CN113781052A - Anti-money laundering monitoring method, device, equipment and storage medium - Google Patents
Anti-money laundering monitoring method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113781052A CN113781052A CN202111042299.3A CN202111042299A CN113781052A CN 113781052 A CN113781052 A CN 113781052A CN 202111042299 A CN202111042299 A CN 202111042299A CN 113781052 A CN113781052 A CN 113781052A
- Authority
- CN
- China
- Prior art keywords
- target
- information
- money laundering
- sample set
- monitoring model
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 208
- 238000004900 laundering Methods 0.000 title claims abstract description 187
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000003860 storage Methods 0.000 title claims abstract description 30
- 238000003062 neural network model Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 44
- 238000012806 monitoring device Methods 0.000 claims abstract 2
- 239000013598 vector Substances 0.000 claims description 86
- 238000009826 distribution Methods 0.000 claims description 51
- 239000011159 matrix material Substances 0.000 claims description 31
- 238000006243 chemical reaction Methods 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 19
- 238000010606 normalization Methods 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 230000003287 optical effect Effects 0.000 description 8
- 230000006978 adaptation Effects 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000013307 optical fiber Substances 0.000 description 3
- 238000013526 transfer learning Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 239000003814 drug Substances 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- GWAOOGWHPITOEY-UHFFFAOYSA-N 1,5,2,4-dioxadithiane 2,2,4,4-tetraoxide Chemical compound O=S1(=O)CS(=O)(=O)OCO1 GWAOOGWHPITOEY-UHFFFAOYSA-N 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Computer Security & Cryptography (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an anti-money laundering monitoring method, an anti-money laundering monitoring device, anti-money laundering monitoring equipment and a storage medium. The method comprises the following steps: acquiring a target domain sample set; inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Description
Technical Field
The embodiment of the invention relates to the technical field of financial risk assessment, in particular to an anti-money laundering monitoring method, device, equipment and storage medium.
Background
The anti-money laundering work in the financial field is a machine learning application scene with complex business, and often faces the problems of small data scale and few samples. Under the scene of cross-mechanism and cross-region cooperation, the characteristics of large characteristic difference, data distribution offset and the like exist. These problems result in that model direct reuse will not satisfy the prerequisite condition of independent and identically distributed modeling data of traditional machine learning.
The current risk monitoring scheme is: and (3) self-defining rules, wherein business experts analyze historical data characteristics based on actual business scenes by researching the financial supervision requirements of China, mine valuable clues and risk doubt points, and self-define a series of quantifiable suspicious monitoring rules by taking risk identification as a core. Therefore, the risk scene discrimination is met, and monitoring rules are expanded and adjusted in time according to the supervision risk prompt and the financial industry. This is called a custom rule, and is widely used in risk monitoring.
The existing traditional custom rule mainly evaluates the model through the experience of professionals, and manually configures the model rule, which has the major disadvantage that 1, a large amount of manpower and time are required to be invested. 2. The manual configuration is to design a rule model on the premise of prejudging possible risk directions, and rules are possibly omitted and are not accurate and comprehensive enough, so that user characteristics cannot be accurately defined. In view of the above problems, some researchers have introduced a machine learning method into the client risk identification, and obtain a rule model through model training. However, for existing model training methods, a large amount of data is usually required to be trained, and training data lacks pre-screening, resulting in higher training costs. And due to the special service scene, the problems of less sample data and small scale are often faced, so that a machine learning algorithm cannot obtain a model for accurately reflecting characteristics due to lack of training data, and the characteristic of data distribution deviation exists in the learning of different monitoring models, so that the modeling condition that the traditional machine learning requires independent and same distribution cannot be met.
Disclosure of Invention
The embodiment of the invention provides an anti-money laundering monitoring method, device, equipment and storage medium, which aim to solve the problems of high training cost, less training data and small scale and can improve the accuracy of anti-money laundering monitoring.
