CN113706291A - Fraud risk prediction method, device, equipment and storage medium - Google Patents
Fraud risk prediction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113706291A CN113706291A CN202111009514.XA CN202111009514A CN113706291A CN 113706291 A CN113706291 A CN 113706291A CN 202111009514 A CN202111009514 A CN 202111009514A CN 113706291 A CN113706291 A CN 113706291A
- Authority
- CN
- China
- Prior art keywords
- historical
- customer
- client
- historical client
- information characteristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 239000013598 vector Substances 0.000 claims description 55
- 230000006870 function Effects 0.000 claims description 34
- 230000006399 behavior Effects 0.000 claims description 28
- 238000012512 characterization method Methods 0.000 claims description 23
- 230000011218 segmentation Effects 0.000 claims description 13
- 230000004913 activation Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 10
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000000875 corresponding effect Effects 0.000 description 22
- 238000004891 communication Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000011835 investigation Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Marketing (AREA)
- Medical Informatics (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an artificial intelligence technology, and discloses a fraud risk prediction method, which comprises the following steps: the method comprises the steps of constructing a knowledge graph of a historical customer, analyzing the knowledge graph by using a pre-constructed graph learning model to generate a prediction score of the historical customer, calculating a loss value between the prediction score and a fraud score of the corresponding historical customer, performing parameter adjustment on the graph learning model according to the loss value to obtain a trained graph learning model, predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain the prediction score of the customer to be analyzed, and judging the fraud risk of the customer to be analyzed according to the prediction score of the customer to be analyzed and a preset warning threshold value. In addition, the invention also relates to a block chain technology, and the knowledge graph can be stored in the nodes of the block chain. The invention also provides a fraud risk prediction device, electronic equipment and a storage medium. The method and the device can improve the accuracy of the fraud risk prediction of pre-credit wind control.
Description
Technical Field
The present invention relates to the field of artificial intelligence technologies, and in particular, to a fraud risk prediction method and apparatus, an electronic device, and a computer-readable storage medium.
Background
At present, various loan businesses are threatened by data loss and confusion in big data wind control, library collision, account number stealing, identity stealing, pseudo base stations and the like, and economic losses are easily brought to users of the loan businesses.
In order to solve the above problem, in the loan, pre-loan wind control needs to be performed. The pre-loan wind control is that a financial institution carries out comprehensive investigation and understanding on a company or an individual to be loaned through various ways to achieve risk prediction before the company or the individual loan.
However, the current pre-credit wind control has the following problems: 1. the method completely depends on the credit investigation information of the client in the credit investigation system, and because the credit investigation system has the condition of incomplete personnel information and only predicts the current information of the client, the input amount of the information is small, and the fraud risk prediction accuracy rate is not high under the conditions; 2. the credit investigation system has long updating time, and the accuracy is low because the change condition can not be supported and predicted in time. Thus, the current pre-credit wind-based fraud risk prediction accuracy is low.
Disclosure of Invention
The invention provides a fraud risk prediction method, a fraud risk prediction device and a computer readable storage medium, and mainly aims to solve the problem of low accuracy of fraud risk prediction of pre-loan wind control.
In order to achieve the above object, the present invention provides a fraud risk prediction method, including:
acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set, and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set;
acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client;
collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set;
constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
weighting and accumulating the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set;
calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model;
acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
Optionally, the extracting a historical customer information feature set from the historical customer information set includes:
performing word segmentation and part-of-speech tagging on the historical client information set to obtain word segmentation and part-of-speech tagging results;
extracting nouns and noun phrases in the participles according to the results of the participles and part-of-speech tagging, counting to obtain a historical client information characteristic frequency set according to the nouns and the noun phrases, and generating a frequent pattern tree according to the historical client information characteristic frequency set;
identifying the characteristics in the frequent pattern tree to obtain a candidate historical client information characteristic set;
and calculating the mutual point information value of each characteristic in the candidate historical client information characteristic set, and filtering out the historical client information characteristics with the mutual point information value smaller than a preset standard threshold value from the candidate historical client information characteristic set to obtain a historical client information characteristic set.
Optionally, the obtaining neighboring clients of each historical client through each historical client information feature in the historical client information feature set includes:
converting the word segmentation of the historical client information characteristic set into a vector to obtain a historical client information characteristic vector;
selecting one client from the historical client set one by one as an initial historical client, and extracting historical client information characteristics of the initial historical client from the historical client information characteristic set;
converting the historical customer information characteristics of the initial historical customer into corresponding vectors to obtain initial customer information characteristic vectors;
and calculating the similarity between the historical customer information characteristic vector and the initial customer information characteristic vector, and obtaining the adjacent customers of the initial historical customer according to the similarity.
