learning Algorithm: An AI algorithm that can learn from data and make predictions or decisions, such as Logistic Regression, XGBoost, AdaBoosting, GDBT, etc.

[0074] Pearson's coefficient: A statistical measure used to assess the linear correlation between two variables, ranging from -1 (perfect negative correlation) to +1 (perfect positive correlation).

[0075] Boosting ensemble algorithm: An ensemble learning method that combines multiple weak learners to build a strong learner to improve the performance of the model.

[0076] NRBO algorithm: Nature-Inspired Optimization Algorithms, an optimization algorithm that simulates the biological behavior of nature and is used to find the optimal parameter settings.

[0077] Feature importance: In machine learning, feature importance refers to the degree of influence of each feature on the model's prediction results.

[0078] Warning threshold: A preset value. When the risk level predicted by the model exceeds this value, the system will issue a warning.

[0079] Data timeliness: the timeliness and accuracy of data. For risk assessment, high data timeliness means that the assessment results are closer to the current actual situation.

[0080] Stratified sampling: A sampling method that ensures that the proportion of each category in the sample is the same as in the population.

[0081] One-Hot Encoding: A method for handling categorical data that converts each category value into a binary vector.

[0082] Feature set screening: The process of selecting a subset of features that are most relevant to the target variable from all available features.

[0083] Data cleaning: refers to cleaning data to correct or remove erroneous information, including handling missing values, outliers, and incorrect data formats.

[0084] The common risk assessment methods for existing bank outsourcing suppliers are mainly to evaluate the comprehensive capabilities of suppliers through quantitative and qualitative indicators, such as financial status, actual staff supply and presence, etc., use data analysis and single mathematical or machine models to predict potential risks, or conduct comprehensive assessments of suppliers through professional third-party institutions. These methods cannot reflect the latest changes in the market and suppliers in a timely manner. In addition, third-party institutions have slow evaluation timelines, collect mixed information, and interfere with comprehensive evaluations. They also cost companies a certain amount of money.

[0085] Based on the above technical problems, an embodiment of the present disclosure provides a supplier risk level assessment method, the method comprising: in response to a supplier risk level assessment instruction, collecting supplier risk data to be assessed according to a target feature, wherein the target feature is determined based on a historical assessment risk level result and a Pearson correlation coefficient; and

[0086] The supplier risk data to be assessed is input into a risk prediction model to output the supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

[0087] Figure 1 The application scenario diagram of the supplier risk level assessment method, apparatus, device, medium and program product according to the embodiment of the present disclosure is schematically shown.

[0088] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a large-scale server installation scenario. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0089] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0090] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0091] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0092] It should be noted that the supplier risk level assessment method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the supplier risk level assessment device provided in the embodiment of the present disclosure can generally be set in the server 105. The supplier risk level assessment method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the supplier risk level assessment device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0093] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements. Figure 1The application scenarios shown are only examples of application scenarios and system architectures to which the embodiments of the present disclosure can be applied, in order to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios. It should be noted that the supplier risk level assessment method and device determined in the present disclosure can be used in the financial field, and can also be used in any field other than the financial field. The application field of the supplier risk level assessment method and device determined in the present disclosure is not limited.

[0094] The following will be based on Figure 1 The scene described by Figure 2~Figure 5 The supplier risk level assessment method of the disclosed embodiment is described in detail.

[0095] Figure 2 The flowchart of the supplier risk level assessment method according to an embodiment of the present disclosure is schematically shown.

[0096] like Figure 2 As shown, the supplier risk level assessment method of this embodiment includes operations S210 to S220, and the transaction processing method can be executed by a server or other computing devices.

[0097] In operation S210, in response to a supplier risk level assessment instruction, supplier risk data to be assessed is collected according to target features, where the target features are determined based on historical assessment risk level results and a Pearson correlation coefficient.

[0098] In one example, information on outsourcing suppliers is collected from multiple channels, including data on demand satisfaction, supplier management capabilities, financial status, and external adverse impacts. The supplier demand satisfaction is obtained from the internal R&D management system, and the personnel demand information proposed within the enterprise is counted monthly, as well as the actual satisfaction of each outsourcing supplier and the internal historical evaluation scores. Specifically, the indicator characteristics involved include the proportion of pushed recruitments, the average time for recruitments to arrive, the personnel demand satisfaction rate, and the evaluation scores of each department.