In a first aspect, an embodiment of the present invention provides an anti-money laundering monitoring method, including: acquiring a target domain sample set;
inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Further, the first anti-money laundering monitoring model is established according to risk prompt information, and comprises:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
Further, the target domain samples include: first transaction information, first customer information, and first account information;
correspondingly, a second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, and the method comprises the following steps:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
Further, inputting the target domain sample set into the target anti-money laundering monitoring model, and obtaining target label information corresponding to the target domain sample set includes:
inputting the target domain sample set into the first anti-money laundering monitoring model to obtain first target label information;
determining the same distribution information of a target domain sample set according to the conversion matrix and the target domain feature vector;
inputting the target domain sample set and distribution information into the second anti-money laundering monitoring model to obtain second target label information;
and determining target label information according to the first target label information and the second target label information.
Further, determining a target anti-money laundering model according to the first anti-money laundering monitoring model and the second anti-money laundering monitoring model includes:
acquiring the maximum mean difference of the same distribution information of the source domain sample set and the same distribution information of the target domain sample set;
determining a weight of the first anti-money laundering monitoring model and a weight of the second anti-money laundering monitoring model according to the maximum mean difference;
and determining a target anti-money laundering model according to the first anti-money laundering monitoring model, the weight of the second anti-money laundering monitoring model and the weight of the second anti-money laundering monitoring model.
In a second aspect, an embodiment of the present invention further provides an anti-money laundering monitoring apparatus, including:
the acquisition module is used for acquiring a target domain sample set;
a monitoring module, configured to input the target domain sample set into the target anti-money laundering monitoring model, to obtain target tag information corresponding to the target domain sample set, where the target anti-money laundering monitoring model includes: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Further, the monitoring module is specifically configured to:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
Further, the target domain samples include: first transaction information, first customer information, and first account information;
correspondingly, the monitoring module is specifically configured to:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the anti-money laundering monitoring method according to any one of the embodiments of the present invention.
In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the anti-money laundering monitoring method according to any one of the embodiments of the present invention.
The embodiment of the invention obtains a target domain sample set; inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, so that the problems of high training cost, less training data and small scale are solved, and the accuracy of anti-money laundering monitoring can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a flow chart of an anti-money laundering monitoring method in an embodiment of the present invention;
FIG. 2 is a schematic diagram of an anti-money laundering monitoring apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device in an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer-readable storage medium containing a computer program in an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures. In addition, the embodiments and features of the embodiments in the present invention may be combined with each other without conflict.
Before discussing exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart may describe the operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently or simultaneously. In addition, the order of the operations may be re-arranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figure. The processes may correspond to methods, functions, procedures, subroutines, and the like. In addition, the embodiments and features of the embodiments in the present invention may be combined with each other without conflict.
The term "include" and variations thereof as used herein are intended to be open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment".
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Fig. 1 is a flowchart of an anti-money laundering monitoring method according to an embodiment of the present invention, where the embodiment is applicable to an anti-money laundering monitoring situation, and the method may be executed by an anti-money laundering monitoring apparatus according to an embodiment of the present invention, where the anti-money laundering monitoring apparatus may be implemented in a software and/or hardware manner, as shown in fig. 1, and the method specifically includes the following steps:
and S110, acquiring a target domain sample set.
Specifically, the target domain sample set may be obtained in a manner as follows: the method includes the steps of presetting a first rule, extracting a sample from a database according to the first rule, and constructing a target domain sample set, where the first rule may be a rule related to anti-money laundering, and the target domain sample set may trigger an anti-money laundering model, and embodiments of the present invention are not limited to this.
S120, inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
The source domain sample set may be obtained in a manner of: presetting a second rule, extracting a sample from the database according to the second rule, and constructing a source domain sample set, wherein the second rule is different from the first rule, the second rule can be a rule related to zero-package vending, and the source domain sample set can trigger a zero-package vending model.
The first anti-money laundering monitoring model can be established according to the risk prompt information in a mode of: acquiring risk prompt information; determining a target rule according to the risk prompt information; and establishing a first anti-money laundering monitoring model according to the target rule.