Optionally, the constructing a knowledge-graph of the historical customer set according to the historical customer information feature set and the adjacent customer information feature set includes:
extracting entity vocabularies and relation vocabularies from the segmentation of the historical customer information characteristic set and the adjacent customer information characteristic set;
classifying the entity vocabulary and the relation vocabulary, and respectively storing the classification results of the entity vocabulary and the relation vocabulary into an entity vocabulary library and a relation vocabulary library;
and constructing a knowledge graph of the historical client set based on the entity vocabulary library and the relation vocabulary library.
Optionally, the weighting and accumulating the knowledge graph of the historical customer by using the pre-constructed graph learning model to generate the prediction score set of the historical customer set includes:
carrying out weighted summation operation on the knowledge graph by using an ith convolution layer in a pre-constructed graph learning model to obtain an ith node characterization vector, wherein i is 1,2 and 3 … n;
transferring the ith node characterization vector to an (i +1) th convolution layer through an activation function to carry out weighted summation to obtain an (i +1) th node characterization vector until i is n-1 to obtain an nth node characterization vector;
and classifying and scoring the n-th node characterization vector to obtain a prediction score set of the historical client set.
Optionally, the transferring the ith node characterization vector to the (i +1) th convolutional layer through an activation function to perform weighted summation to obtain an (i +1) th node characterization vector, where the method includes:
the step of transferring the ith node characterization vector to the (i +1) th convolutional layer through an activation function to perform weighted summation to obtain an (i +1) th node characterization vector includes:
where σ () represents an activation function; l represents the first layer convolution layer, i represents the current node; r represents the relationship between nodes; r represents all relationships between nodes;
representing all node sets with a relation R with the current node i under the condition that R belongs to the R relation; c. Ci,rRepresenting a regularization constant;
a weight representing a self-loop;
a weight representing the node relationship r;
representing the feature vector of the current node i in the first layer convolution layer;
represents the feature vector of node j in the first convolutional layer.
Optionally, the calculating a fraud score set corresponding to the historical customer set according to the historical customer information feature set includes:
selecting one of the historical clients from the historical client set, and acquiring the repayment date, the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data of the selected historical client from the historical client information characteristic set;
extracting the on-time repayment days, the advance repayment days and the overdue repayment days from the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data respectively;
counting the repayment behavior days of the selected historical client according to the on-time repayment days, the advance repayment days and the overdue repayment days;
calculating a historical customer fraud score S for the selected historical customer by the following formula:
S＝D×F(x)×P
d represents the repayment behavior days, P represents the repayment period number, F (x) represents a normal distribution function, mu represents the average number of the historical customer information set, sigma represents the standard deviation of the historical customer information set, and x is a preset factor.
In order to solve the above problem, the present invention also provides a fraud risk prediction apparatus, including:
the knowledge map building module is used for acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set; acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client; collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set; constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
the score prediction module is used for performing weighting and accumulation operation on the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set; calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model; acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and the risk decision module is used for judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
In order to solve the above problem, the present invention also provides an electronic device, including:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the fraud risk prediction method described above.
In order to solve the above problem, the present invention also provides a computer-readable storage medium, in which at least one computer program is stored, the at least one computer program being executed by a processor in an electronic device to implement the fraud risk prediction method described above.
The embodiment of the invention extracts a historical client information characteristic set from the historical client information set, calculates a fraud score set corresponding to the historical client set according to the historical client information characteristic set, associates an adjacent client set through the historical client information characteristic set, constructs a knowledge graph of the historical client set through the historical client information characteristic set and the adjacent client information characteristic set, constructs the knowledge graph by using the characteristics of the client information and the adjacent client information together, supplements and perfects the client information, thereby improving the accuracy of predicting the pre-credit wind-controlled fraud risk, trains a graph learning model through the knowledge graph and the fraud score set to obtain a trained graph learning model, analyzes the knowledge graph of the client by using the trained graph learning model to obtain the fraud risk of the client to be analyzed, continuously trains and optimizes by using the graph learning model, and the model prediction accuracy is improved. Therefore, the fraud risk prediction method, the fraud risk prediction device, the electronic equipment and the computer readable storage medium can improve the accuracy of fraud risk prediction of the pre-loan wind control of the customer to be analyzed.