[0099] In one example, the supplier management capability is also obtained from the internal R&D management system, including the internal employee satisfaction evaluation of outsourced supplier personnel, whether there are complaints from peers or employees due to unfair means, and the R&D and production quality of outsourced personnel. Specifically, the indicator characteristics involved include outsourced employee satisfaction, historical issues such as forged resumes or unfair competition, the number of employee complaints and event types, outsourced R&D and production issues, and historical risk assessment levels.

[0100] In one example, a financial status assessment can be obtained from the latest financial statements provided by the supplier, such as the balance sheet, income statement and cash flow statement, which can be automatically captured and entered. Specific indicator features include asset and liability data, revenue and profit growth rates, and whether there are audit issues or irregularities in financial statements.

[0101] In one example, in terms of external adverse impacts, administrative penalty information and amount, court execution records, consumption restrictions on related legal persons, and illegal and dishonest blacklist records are queried based on the supplier name.

[0102] Figure 3 One of the flowcharts for determining target features based on historical risk level assessment results and Pearson correlation coefficient according to an embodiment of the present disclosure is schematically shown.

[0103] like Figure 3 As shown, operation S310 and operation S320 are included.

[0104] In operation S310, the collected supplier risk data is cleaned and a data feature set is extracted.

[0105] In one example, a Python script is used to extract and process key indicator data to complete data cleaning and integration. Specifically, the Excel / CSV file data provided by the supplier is extracted and the pandas library is installed for data processing. The read_excel and read_csv functions are used to read Excel or CSV files, or the requests and json libraries are used to obtain JSON data from external API interfaces. The code checks whether the data has missing values, outliers, or incorrect data formats.

[0106] Use Python code to extract key indicators, obtain the corresponding column data df['ColumnName'], and calculate the average arrival time of recruited personnel and other indicator results.

[0107] In operation S320, the data feature set is screened based on historical risk level assessment results and the Pearson coefficient to determine a target feature.

[0108] Figure 4 The second flowchart of determining target features based on historical risk level assessment results and Pearson correlation coefficient according to an embodiment of the present disclosure is schematically shown. Figure 4 As shown, operation S320 includes operations S321 to S323.

[0109] In operation S321 , the data feature set is converted into a data value type set based on a one-hot encoding technique.

[0110] In operation S322 , the Pearson coefficient between each data feature and the risk level is calculated.

[0111] In operation S323, a target feature is determined according to the Pearson coefficient.

[0112] In one example, in order to improve the accuracy of subsequent model training data, the disclosed embodiment screens the data feature set to determine the target feature, uses the Pearson coefficient to evaluate each indicator in the data set, and identifies the relevant feature set in combination with the historical risk level assessment results.

[0113] In one example, in order to calculate the maximum correlation between each evaluation indicator and the risk level, the data set is converted into a numerical set. Assuming that there are 4 types of risk levels (such as A high risk, B medium-high risk, C general risk, D low risk), a custom conversion can be set to map the risk level to a specific number ("A" = 4, "B" = 3, "C" = 2, "D" = 1).

[0114] Use one-hot encoding for other character-type indicator data, creating new binary columns for each indicator value. For example, one-hot encoding can be achieved using the get_dummies method of pandas:

[0115] df = pd.get_dummies(df, columns=['CategoricalColumn'])

[0116] In an example, the Pearson coefficient between each data feature and the risk level is calculated, and the correlation between each indicator and the risk level assessment result is compared. The calculation formula is as follows:

[0117]

[0118] in is the result of Pearson coefficient calculation, n is the number of samples, is the index value of the i-th sample, The risk level result of the i-th sample; is the indicator mean, is the mean risk level.

[0119] In an example, after calculating the Pearson coefficients between all indicators and risk levels, these coefficients are sorted from large to small by absolute value. The range of the Pearson correlation coefficient is between [-1,1]. If the correlation coefficient is close to 1, it means that there is a high positive correlation between the two variables, where 0.8-1 is a very strong correlation, 0.6-0.8 is a strong correlation, 0.4-0.6 is a moderate correlation, 0.2-0.4 is a weak correlation, and 0-0.2 is a very weak correlation or no correlation. According to the following analysis results, the Pearson coefficient of indicator X is the highest and is most relevant to the supplier risk assessment results. According to the Pearson coefficient correlation, the corresponding data set is split into 2 feature sets, and the relevant indicator data with a correlation of less than 0.2 is removed, and the other data is taken as a set .