Specifically, the way of obtaining the second anti-money laundering monitoring model by iteratively training the neural network model through the target domain sample set and the source domain sample set may be: obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information; respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector; respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector; inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information; inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information; determining target domain data according to the target domain feature vector and the first label information; determining source domain data according to the source domain feature vector and the second label information; determining a conversion matrix according to the target domain data and the source domain data; determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector; inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information; training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information; and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained. The method for obtaining the second anti-money laundering monitoring model by iteratively training the neural network model through the target domain sample set and the source domain sample set can also be as follows: determining a conversion matrix according to a target domain sample set and a source domain sample set, and determining the same distribution information of the source domain sample set according to the conversion matrix and the source domain eigenvector; inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information; training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information; and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
Optionally, the first anti-money laundering monitoring model is established according to risk prompt information, and includes:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
The risk prompt information may be a risk prompt text or other content related to the risk prompt information, which is not limited in this embodiment of the present invention.
Specifically, the first anti-money laundering monitoring model is established according to the target rule, for example, a pedestrian risk prompting message and expert experience are obtained to establish a rule main point corresponding to the first anti-money laundering monitoring model, and the first anti-money laundering monitoring model is established according to the rule main point.
Optionally, the target domain samples include: first transaction information, first customer information, and first account information;
correspondingly, a second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, and the method comprises the following steps:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
The zero-package vending monitoring model may be a pre-trained model, which is not limited in this embodiment of the present invention.
The method for determining the target domain data according to the target domain feature vector and the first tag information may be: obtaining a target domain feature vector corresponding to each target domain sample in a target domain sample set and first label information corresponding to each target domain feature vector, and determining target domain data according to the target domain feature vector corresponding to each target domain sample and the first label information corresponding to each target domain feature vector, for example, obtaining a target domain feature vector X corresponding to a target domain sampletThe target domain feature vector set corresponding to the target domain sample set is XT,XTComprising a plurality of target domain feature vectors Xt. Target domain feature vector XtInputting a zero package drug vending monitoring model to obtain first label information YtFirst label information set Y corresponding to target domain sample setTIncluding a plurality of first label information Yt. According to the target domain feature vector XtAnd first label information YtDetermining target domain data (X)T,YT)。
The method for determining the source domain data according to the source domain feature vector and the second tag information may be: obtaining a source domain feature vector corresponding to each source domain sample in a source domain sample set and second label information corresponding to each source domain feature vector, and determining source domain data according to the source domain feature vector corresponding to each source domain sample and the second label information corresponding to each source domain feature vector, for example, obtaining a source domain feature vector X corresponding to a source domain samplesThe source domain feature vector set corresponding to the source domain sample set is XS,XSComprising a plurality of source domain feature vectors XsThe source domain feature vector XsInputting a zero package drug vending monitoring model to obtain second label information YsSecond label information set Y corresponding to source domain sample setSIncluding a plurality of second label information Ys. According to the source domain feature vector XsAnd second label information YsDetermining source domain data (X)S,YS)。
The method for determining the transformation matrix according to the target domain data and the source domain data may be: based on the target domain data and the source domain data, a transformation matrix is obtained by using a joint distribution adaptation method, for example, the transformation matrix can be source domain data (X)S,YS) Target Domain data (X)T,YT). The joint distribution adaptation target is to find a transformation matrix A such that P (A) after transformationTXS) And P (A)TXT) Can be as close as possible, while P (Y) is presentS|ATXS) And P (Y)T|ATXT) Is also small.
The method for determining the same distribution information of the source domain sample set according to the transformation matrix and the source domain feature vector may be: and mapping the source domain feature vector by using the conversion matrix to obtain the same distribution information of the source domain sample set.
Optionally, inputting the target domain sample set into the target anti-money laundering monitoring model, and obtaining target label information corresponding to the target domain sample set includes:
inputting the target domain sample set into the first anti-money laundering monitoring model to obtain first target label information;
determining the same distribution information of a target domain sample set according to the conversion matrix and the target domain feature vector;
inputting the target domain sample set and distribution information into the second anti-money laundering monitoring model to obtain second target label information;
and determining target label information according to the first target label information and the second target label information.