Drawings
Fig. 1 is a schematic flow chart of a fraud risk prediction method according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating a detailed implementation of one of the steps in the fraud risk prediction method shown in FIG. 1;
FIG. 3 is a flow chart illustrating a detailed implementation of one of the steps in the fraud risk prediction method shown in FIG. 1;
FIG. 4 is a flowchart illustrating a detailed implementation of one of the steps in the fraud risk prediction method shown in FIG. 1;
FIG. 5 is a functional block diagram of a fraud risk prediction apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device implementing the fraud risk prediction method according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the application provides a fraud risk prediction method. The execution subject of the fraud risk prediction method includes, but is not limited to, at least one of electronic devices that can be configured to execute the method provided by the embodiment of the present application, such as a server, a terminal, and the like. In other words, the fraud risk prediction method may be performed by software or hardware installed in the terminal device or the server device, and the software may be a block chain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a Network service, cloud communication, a middleware service, a domain name service, a security service, a Content Delivery Network (CDN), a big data and artificial intelligence platform, and the like.
Fig. 1 is a schematic flow chart of a fraud risk prediction method according to an embodiment of the present invention. In this embodiment, the fraud risk prediction method includes:
s1, acquiring a historical client information set of a historical client set, extracting a historical client information feature set from the historical client information set, and calculating a fraud score set corresponding to the historical client set according to the historical client information feature set;
the historical customer information in the embodiment of the invention comprises basic information, behavior data, operator data, network public information and the like of the customer, and the main sources are credit investigation platforms, various nominal loan platforms, banks, social media and the like.
In the embodiment of the invention, the historical customer information characteristics are descriptions of basic attributes, information requirements, information behaviors, psychological states, environmental influences and the like of a user, and are mainly divided into dimensional attributes such as identity traits, performance capability, credit history, interpersonal relationships, behavior preference and the like, and the final model predicts corresponding customer scores according to the dimensional information.
In detail, referring to fig. 2, the extracting a historical customer information feature set from the historical customer information feature set includes:
s11, performing word segmentation and part-of-speech tagging on the historical client information set to obtain word segmentation and part-of-speech tagging results;
s12, extracting nouns and noun phrases in the participles according to the participle and part-of-speech tagging results, counting to obtain a historical client information characteristic frequency set according to the nouns and the noun phrases, and generating a frequent pattern tree according to the historical client information characteristic frequency set;
in the embodiment of the invention, the transaction file is scanned, the frequency of the historical client information characteristics appearing in the transaction file is counted, the historical client information characteristics in the historical client information characteristic frequency set are sorted in a descending order according to the frequency, and the frequent pattern tree based on the historical client information characteristic frequency set is obtained.
S13, identifying the features in the frequent pattern tree to obtain a candidate historical customer information feature set;
in the embodiment of the invention, the frequent pattern tree is decomposed into a plurality of conditional frequent pattern trees, each conditional frequent pattern tree is subjected to frequent pattern mining, and historical customer information characteristics which are lower than a preset frequency in each conditional frequent pattern tree are filtered out, so that a candidate historical customer information characteristic set is obtained.
S14, calculating the mutual point information value of each feature in the candidate historical client information feature set, and filtering out the historical client information features of which the mutual point information value is smaller than a preset standard threshold value from the candidate historical client information feature set to obtain a historical client information feature set.
In the embodiment of the invention, the PMI algorithm is utilized to calculate the point mutual information value of each characteristic in the candidate historical client information characteristic set, and the calculation formula is defined as follows:
the higher the point mutual information value is, the higher the correlation degree of the historical client and the characteristics is, and when the point mutual information value is lower than a preset standard threshold value, the corresponding historical client information characteristics are filtered; and when the point mutual information value is higher than a preset standard threshold value, retaining corresponding historical customer information characteristics, and combining to obtain the historical customer information characteristic set.
Further, the calculating a fraud score set corresponding to the historical customer set according to the historical customer information feature set according to the embodiment of the present invention includes:
selecting one of the historical clients from the historical client set, and acquiring the repayment date, the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data of the selected historical client from the historical client information characteristic set;
extracting the on-time repayment days, the advance repayment days and the overdue repayment days from the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data respectively;
counting the repayment behavior days of the selected historical client according to the on-time repayment days, the advance repayment days and the overdue repayment days;
calculating a historical customer fraud score S for the selected historical customer by the following formula:
S＝D×F(x)×P
d represents the repayment behavior days, P represents the repayment period number, F (x) represents a normal distribution function, mu represents the average number of the historical customer information set, sigma represents the standard deviation of the historical customer information set, and x is a preset factor.
In the embodiment of the invention, the historical client fraud score can directly reflect the historical client fraud risk, and the lower the fraud score is, the greater the fraud risk is.
S2, acquiring the adjacent client of each history client according to the information characteristics of each history client, and associating each history client with the corresponding adjacent client;
in detail, referring to fig. 3, the obtaining neighboring clients of each historical client through each historical client information feature in the historical client information feature set includes:
s21, converting the word segmentation of the historical customer information feature set into a vector to obtain a historical customer information feature vector;
the embodiment of the invention carries out word segmentation on the historical client information characteristics to obtain historical client information characteristic words, and converts the historical client information characteristic words into historical client information characteristic vectors according to a preset vector conversion algorithm.