[0120]

[0121] In operation S220, the supplier risk data to be assessed is input into a risk prediction model to output a supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

[0122] Figure 5 The flowchart schematically shows a method for inputting the supplier risk data to be assessed into a risk prediction model to output a supplier risk level according to an embodiment of the present disclosure.

[0123] like Figure 5 As shown, operation S220 includes operations S221 to S223.

[0124] In operation S221, the importance of each target feature is calculated.

[0125] In operation S222 , the features are sorted according to their importance and corresponding feature indexes are obtained.

[0126] In operation S223, the names of suppliers with medium-high and above risks, risk prediction results, and key target characteristics are output.

[0127] In one example, the best ensemble learning algorithm and parameter combination obtained through pre-training, for example, assuming that the algorithm with the highest final evaluation accuracy is the XGBoost model, the feature_importances_ attribute can be used to obtain the importance of each feature. Use the feature_importances_ attribute in python to calculate the importance of each feature element, sort the feature importance, and obtain the corresponding feature index. According to the degree of control of the enterprise over each outsourcing supplier, a reasonable warning threshold is set. Assuming that the model predicts that a certain outsourcing supplier reaches a medium-high risk level or above, the corresponding enterprise outsourcing supplier manager is immediately reminded through the background. The outsourcing supplier manager receives the warning information, including the supplier name, the risk level result predicted by the model, and the first three key target feature indicator results that cause medium-high risk and high risk obtained from the above steps. For example, after calculating and determining the following three key target features: low satisfaction of outsourced employees, low revenue and profit growth rate, and the existence of historical administrative penalties, the feature can be sent to the supplier for rectification.

[0128] The supplier risk level assessment method provided by the embodiment of the present disclosure uses the Pearson coefficient to screen the indicators in the data set, remove invalid indicators, and improve the effectiveness of the model training data. The risk prediction model pre-trained by the ensemble learning algorithm is used to perform risk prediction on the collected supplier information to obtain the risk level assessment results of each supplier. Since the risk prediction model uses the ensemble learning algorithm to determine the optimal algorithm and parameters, it is more accurate than a single machine learning prediction model.

[0129] The following will be combined Figure 6 The training process of the risk prediction model is explained. Figure 6 A flowchart of a risk prediction model training method according to an embodiment of the present disclosure is schematically shown.

[0130] like Figure 6 As shown, operation S410 to operation S450 are included.

[0131] In operation S410, model training data is generated based on the target features.

[0132] In operation S420, the model training data is divided into layers according to risk level types.

[0133] In one example, according to the target feature set Prepare model training data based on supplier risk assessment results, adopt stratified sampling, use Scikit-learn's train_test_split function, set the stratify parameter to implement stratified sampling, and extract 80% of each risk level as the training set, and the rest as the test set, ensuring that the proportion of each risk level category in the training set and the test set is the same as that in the original data set.

[0134] In operation S430, a plurality of ensemble learning algorithms are selected for prediction training.

[0135] According to an embodiment of the present disclosure, the integrated learning algorithm includes an XGBoost algorithm, an AdaBoosting algorithm and a GDBT algorithm.

[0136] Through Python programs, three typical integrated learning algorithms, XGBoost, AdaBoosting, and GDBT, are called to carry out training and prediction.

[0137] In operation S440, a natural rhythm optimization algorithm is called to perform iterative optimization to determine an optimal parameter combination of the model.

[0138] In operation S450, a risk prediction model is determined according to the prediction accuracy of each algorithm.

[0139] In an example, the NRBO algorithm is called for parameter optimization. The NRBO algorithm is initialized with the default parameter combination and the optimal parameter settings are found through an iterative process. NRBO is a natural heuristic optimization algorithm used to find the optimal parameter settings. The following are the steps for using NRBO for parameter optimization: Define the objective function, that is, the performance indicator of the model to be optimized. Initialize the parameters of the NRBO algorithm, including population size, number of iterations, etc. Run the NRBO algorithm and find the optimal parameter settings through an iterative process. During the running process, the best parameters and fitness of each iteration are output. After reaching the maximum number of iterations, the NRBO convergence results are visualized and the trained model is saved through joblib.dump. Compare the joint prediction effects of the three algorithms, select the algorithm with the highest accuracy as the final risk prediction model, and output the corresponding parameter combination.