Optionally, determining a target anti-money laundering model according to the first anti-money laundering monitoring model and the second anti-money laundering monitoring model includes:
acquiring the maximum mean difference of the same distribution information of the source domain sample set and the same distribution information of the target domain sample set;
determining a weight of the first anti-money laundering monitoring model and a weight of the second anti-money laundering monitoring model according to the maximum mean difference;
and determining a target anti-money laundering model according to the first anti-money laundering monitoring model, the weight of the second anti-money laundering monitoring model and the weight of the second anti-money laundering monitoring model.
The method for determining the weight of the first anti-money laundering monitoring model and the weight of the second anti-money laundering monitoring model according to the maximum mean difference may be: determining a weight of the second anti-money laundering monitoring model based on the maximum mean difference, determining a weight of the first anti-money laundering monitoring model based on the weight of the second anti-money laundering monitoring model, which may be, for example,weight, w, for the second anti-money laundering monitoring model1+w21, and may further be according to w2To obtain w1,w1A weight of the first anti-money laundering monitoring model.
Specifically, a source domain sample set and a target domain sample set are obtained, and features in the source domain sample set and features in the target domain sample set are mapped to a common domain of the source domain sample set and the target domain sample set by using joint distribution adaptation. And measuring the difference of the adapted source domain and the target domain through the maximum mean difference, and taking the reciprocal of the difference as an adaptive weight.
In a specific example, the embodiment of the invention provides a training method of an anti-money laundering monitoring model, and mainly aims to combine transfer learning with a custom rule to improve the risk identification effect of anti-money laundering.
The main technical scheme comprises the following steps:
and establishing a first anti-money laundering monitoring model according to the risk prompt text and expert experience, wherein the first anti-money laundering monitoring model is f (x).
And acquiring a source domain sample set and a target domain sample set, wherein the source domain sample and the target domain sample are both samples used for training the target anti-money laundering monitoring model. The source domain sample set is a sample set capable of triggering a zero package vending monitoring model, and the target domain sample set is a sample set capable of triggering an anti-money laundering monitoring model. The source domain samples include: the attributes and transaction type sample information of the client, such as transaction amount, transaction time, transaction direction, transaction mode, client age, role, occupation, associator and the like. The target domain sample set comprises: the attributes and transaction type sample information of the client, such as transaction amount, transaction time, transaction direction, transaction mode, client age, role, occupation, associator and the like.
Mapping the source domain sample and the target domain sample to the same feature space through feature transformation, enabling the distance between the transformed source domain sample and the transformed target domain sample to be as close as possible, and training to obtain a second anti-money laundering monitoring model g (x), wherein the specific training method is as follows:
the method comprises the following steps: the source domain sample set and the target domain sample set are preprocessed. The target domain samples include: first transaction information, first customer information, and first account information, the source domain sample comprising: and respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information and then splicing to obtain a target domain feature vector, and respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information and then splicing to obtain a source domain feature vector, wherein the source domain feature vector is used for fully mining the potential values of the two features so as to enable the anti-money laundering model learning sample to be more comprehensive.
Step two: and inputting the source domain feature vector subjected to the normalization processing into a zero-package vending monitoring model to obtain second label information.
Step three: inputting the target domain feature vector into a zero-packet vending monitoring model to obtain first label information Yt. The first label information is predicted by the zero-packet vending monitoring model according to the target domain feature vectorRather than in the actual service.
Step four: determining target domain data according to the target domain feature vector and the first label information, determining source domain data according to the source domain feature vector and the second label information, and obtaining a transformation matrix according to the source domain data and the target domain data by using a joint distribution adaptation method. The method comprises the following specific steps:
source domain data (X)S,YS) Target Domain data (X)T,YT). The joint distribution adaptation target is to find a transformation matrix A such that P (A) after transformationTXS) And P (A)TXT) Can be as close as possible, while P (Y) is presentS|ATXS) And P (Y)T|ATXT) Is also small. When edge distribution adaption is carried out, the maximum mean difference MMD of the source domain characteristic vector and the target domain characteristic vector is obtained, and the source domain characteristic vector and the target domain characteristic vector are mapped to a public domain space by searching a conversion matrix A, so that the MMD is minimum; when condition distribution is adapted, mapping the source domain eigenvector and the target domain eigenvector to a public domain space by searching a transformation matrix A, wherein A is respectivelyTXS、ATXTThen measures P (Y) by MMDS|ATXS) And P (Y)T|ATXT) Such that P (Y)S|ATXS) And P (Y)T|ATXT) Is the smallest.