S22, selecting one client from the historical client set as an initial historical client one by one, and extracting the historical client information characteristics of the initial historical client from the historical client information characteristic set;
s23, converting the historical customer information characteristics of the initial historical customer into corresponding vectors to obtain initial customer information characteristic vectors;
and S24, calculating the similarity between the historical customer information characteristic vector and the initial customer information characteristic vector, and obtaining the adjacent customers of the initial historical customer according to the similarity.
According to the embodiment of the invention, whether the adjacent customers are in a neighborhood is judged according to the preset threshold tau of the similarity; when the similarity is smaller than the preset threshold value tau, judging that the client is not an adjacent client of the historical client; and when the similarity is greater than or equal to the preset threshold value tau, judging that the client is a neighboring client of the historical client.
Specifically, the similarity between the historical client information feature vector and the initial client feature vector is calculated by using a cosine similarity algorithm, and a calculation formula is defined as follows:
wherein a is a history client, b is an initial client, AiFor historical customer information feature vectors, BiIs the initial customer feature vector. When cos (a, b) < τ, determining as not a neighbor client of the history client; and when cos (a, b) ≧ τ, judging that the client is the neighbor of the historical client, and obtaining the neighbor of the historical client.
According to the embodiment of the invention, all history clients are traversed to obtain the adjacent client of each history client; the historical customer information features are correlated to the adjacent customers with similar features, so that the graph learning model with higher accuracy can be obtained by utilizing the adjacent customer information training.
S3, summarizing the adjacent client information of each historical client to obtain an adjacent client information feature set;
in the embodiment of the invention, the adjacent customer information comprises basic information, behavior data, operator data, network public information and the like of the customer. The neighbor client information feature set is obtained by traversing the historical client information set to the last historical client, referring to the S1 to extract neighbor client information features, and summarizing the neighbor client information features of each client.
S4, constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
the knowledge graph is a multi-relation graph, and is a network knowledge graph formed by connecting entities with attributes through relations. The entities are real world things such as history clients, neighboring clients, history client companies, neighboring client companies, etc., and the relationships are used to express some kind of connection between different entities. The Entity (Entity) expresses a node in the graph and the relationship (relationship) expresses an edge in the graph. For example: "historical clients and neighboring clients are friendships," which are entities in the graph and friendships are edges in the graph.
In detail, the method for constructing the knowledge graph of the historical client according to the historical client information characteristic set and the adjacent client information characteristic set comprises the following steps:
extracting entity vocabularies and relation vocabularies from the segmentation of the historical customer information characteristic set and the adjacent customer information characteristic set;
classifying the entity vocabulary and the relation vocabulary, and respectively storing the classification results of the entity vocabulary and the relation vocabulary into an entity vocabulary library and a relation vocabulary library;
and constructing a knowledge graph of the historical client set based on the entity vocabulary library and the relation vocabulary library. Further, the knowledge graph of the history client is represented as:
G＝(E,R,S)
wherein E represents a node, R represents a relationship, S represents Ei×R×EjThe triplet of (2).
S5, carrying out weighting and accumulation operation on the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set;
in the embodiment of the present invention, the pre-constructed graph learning model may be an R-GCN (Relational graph convolutional network) model, and a graph convolutional network is formed by a plurality of convolutional layers. For example: the SqueezeNet network comprises a plurality of convolutional layers and is a lightweight network, and the computation capability is more efficient than that of a common network.
In detail, referring to fig. 4, the S5 includes:
s51, carrying out weighted summation operation on the knowledge graph by using the ith convolution layer in the pre-constructed graph learning model to obtain an ith node characterization vector, wherein i is 1,2,3 … n;
specifically, the node characterization vectors are obtained by performing weighted summation on the node characteristics and the node relation in the knowledge graph spectrum of the historical client by utilizing the convolution layer in the constructed graph learning model; the (i +1) th node characterization vector is calculated, for example, by the following formula:
where σ () represents an activation function; l represents the first layer convolution layer, i represents the current node; r represents the relationship between nodes; r represents all relationships between nodes;
representing all node sets with a relation R with the current node i under the condition that R belongs to the R relation; c. Ci,rRepresenting a regularization constant;
a weight representing a self-loop;
a weight representing the node relationship r;
representing the feature vector of the current node i in the first layer convolution layer;
represents the feature vector of node j in the first convolutional layer.
