CN113191072A - Suspicious transaction monitoring method and device based on longitudinal federal logistic regression - Google Patents
Suspicious transaction monitoring method and device based on longitudinal federal logistic regression Download PDFInfo
- Publication number
- CN113191072A CN113191072A CN202110339256.5A CN202110339256A CN113191072A CN 113191072 A CN113191072 A CN 113191072A CN 202110339256 A CN202110339256 A CN 202110339256A CN 113191072 A CN113191072 A CN 113191072A
- Authority
- CN
- China
- Prior art keywords
- financial institution
- institution system
- suspicious transaction
- transaction monitoring
- tagged
- 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 175
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000007477 logistic regression Methods 0.000 title claims abstract description 62
- 238000012546 transfer Methods 0.000 claims abstract description 127
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 24
- 230000002776 aggregation Effects 0.000 claims abstract description 22
- 238000004220 aggregation Methods 0.000 claims abstract description 22
- 238000012806 monitoring device Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 7
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000004891 communication Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 13
- 230000006870 function Effects 0.000 description 12
- 238000010801 machine learning Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 239000000872 buffer Substances 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000001617 migratory effect Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computer Security & Cryptography (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a suspicious transaction monitoring method and a suspicious transaction monitoring device based on longitudinal federal logistic regression, which relate to the field of artificial intelligence, and comprise the following steps: acquiring modeling demand information of each financial institution system for establishing an illegal fund transfer suspicious transaction monitoring model based on a longitudinal federal learning algorithm; determining a tagged financial institution system and a non-tagged financial institution system according to the modeling demand information; the non-tag financial institution system calculates model parameters based on locally stored user data; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; based on the model, suspicious transactions of illegal funds transfer in various financial institution systems are monitored. The invention can break through the data isolated island to complete the joint modeling of the illegal fund transfer suspicious transaction monitoring model.
Description
Technical Field
The invention relates to the field of artificial intelligence, in particular to a suspicious transaction monitoring method and device based on longitudinal federal logistic regression.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
With the application of emerging technologies such as big data and machine learning to risk prevention and control in the financial field, illegal fund transfer systems of large financial institutions begin to establish efficient suspicious transaction monitoring models by utilizing machine learning in many cases. However, most financial institutions have limited data quality, and are difficult to break data islands, so that the responsibility of ensuring privacy and safety of users is great, and the realization requirement of machine learning is difficult and serious.
At present, both the traditional rule-based suspicious transaction monitoring model and the advanced machine learning-based suspicious transaction monitoring model have obvious data quality bottlenecks in the aspects of optimizing iteration speed and accuracy, and are difficult to meet flexible human supervision requirements and illegal fund transfer means of changing endless tests.
Therefore, how to provide a suspicious transaction monitoring model for illegal fund transfer in most scenes under the condition of ensuring data security and user privacy is a technical problem to be solved urgently by an illegal fund transfer system of each large financial institution.
Disclosure of Invention
The embodiment of the invention provides a suspicious transaction monitoring method based on longitudinal federal logistic regression, which is used for solving the technical problems of iteration speed and accuracy bottleneck caused by limited data quality of the existing rule-based suspicious transaction monitoring model and the existing machine learning-based suspicious transaction monitoring model, and comprises the following steps: acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm; according to the modeling demand information, determining a tagged financial institution system and a non-tagged financial institution system, wherein the tagged financial institution system is a financial institution system with tag data, and the non-tagged financial institution system is a financial institution system without tag data; the non-tag financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data, and uploads the model parameters to the tag financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
Further, after determining the tagged financial institution system and the untagged financial institution system according to the modeling demand information, the method further comprises: determining user information with intersection among financial institution systems; and according to the user information with intersection among the financial institution systems, establishing data samples of the illegal fund transfer suspicious transaction monitoring models established by the financial institution systems based on a longitudinal federal learning algorithm.
Further, determining user information that an intersection exists between the financial institution systems comprises: encrypting the unique identity of each user in each financial institution system; and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
Further, before the untagged financial institution system calculates the model parameters of the illegal fund transfer suspicious transaction monitoring model based on the locally stored user data and uploads the model parameters to the tagged financial institution system, the method further comprises the following steps: the tagged financial institution system sends key information to the non-tagged financial institution system, wherein the key information is used for the non-tagged financial institution system to encrypt data to be exchanged with the tagged financial institution system.
Further, before the untagged financial institution system calculates the model parameters of the illegal fund transfer suspicious transaction monitoring model based on the locally stored user data and uploads the model parameters to the tagged financial institution system, the method further comprises the following steps: and the non-tag financial institution system encrypts the model parameters to be uploaded according to the key information.