[0140] Based on the above supplier risk level assessment method, the present disclosure also provides a supplier risk level assessment device. Figure 7 The device is described in detail.

[0141] Figure 7 The structural block diagram of the supplier risk level assessment device according to an embodiment of the present disclosure is schematically shown.

[0142] like Figure 7As shown, the supplier risk level assessment device 700 of this embodiment includes a collection module 710 and a risk assessment module 720 .

[0143] The collection module 710 is used to respond to the supplier risk level assessment instruction and collect supplier risk data to be assessed according to the target characteristics, wherein the target characteristics are determined based on the historical assessment risk level results and the Pearson correlation coefficient. In one embodiment, the collection module 710 can be used to perform the operation S210 described above, which will not be repeated here.

[0144] The risk assessment module 720 is used to input the supplier risk data to be assessed into a risk prediction model to output a supplier risk level, wherein the risk prediction model is pre-trained based on an ensemble learning algorithm. In one embodiment, the risk assessment module 720 can be used to perform the operation S220 described above, which will not be described in detail here.

[0145] According to an embodiment of the present disclosure, a target feature determination module is also included.

[0146] The target feature determination module is used to determine the target feature based on the historical risk level assessment results and the Pearson correlation coefficient. In one embodiment, the target feature determination module can be used to perform the operations S310 and S320 described above, which will not be described in detail here.

[0147] According to an embodiment of the present disclosure, the target feature determination module includes a data extraction submodule and a first determination submodule.

[0148] The data extraction submodule is used to clean the collected supplier risk data and extract a data feature set; in one embodiment, the data extraction submodule can be used to perform the operation S310 described above, which will not be repeated here.

[0149] The first determination submodule is used to screen the data feature set based on the historical risk level assessment results and the Pearson coefficient to determine the target feature. In one embodiment, the first determination submodule can be used to perform the operation S320 described above, which will not be repeated here.

[0150] According to an embodiment of the present disclosure, the first determining submodule includes a data conversion unit, a calculation unit and a determination unit.

[0151] The data conversion unit is used to convert the data feature set into a data value type set based on the one-hot encoding technology; in one embodiment, the data conversion unit can be used to perform the operation S321 described above, which will not be repeated here.

[0152] The calculation unit is used to calculate the Pearson coefficient between each data feature and the risk level. In one embodiment, the calculation unit can be used to perform the operation S322 described above, which will not be repeated here.

[0153] The determination unit is configured to determine the target feature according to the Pearson coefficient. In one embodiment, the determination unit may be configured to perform the operation S323 described above, which will not be described in detail herein.

[0154] According to an embodiment of the present disclosure, a risk prediction model training module is also included.

[0155] The risk prediction model training module includes: a generation submodule, a splitting submodule, a training submodule, a parameter optimization submodule and a second determination submodule.

[0156] A generation submodule is used to generate model training data based on the target features. In one embodiment, the generation submodule can be used to perform the operation S410 described above, which will not be repeated here.

[0157] The splitting submodule is used to split the model training data into layers according to the risk level type; in one embodiment, the splitting submodule can be used to perform the operation S420 described above, which will not be repeated here.

[0158] The training submodule is used to select multiple ensemble learning algorithms for prediction training. In one embodiment, the training submodule can be used to perform the operation S430 described above, which will not be repeated here.

[0159] The parameter optimization submodule is used to call the natural rhythm optimization algorithm for iterative optimization to determine the optimal parameter combination of the model; and in one embodiment, the parameter optimization submodule can be used to perform the operation S440 described above, which will not be repeated here.

[0160] The second determination submodule is used to determine the risk prediction model according to the prediction accuracy of each algorithm. In one embodiment, the second determination submodule can be used to perform the operation S450 described above, which will not be repeated here.

[0161] According to an embodiment of the present disclosure, the risk assessment module includes a calculation submodule, a sorting submodule and an output submodule.

[0162] The calculation submodule is used to calculate the importance of each target feature. In one embodiment, the calculation submodule can be used to perform the operation S221 described above, which will not be repeated here.

[0163] The sorting submodule is used to sort according to feature importance and obtain corresponding feature indexes; in one embodiment, the sorting submodule can be used to perform the operation S222 described above, which will not be repeated here.