S52, transferring the ith node characterization vector to an (i +1) th convolution layer through an activation function for weighted summation to obtain an (i +1) th node characterization vector until i is equal to n-1 to obtain an nth node characterization vector;
in the embodiment of the invention, the activation function is a Relu function, the calculation speed of the function is high, and the calculation speed of the graph learning model is improved. The Relu function formula is as follows:
Relu＝max(0,x)
and S53, classifying and scoring the n-th node characterization vector to obtain a prediction score set of the historical client set.
According to the embodiment of the invention, the characterization vectors are classified and scored to obtain the historical customer predicted fraud scores, the output of multiple dimensions is mapped into a (0,1) interval by using a softmax function, and the occurrence probability is understood as the scores, so that the historical customer predicted scores are obtained. For example: vi represents the ith node in V, then the softmax value for this node is:
Sei
s6, calculating a loss value between the prediction score set and the fraud score set by using a loss function, and carrying out parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model;
the embodiment of the invention calculates the loss value between the prediction score of the historical customer and the fraud score of the historical customer by using the following loss function:
L(x,y)＝-(1-x)*log(1-y)
wherein L (x, y) is the loss value, x is the historical customer fraud score, and y is the historical customer forecast score.
In detail, in the embodiment of the present invention, the performing parameter adjustment on the graph learning model according to the loss value to obtain the trained graph learning model includes:
when the loss value of the loss function is larger than or equal to a preset loss threshold value, optimizing the parameters of the graph learning model by using an optimization algorithm;
and when the loss value of the loss function is smaller than the loss threshold value, obtaining the trained image learning model.
In the embodiment of the invention, when the loss value of the loss function is greater than the preset loss threshold value, the Adadelta optimization algorithm is used for optimizing the parameters of the graph learning model, and the Adadelta optimization algorithm can adaptively adjust the learning rate in the training process of the graph learning model, so that the graph learning model is more accurate, and the fraud risk prediction accuracy is improved.
S7, acquiring a knowledge graph of the customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained image learning model to obtain the prediction score of the customer to be analyzed;
in the embodiment of the present invention, the customer knowledge graph to be analyzed obtaining method refers to the steps S1, S2, S3, and S4.
According to the embodiment of the invention, the information characteristics of the customer to be analyzed can be constructed into the knowledge graph of the customer to be analyzed in a knowledge graph mode, and when the knowledge graph of the customer to be analyzed is obtained, the knowledge graph of the customer to be analyzed is input into the trained graph learning model for analysis, so that the prediction score of the customer to be analyzed is obtained.
And S8, judging the fraud risk of the customer according to the forecast score of the customer to be analyzed and a preset warning threshold value.
In the embodiment of the present invention, when the prediction score of the customer to be analyzed is greater than or equal to the preset warning threshold, it is determined that the fraud risk of the customer to be analyzed is low, and when the prediction score of the customer to be analyzed is less than or equal to the preset warning threshold, it is determined that the fraud risk of the customer to be analyzed is high.
The embodiment of the invention extracts a historical client information characteristic set from the historical client information set, calculates a fraud score set corresponding to the historical client set according to the historical client information characteristic set, associates an adjacent client set through the historical client information characteristic set, constructs a knowledge graph of the historical client set through the historical client information characteristic set and the adjacent client information characteristic set, constructs the knowledge graph by using the characteristics of the client information and the adjacent client information together, supplements and perfects the client information, thereby improving the accuracy of predicting the pre-credit wind-controlled fraud risk, trains a graph learning model through the knowledge graph and the fraud score set to obtain a trained graph learning model, analyzes the knowledge graph of the client by using the trained graph learning model to obtain the fraud risk of the client to be analyzed, continuously trains and optimizes by using the graph learning model, and the model prediction accuracy is improved. Therefore, the fraud risk prediction method, the fraud risk prediction device, the electronic equipment and the computer readable storage medium can improve the accuracy of fraud risk prediction of the pre-loan wind control of the customer to be analyzed.
Fig. 5 is a functional block diagram of a fraud risk prediction apparatus according to an embodiment of the present invention.
The fraud risk prediction apparatus 100 of the present invention may be installed in an electronic device. According to the implemented functions, the fraud risk prediction apparatus 100 may include a knowledge graph construction module 101, a score prediction module 102, and a risk decision module 103. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
In the present embodiment, the functions regarding the respective modules/units are as follows:
the knowledge graph building module 101 is configured to acquire a historical client information set of a historical client set, extract a historical client information feature set from the historical client information set, and calculate a fraud score set corresponding to the historical client set according to the historical client information feature set; acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client; collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set; constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
the score prediction module 102 is configured to perform weighting and accumulation operations on the knowledge graph by using a pre-constructed graph learning model, and generate a prediction score set of the historical client set; calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model; acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and the risk decision module 103 is used for judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
In detail, when the modules in the fraud risk prediction apparatus 100 according to the embodiment of the present invention are used, the same technical means as the fraud risk prediction method described in fig. 1 to 4 are adopted, and the same technical effects can be produced, which is not described herein again.