The two distances are combined to obtain a total optimization target, and the optimization target is minimized through continuous iteration to obtain a transformation matrix A.
Step five: mapping the source domain characteristic vector by using a conversion matrix A to obtain source domain sample set homography information, and inputting the source domain sample set homography information into a neural network model to obtain prediction label information; training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information; and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
Step six: inputting a target domain sample set into the first anti-money laundering monitoring model to obtain first target label information, and inputting the target domain sample set and distribution information into the second anti-money laundering monitoring model to obtain second target label information.
Weighting the first anti-money laundering monitoring model and the second anti-money laundering monitoring model, and processing to obtain a final target anti-money laundering monitoring model, namely w1f(x)+w2g (x) due to w1+w21, the above formula can be converted to (1-w)2)f(x)+w2g (x). Calculating w2The formula of (1) is as follows:
measure the difference between the source domain and the target domain by the maximum mean difference, thenThe final anti-money laundering model is then:
the embodiment of the invention provides an anti-money laundering model training method combining transfer learning and custom rules. Establishing an incidence relation A between the customer information and the label information based on a rule model; and establishing an incidence relation B between the customer information and the label information based on the transfer learning model, and weighting the incidence relation A and the incidence relation B to obtain a target anti-money laundering monitoring model. The rule model is a money laundering suspected transaction monitoring model formed on the basis of a user-defined rule, the migration learning model is a money laundering suspected transaction monitoring model based on a joint distribution adaptation algorithm, and the customer information is transaction data, account information, user identity and the like.
According to the technical scheme of the embodiment, a target domain sample set is obtained; inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, so that the problems of high training cost, less training data and small scale are solved, and the accuracy of anti-money laundering monitoring can be improved.
Fig. 2 is a schematic structural diagram of an anti-money laundering monitoring apparatus according to an embodiment of the present invention. The present embodiment may be applicable to the anti-money laundering monitoring, the apparatus may be implemented in software and/or hardware, and the apparatus may be integrated into any device providing an anti-money laundering monitoring function, as shown in fig. 2, where the anti-money laundering monitoring apparatus specifically includes: an acquisition module 210 and a monitoring module 220.
The obtaining module 210 is configured to obtain a target domain sample set;
a monitoring module 220, configured to input the target domain sample set into the target anti-money laundering monitoring model, to obtain target tag information corresponding to the target domain sample set, where the target anti-money laundering monitoring model includes: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Optionally, the monitoring module is specifically configured to:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
Optionally, the target domain samples include: first transaction information, first customer information, and first account information;
correspondingly, the monitoring module is specifically configured to:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
The product can execute the method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
According to the technical scheme of the embodiment, a target domain sample set is obtained; inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, so that the problems of high training cost, less training data and small scale are solved, and the accuracy of anti-money laundering monitoring can be improved.
Fig. 3 is a schematic structural diagram of an electronic device according to a third embodiment of the present invention. FIG. 3 illustrates a block diagram of an electronic device 312 suitable for use in implementing embodiments of the present invention. The electronic device 312 shown in fig. 3 is only an example, and should not bring any limitation to the functions and the scope of the use of the embodiment of the present invention. Device 312 is a computing device for typical trajectory fitting functions.
As shown in fig. 3, electronic device 312 is in the form of a general purpose computing device. The components of the electronic device 312 may include, but are not limited to: one or more processors 316, a storage device 328, and a bus 318 that couples the various system components including the storage device 328 and the processors 316.
The processor 316 executes various functional applications and data processing by executing programs stored in the storage 328, for example, implementing the anti-money laundering monitoring method provided by the above-described embodiment of the present invention:
acquiring a target domain sample set;
inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Fig. 4 is a schematic structural diagram of a computer-readable storage medium containing a computer program according to an embodiment of the present invention. Embodiments of the present invention provide a computer-readable storage medium 61 having stored thereon a computer program 610, which when executed by one or more processors, implements an anti-money laundering monitoring method as provided by all embodiments of the invention of the present application:
acquiring a target domain sample set;
inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
Any combination of one or more computer-readable media may be employed. The computer readable medium may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (Hyper Text Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, or the like, as well as conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. Where the name of an element does not in some cases constitute a limitation on the element itself.