[0164] The output submodule is used to output the names of suppliers with medium-high or higher risks, risk prediction results, and key target features. In one embodiment, the output submodule can be used to perform the operation S223 described above, which will not be repeated here.

[0165] According to an embodiment of the present disclosure, any multiple modules in the acquisition module 710 and the risk assessment module 720 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the acquisition module 710 and the risk assessment module 720 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the acquisition module 710 and the risk assessment module 720 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0166] Figure 8 A block diagram of an electronic device suitable for implementing a supplier risk level assessment method according to an embodiment of the present disclosure is schematically shown.

[0167] like Figure 8 As shown, the electronic device 900 according to an embodiment of the present disclosure includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 to a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include an onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0168] In RAM 903, various programs and data required for the operation of electronic device 900 are stored. Processor 901, ROM 902 and RAM 903 are connected to each other via bus 904. Processor 901 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 902 and / or RAM 903. It should be noted that the program can also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0169] According to an embodiment of the present disclosure, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to the bus 904. The electronic device 900 may further include one or more of the following components connected to the input / output (I / O) interface 905: an input portion 906 including a keyboard, a mouse, etc.; an output portion 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 908 including a hard disk, etc.; and a communication portion 909 including a network interface card such as a LAN card, a modem, etc. The communication portion 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage portion 908 as needed.

[0170] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the supplier risk level assessment method according to the embodiment of the present disclosure is implemented.

[0171] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: 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), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 902 and / or RAM 903 described above and / or one or more memories other than ROM 902 and RAM 903.

[0172] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the supplier risk level assessment method provided by the embodiment of the present disclosure.

[0173] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 901. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0174] In one embodiment, the computer program may be based on a tangible storage medium such as an optical storage device, a magnetic storage device, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of a signal on a network medium, and downloaded and installed through the communication part 909, and / or installed from a removable medium 911. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0175] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the processor 901, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0176] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0177] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two 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 box in the block diagram or flow chart, and the combination of the boxes in the block diagram 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.

[0178] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0179] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A supplier risk level assessment method, characterized in that: The method comprises: In response to a supplier risk level assessment instruction, collecting supplier risk data to be assessed according to target characteristics, wherein the target characteristics are determined based on historical assessment risk level results and a Pearson correlation coefficient; and The supplier risk data to be assessed is input into a risk prediction model to output the supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

2. The method according to claim 1, characterized in that Target characteristics determined based on historical risk assessment results and Pearson correlation coefficient include: Clean the collected supplier risk data and extract data feature sets; The data feature set is screened based on historical assessment risk level results and the Pearson coefficient to determine the target feature.

3. The method according to claim 2, characterized in that The data feature set is screened based on the historical risk level assessment results and the Pearson coefficient to determine the target features include: Convert the data feature set into a data value set based on the one-hot encoding technology; Calculate the Pearson coefficient between each data feature and risk level; and The target feature is determined according to the Pearson coefficient.

4. The method according to any one of claims 1 to 3, characterized in that The risk prediction models pre-trained based on the ensemble learning algorithm include: Generate model training data based on the target features; The model training data is divided into layers according to the risk level type; Select multiple ensemble learning algorithms for prediction training; Call the natural rhythm optimization algorithm for iterative optimization to determine the optimal parameter combination of the model; and The risk prediction model is determined based on the prediction accuracy of each algorithm.

5. The method according to claim 4, characterized in that The ensemble learning algorithms include XGBoost algorithm, AdaBoosting algorithm and GDBT algorithm.

6. The method according to claim 1, characterized in that The step of inputting the supplier risk data to be assessed into a risk prediction model to output the supplier risk level includes: Calculate the importance of each target feature; Sort by feature importance and obtain the corresponding feature index; and Output the names of suppliers with medium, high or above risks, risk prediction results and key target characteristics.

7. A supplier risk level assessment device, characterized in that: The device comprises: A collection module, configured to respond to a supplier risk level assessment instruction and collect supplier risk data to be assessed according to a target feature, wherein the target feature is determined based on a historical assessment risk level result and a Pearson correlation coefficient; and The risk assessment module is used to input the supplier risk data to be assessed into a risk prediction model to output the supplier risk level, wherein the risk prediction model is pre-trained based on an integrated learning algorithm.

8. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images