Fig. 6 is a schematic structural diagram of an electronic device implementing a fraud risk prediction method according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may further comprise a computer program, such as a fraud risk prediction program, stored in the memory 11 and executable on the processor 10.
In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same function or different functions, and includes one or more Central Processing Units (CPUs), a microprocessor, a digital Processing chip, a graphics processor, a combination of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the whole electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device by running or executing programs or modules (for example, executing fraud risk prediction programs and the like) stored in the memory 11 and calling data stored in the memory 11.
The memory 11 includes at least one type of readable storage medium including flash memory, removable hard disks, multimedia cards, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disks, optical disks, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, for example a removable hard disk of the electronic device. The memory 11 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device. The memory 11 may be used not only to store application software installed in the electronic device and various types of data, such as codes of fraud risk prediction programs, etc., but also to temporarily store data that has been output or is to be output.
The communication bus 12 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
The communication interface 13 is used for communication between the electronic device and other devices, and includes a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), which are typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable, among other things, for displaying information processed in the electronic device and for displaying a visualized user interface.
Fig. 6 only shows an electronic device with components, and it will be understood by a person skilled in the art that the structure shown in fig. 6 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or a combination of certain components, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management and the like are realized through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The fraud risk prediction program stored in the memory 11 of the electronic device 1 is a combination of instructions that, when executed in the processor 10, enable:
acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set, and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set;
acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client;
collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set;
constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
weighting and accumulating the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set;
calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model;
acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
Specifically, the specific implementation method of the instruction by the processor 10 may refer to the description of the relevant steps in the embodiment corresponding to the drawings, which is not described herein again.
Further, the integrated modules/units of the electronic device 1, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. The computer readable storage medium may be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying said computer program code, recording medium, U-disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM).
The present invention also provides a computer-readable storage medium, storing a computer program which, when executed by a processor of an electronic device, may implement:
acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set, and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set;
acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client;
collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set;
constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
weighting and accumulating the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set;
calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model;
acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method can be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product service layer, an application service layer, and the like.
The embodiment of the application can acquire and process related data based on an artificial intelligence technology. Among them, Artificial Intelligence (AI) is a theory, method, technique and application system that simulates, extends and expands human Intelligence using a digital computer or a machine controlled by a digital computer, senses the environment, acquires knowledge and uses the knowledge to obtain the best result.
Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the system claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. A fraud risk prediction method, characterized in that the method comprises:
acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set, and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set;
acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client;
collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set;
constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
weighting and accumulating the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set;
calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model;
acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
2. The fraud risk prediction method of claim 1, wherein said extracting a set of historical customer information characteristics from the set of historical customer information comprises:
performing word segmentation and part-of-speech tagging on the historical client information set to obtain word segmentation and part-of-speech tagging results;
extracting nouns and noun phrases in the participles according to the results of the participles and part-of-speech tagging, counting to obtain a historical client information characteristic frequency set according to the nouns and the noun phrases, and generating a frequent pattern tree according to the historical client information characteristic frequency set;
identifying the characteristics in the frequent pattern tree to obtain a candidate historical client information characteristic set;
and calculating the mutual point information value of each characteristic in the candidate historical client information characteristic set, and filtering out the historical client information characteristics with the mutual point information value smaller than a preset standard threshold value from the candidate historical client information characteristic set to obtain a historical client information characteristic set.
3. The fraud risk prediction method of claim 2, wherein said obtaining neighboring customers to each of the historical customers through each of the historical customer information features in the set of historical customer information features comprises:
converting the word segmentation of the historical client information characteristic set into a vector to obtain a historical client information characteristic vector;
selecting one client from the historical client set one by one as an initial historical client, and extracting historical client information characteristics of the initial historical client from the historical client information characteristic set;
converting the historical customer information characteristics of the initial historical customer into corresponding vectors to obtain initial customer information characteristic vectors;
and calculating the similarity between the historical customer information characteristic vector and the initial customer information characteristic vector, and obtaining the adjacent customers of the initial historical customer according to the similarity.
4. The fraud risk prediction method of claim 3, wherein said constructing a knowledge-graph of the historical set of clients from the historical set of client informational characteristics and the neighboring set of client informational characteristics comprises:
extracting entity vocabularies and relation vocabularies from the segmentation of the historical customer information characteristic set and the adjacent customer information characteristic set;
classifying the entity vocabulary and the relation vocabulary, and respectively storing the classification results of the entity vocabulary and the relation vocabulary into an entity vocabulary library and a relation vocabulary library;
and constructing a knowledge graph of the historical client set based on the entity vocabulary library and the relation vocabulary library.