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. An anti-money laundering monitoring method, comprising:
acquiring a target domain sample set;
inputting the target domain sample set into the target anti-money laundering monitoring model to obtain target label information corresponding to the target domain sample set, wherein the target anti-money laundering monitoring model comprises: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
2. The method of claim 1, wherein the first anti-money laundering monitoring model is established based on risk hint information, comprising:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
3. The method of claim 1, wherein the target domain samples comprise: first transaction information, first customer information, and first account information;
correspondingly, a second anti-money laundering monitoring model is obtained by iteratively training a neural network model through the target domain sample set and the source domain sample set, and the method comprises the following steps:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
4. The method of claim 3, wherein inputting the target domain sample set into the target anti-money laundering monitoring model, and obtaining target label information corresponding to the target domain sample set comprises:
inputting the target domain sample set into the first anti-money laundering monitoring model to obtain first target label information;
determining the same distribution information of a target domain sample set according to the conversion matrix and the target domain feature vector;
inputting the target domain sample set and distribution information into the second anti-money laundering monitoring model to obtain second target label information;
and determining target label information according to the first target label information and the second target label information.
5. The method of claim 4, wherein determining a target anti-money laundering model from the first anti-money laundering monitoring model and the second anti-money laundering monitoring model comprises:
acquiring the maximum mean difference of the same distribution information of the source domain sample set and the same distribution information of the target domain sample set;
determining a weight of the first anti-money laundering monitoring model and a weight of the second anti-money laundering monitoring model according to the maximum mean difference;
and determining a target anti-money laundering model according to the first anti-money laundering monitoring model, the weight of the second anti-money laundering monitoring model and the weight of the second anti-money laundering monitoring model.
6. An anti-money laundering monitoring device, comprising:
the acquisition module is used for acquiring a target domain sample set;
a monitoring module, configured to input the target domain sample set into the target anti-money laundering monitoring model, to obtain target tag information corresponding to the target domain sample set, where the target anti-money laundering monitoring model includes: the system comprises a first anti-money laundering monitoring model and a second anti-money laundering monitoring model, wherein the first anti-money laundering monitoring model is established according to risk prompt information, and the second anti-money laundering monitoring model is obtained by iteratively training a neural network model through a target domain sample set and a source domain sample set.
7. The apparatus according to claim 6, wherein the monitoring module is specifically configured to:
acquiring risk prompt information;
determining a target rule according to the risk prompt information;
and establishing a first anti-money laundering monitoring model according to the target rule.
8. The apparatus of claim 6, wherein the target domain samples comprise: first transaction information, first customer information, and first account information;
correspondingly, the monitoring module is specifically configured to:
obtaining a set of source domain samples, wherein the source domain samples comprise: second transaction information, second customer information, and second account information;
respectively carrying out normalization processing on the first transaction information, the first customer information and the first account information, and then splicing to obtain a target domain feature vector;
respectively carrying out normalization processing on the second transaction information, the second customer information and the second account information, and then splicing to obtain a source domain feature vector;
inputting the target domain feature vector into a zero-package vending monitoring model to obtain first label information;
inputting the source domain feature vector into the zero package vending monitoring model to obtain second label information;
determining target domain data according to the target domain feature vector and the first label information;
determining source domain data according to the source domain feature vector and the second label information;
determining a conversion matrix according to the target domain data and the source domain data;
determining the same distribution information of a source domain sample set according to the conversion matrix and the source domain feature vector;
inputting the source domain sample set and the distribution information into a neural network model to obtain prediction label information;
training parameters of the neural network model according to an objective function generated by the predicted label information and the second label information;
and returning to execute the operation of inputting the source domain sample set and the distribution information into the neural network model to obtain the predicted label information until a second anti-money laundering monitoring model is obtained.
9. An electronic device, comprising:
one or more processors;
a memory for storing one or more programs;
the one or more programs, when executed by the one or more processors, cause the processors to implement the method of any of claims 1-5.
10. A computer-readable storage medium containing a computer program, on which the computer program is stored, characterized in that the program, when executed by one or more processors, implements the method according to any one of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111042299.3A CN113781052A (en) | 2021-09-07 | 2021-09-07 | Anti-money laundering monitoring method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111042299.3A CN113781052A (en) | 2021-09-07 | 2021-09-07 | Anti-money laundering monitoring method, device, equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113781052A true CN113781052A (en) | 2021-12-10 |
Family
ID=78841450
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111042299.3A Pending CN113781052A (en) | 2021-09-07 | 2021-09-07 | Anti-money laundering monitoring method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113781052A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20200372424A1 (en) * | 2018-02-12 | 2020-11-26 | Alibaba Group Holding Limited | System and method for generating risk-control rules |
CN112347392A (en) * | 2020-10-21 | 2021-02-09 | 上海淇玥信息技术有限公司 | Anti-fraud assessment method and device based on transfer learning and electronic equipment |
CN112634048A (en) * | 2020-12-30 | 2021-04-09 | 第四范式(北京)技术有限公司 | Anti-money laundering model training method and device |
CN113095927A (en) * | 2021-02-23 | 2021-07-09 | 广发证券股份有限公司 | Method and device for identifying suspicious transactions of anti-money laundering |
-
2021
- 2021-09-07 CN CN202111042299.3A patent/CN113781052A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20200372424A1 (en) * | 2018-02-12 | 2020-11-26 | Alibaba Group Holding Limited | System and method for generating risk-control rules |
CN112347392A (en) * | 2020-10-21 | 2021-02-09 | 上海淇玥信息技术有限公司 | Anti-fraud assessment method and device based on transfer learning and electronic equipment |
CN112634048A (en) * | 2020-12-30 | 2021-04-09 | 第四范式(北京)技术有限公司 | Anti-money laundering model training method and device |
CN113095927A (en) * | 2021-02-23 | 2021-07-09 | 广发证券股份有限公司 | Method and device for identifying suspicious transactions of anti-money laundering |
Non-Patent Citations (1)
Title |
---|
"基于迁移学习技术的反洗钱建模研究和实现", 中国金融电脑, no. 10, 7 October 2020 (2020-10-07), pages 48 - 53 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11348352B2 (en) | Contract lifecycle management | |
WO2019233421A1 (en) | Image processing method and device, electronic apparatus, and storage medium | |
WO2022077646A1 (en) | Method and apparatus for training student model for image processing | |
CN112949415B (en) | Image processing method, apparatus, device and medium | |
EP3637310A1 (en) | Method and apparatus for generating vehicle damage information | |
US20220292861A1 (en) | Docket Analysis Methods and Systems | |
US20220092407A1 (en) | Transfer learning with machine learning systems | |
CN110490304B (en) | Data processing method and device | |
CN110929802A (en) | Information entropy-based subdivision identification model training and image identification method and device | |
CN112883990A (en) | Data classification method and device, computer storage medium and electronic equipment | |
CN110781970A (en) | Method, device and equipment for generating classifier and storage medium | |
CN114357170A (en) | Model training method, analysis method, device, equipment and medium | |
CN112766284A (en) | Image recognition method and device, storage medium and electronic equipment | |
CN113379059A (en) | Model training method for quantum data classification and quantum data classification method | |
CN110009154B (en) | Refund prediction method and device, terminal equipment and storage medium | |
CN111161238A (en) | Image quality evaluation method and device, electronic device, and storage medium | |
CN111160410A (en) | Object detection method and device | |
CN115482436B (en) | Training method and device for image screening model and image screening method | |
CN114741697B (en) | Malicious code classification method and device, electronic equipment and medium | |
CN113781052A (en) | Anti-money laundering monitoring method, device, equipment and storage medium | |
CN113255819B (en) | Method and device for identifying information | |
CN113360672B (en) | Method, apparatus, device, medium and product for generating knowledge graph | |
CN112541557B (en) | Training method and device for generating countermeasure network and electronic equipment | |
CN113989562A (en) | Model training and image classification method and device | |
CN113496231B (en) | Classification model training method, image classification method, device, equipment and 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 |