5. The fraud risk prediction method of claim 4, wherein the weighting and accumulating the knowledge-graph of the historical customer using the pre-constructed graph learning model to generate the set of prediction scores for the set of historical customers comprises:
carrying out weighted summation operation on the knowledge graph by using an ith convolution layer in a pre-constructed graph learning model to obtain an ith node characterization vector, wherein i is 1,2 and 3 … n;
transferring the ith node characterization vector to an (i +1) th convolution layer through an activation function to carry out weighted summation to obtain an (i +1) th node characterization vector until i is n-1 to obtain an nth node characterization vector;
and classifying and scoring the n-th node characterization vector to obtain a prediction score set of the historical client set.
6. The fraud risk prediction method of claim 5, wherein the passing the i-th node token vector to the (i +1) -th convolutional layer through an activation function for weighted summation to obtain an (i +1) -th node token vector comprises:
calculating an (i +1) th node characterization vector by the following formula:
where σ () represents an activation function; l represents the first layer convolution layer, i represents the current node; r represents the relationship between nodes; r represents all relationships between nodes;
representing all node sets with a relation R with the current node i under the condition that R belongs to the R relation; c. Ci,rRepresenting a regularization constant;
a weight representing a self-loop;
a weight representing the node relationship r;
indicating that the current node i is at the l-th levelA feature vector of the convolutional layer;
represents the feature vector of node j in the first convolutional layer.
7. The fraud risk prediction method of claim 1, wherein said computing a set of fraud scores for a set of historical clients from the set of historical client information characteristics comprises:
selecting one of the historical clients from the historical client set, and acquiring the repayment date, the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data of the selected historical client from the historical client information characteristic set;
extracting the on-time repayment days, the advance repayment days and the overdue repayment days from the on-time repayment behavior data, the advance repayment behavior data and the overdue repayment behavior data respectively;
counting the repayment behavior days of the selected historical client according to the on-time repayment days, the advance repayment days and the overdue repayment days;
calculating a historical customer fraud score S for the selected historical customer by the following formula:
S＝D×F(x)×P
d represents the repayment behavior days, P represents the repayment period number, F (x) represents a normal distribution function, mu represents the average number of the historical customer information set, sigma represents the standard deviation of the historical customer information set, and x is a preset factor.
8. A fraud risk prediction apparatus, characterized in that the apparatus comprises:
the knowledge map building module is used for acquiring a historical client information set of a historical client set, extracting a historical client information characteristic set from the historical client information set and calculating a fraud score set corresponding to the historical client set according to the historical client information characteristic set; acquiring adjacent clients of each historical client through each historical client information characteristic in the historical client information characteristic set, and associating each historical client with the corresponding adjacent client; collecting the adjacent client information of each historical client to obtain an adjacent client information characteristic set; constructing a knowledge graph of the historical customer set according to the historical customer information characteristic set and the adjacent customer information characteristic set;
the score prediction module is used for performing weighting and accumulation operation on the knowledge graph by using a pre-constructed graph learning model to generate a prediction score set of the historical client set; calculating a loss value between the prediction score set and the fraud score set by using a loss function, and performing parameter adjustment on the graph learning model according to the loss value until the loss value is smaller than a preset loss threshold value to obtain a trained graph learning model; acquiring a knowledge graph of a customer to be analyzed, and predicting the knowledge graph of the customer to be analyzed by using the trained graph learning model to obtain a prediction score of the customer to be analyzed;
and the risk decision module is used for judging the fraud risk of the customer according to the prediction score of the customer to be analyzed and a preset warning threshold value.
9. An electronic device, characterized in that the electronic device comprises:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform a fraud risk prediction method according to any one of claims 1 to 7.
10. A computer-readable storage medium, storing a computer program, wherein the computer program, when executed by a processor, implements a fraud risk prediction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111009514.XA CN113706291A (en) | 2021-08-31 | 2021-08-31 | Fraud risk prediction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111009514.XA CN113706291A (en) | 2021-08-31 | 2021-08-31 | Fraud risk prediction method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113706291A true CN113706291A (en) | 2021-11-26 |
Family
ID=78657612
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111009514.XA Pending CN113706291A (en) | 2021-08-31 | 2021-08-31 | Fraud risk prediction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113706291A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115391414A (en) * | 2022-10-28 | 2022-11-25 | 北京双赢天下管理咨询有限公司 | Bank market expanding system and method based on big data |
| JP7292454B1 (en) | 2022-03-04 | 2023-06-16 | ＭＯＮＥＴ Ｔｅｃｈｎｏｌｏｇｉｅｓ株式会社 | Information processing device, information processing method and information processing program |
| CN116823451A (en) * | 2023-08-10 | 2023-09-29 | 杭银消费金融股份有限公司 | Credit risk control method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685647A (en) * | 2018-12-27 | 2019-04-26 | 阳光财产保险股份有限公司 | The training method of credit fraud detection method and its model, device and server |
| CN111813933A (en) * | 2020-06-18 | 2020-10-23 | 国网上海市电力公司 | Automatic identification method for technical field in technical atlas |
| CN112053222A (en) * | 2020-08-14 | 2020-12-08 | 百维金科(上海)信息科技有限公司 | Knowledge graph-based internet financial group fraud detection method |
| CN112561684A (en) * | 2020-12-15 | 2021-03-26 | 平安科技（深圳）有限公司 | Financial fraud risk identification method and device, computer equipment and storage medium |
-
2021
- 2021-08-31 CN CN202111009514.XA patent/CN113706291A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109685647A (en) * | 2018-12-27 | 2019-04-26 | 阳光财产保险股份有限公司 | The training method of credit fraud detection method and its model, device and server |
| CN111813933A (en) * | 2020-06-18 | 2020-10-23 | 国网上海市电力公司 | Automatic identification method for technical field in technical atlas |
| CN112053222A (en) * | 2020-08-14 | 2020-12-08 | 百维金科(上海)信息科技有限公司 | Knowledge graph-based internet financial group fraud detection method |
| CN112561684A (en) * | 2020-12-15 | 2021-03-26 | 平安科技（深圳）有限公司 | Financial fraud risk identification method and device, computer equipment and storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7292454B1 (en) | 2022-03-04 | 2023-06-16 | ＭＯＮＥＴ Ｔｅｃｈｎｏｌｏｇｉｅｓ株式会社 | Information processing device, information processing method and information processing program |
| JP2023129069A (en) * | 2022-03-04 | 2023-09-14 | ＭＯＮＥＴ Ｔｅｃｈｎｏｌｏｇｉｅｓ株式会社 | Information processing device, information processing method, and information processing program |
| CN115391414A (en) * | 2022-10-28 | 2022-11-25 | 北京双赢天下管理咨询有限公司 | Bank market expanding system and method based on big data |
| CN116823451A (en) * | 2023-08-10 | 2023-09-29 | 杭银消费金融股份有限公司 | Credit risk control method and system |
| CN116823451B (en) * | 2023-08-10 | 2024-03-26 | 杭银消费金融股份有限公司 | Credit risk control method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| CN113706291A (en) | Fraud risk prediction method, device, equipment and storage medium | |
| CN115002200B (en) | Message pushing method, device, equipment and storage medium based on user portrait | |
| CN112883190A (en) | Text classification method and device, electronic equipment and storage medium | |
| CN112988963B (en) | User intention prediction method, device, equipment and medium based on multi-flow nodes | |
| CN113626607B (en) | Abnormal work order identification method and device, electronic equipment and readable storage medium | |
| CN112883730B (en) | Similar text matching method and device, electronic equipment and storage medium | |
| CN115081025A (en) | Sensitive data management method and device based on digital middlebox and electronic equipment | |
| CN104951843A (en) | Sales forecasting system and method | |
| CN114880449B (en) | Method and device for generating answers of intelligent questions and answers, electronic equipment and storage medium | |
| CN113628043B (en) | Complaint validity judging method, device, equipment and medium based on data classification | |
| CN117155771B (en) | Equipment cluster fault tracing method and device based on industrial Internet of things | |
| CN113918704A (en) | Question-answering method and device based on machine learning, electronic equipment and medium | |
| CN114708073B (en) | Intelligent detection method and device for surrounding mark and serial mark, electronic equipment and storage medium | |
| CN113515591B (en) | Text defect information identification method and device, electronic equipment and storage medium | |
| CN114662794A (en) | Enterprise default risk prediction method, device, equipment and storage medium | |
| CN115099680A (en) | Risk management method, device, equipment and storage medium | |
| CN115099339A (en) | Fraud behavior identification method and device, electronic equipment and storage medium | |
| CN114202337A (en) | Risk identification method, device, equipment and storage medium | |
| CN113706207A (en) | Order transaction rate analysis method, device, equipment and medium based on semantic analysis | |
| CN114006986A (en) | Outbound call compliance early warning method, device, equipment and storage medium | |
| CN113888265A (en) | Product recommendation method, device, equipment and computer-readable storage medium | |
| CN112734205A (en) | Model confidence degree analysis method and device, electronic equipment and computer storage medium | |
| CN113688924B (en) | Abnormal order detection method, device, equipment and medium | |
| CN111680513B (en) | Feature information identification method and device and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20240517 |