Further, before the non-tagged financial institution system calculates the model parameters of the illegal funds transfer suspicious transaction monitoring model based on locally stored user data, the method further comprises: according to the modeling demand information of each financial institution system and locally stored user data, determining initial parameters and a characteristic vector matrix of the illegal fund transfer suspicious transaction monitoring model in each financial institution system.
Further, the method further comprises: determining a positive sample label and a negative sample label, wherein the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually approved as non-suspicious transaction information.
The embodiment of the invention also provides a suspicious transaction monitoring device based on longitudinal federal logistic regression, which is used for solving the technical problems of iteration speed and accuracy bottleneck caused by limited data quality of the existing rule-based suspicious transaction monitoring model and the existing machine learning-based suspicious transaction monitoring model, and the device comprises: the modeling demand acquisition module is used for acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm; the joint learning participant determining module is used for determining a tagged financial institution system and a non-tagged financial institution system according to the modeling demand information, wherein the tagged financial institution system is a financial institution system with tagged data, and the non-tagged financial institution system is a financial institution system without tagged data; the model local calculation module is used for calculating model parameters of the illegal fund transfer suspicious transaction monitoring model by the non-tag financial institution system based on locally stored user data and uploading the model parameters to the tag financial institution system; the local calculation result aggregation module is used for aggregating the model parameters of each non-tag financial institution system by using a logistic regression equation by the tagged financial institution system, and updating the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and the transaction monitoring module is used for monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
Further, the apparatus further comprises: the intersection user information determining module is used for determining user information with intersection among the financial institution systems; and the longitudinal federal learning sample determining module is used for constructing data samples of the suspicious transaction monitoring model for illegal fund transfer established by each financial institution system based on a longitudinal federal learning algorithm according to the user information with intersection among the financial institution systems.
Further, the intersection user information determination module is further configured to: encrypting the unique identity of each user in each financial institution system; and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
Further, the apparatus further comprises: and the key information issuing module is used for sending the key information to the non-tag financial institution system by the tag financial institution system, wherein the key information is used for encrypting data to be exchanged with the tag financial institution system by the non-tag financial institution system.
Further, the apparatus further comprises: and the encryption module is used for encrypting the model parameters to be uploaded by the non-tag financial institution system according to the key information.
Further, the apparatus further comprises: and the local model determining module is used for determining initial parameters and a characteristic vector matrix of the illegal fund transfer suspicious transaction monitoring model in each financial institution system according to the modeling demand information of each financial institution system and locally stored user data.
Further, the apparatus further comprises: the positive and negative sample label determining module is used for determining a positive sample label and a negative sample label, wherein the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually approved as non-suspicious transaction information.
The embodiment of the invention also provides electronic equipment for solving the technical problems of iteration speed and accuracy bottleneck caused by limited data quality of the conventional rule-based suspicious transaction monitoring model and the machine learning-based suspicious transaction monitoring model.
The embodiment of the invention also provides a computer readable storage medium, which stores a computer program for executing the suspicious transaction monitoring method based on the longitudinal federal logistic regression.
After modeling demand information of each financial institution system for establishing an illegal fund transfer suspicious transaction monitoring model based on a longitudinal federated learning algorithm is acquired, a labeled financial institution system with label data and a non-labeled financial institution system without label data are determined according to the modeling demand information, so that the non-labeled financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data and uploads the model parameters to the labeled financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and finally, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
By the embodiment of the invention, under the condition that all financial institutions do not share data, the data island is broken, the joint modeling of the illegal fund transfer suspicious transaction monitoring model is completed, and the monitoring of illegal fund transfer transactions in all financial institution systems is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
fig. 1 is a flowchart of a suspicious transaction monitoring method based on longitudinal federal logistic regression according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for constructing a longitudinal federated learning data sample provided in an embodiment of the present invention;
fig. 3 is a flow chart of data encryption provided in the embodiment of the present invention;
fig. 4 is a flowchart of a specific implementation of a suspicious transaction monitoring method based on longitudinal federal logistic regression according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a suspicious transaction monitoring device based on longitudinal federal logistic regression according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an optional longitudinal federal logistic regression-based suspicious transaction monitoring device according to an embodiment of the present invention;
fig. 7 is a schematic diagram of an electronic device provided in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
In order to deeply practice the working principles of risk, comprehensiveness, applicability, dynamic management and the like, establish a transaction monitoring standard of a sound financial institution, practically improve the effectiveness of suspicious transaction reporting work, and combine the self business characteristics, risk conditions and management modes of the financial institution.
Federated learning enables joint modeling based on distributed machine learning without data sharing. In the process of cooperatively training the combined model by each mechanism, the training data of each mechanism can be kept locally, and the data does not need to be summarized like the traditional machine learning method. And a global sharing model is jointly established through a parameter exchange mode under a federal system encryption mechanism, and the established model only serves local targets in respective areas.
The federal learning framework has a breakthrough in overcoming the computational performance and improving the safety of federal learning, and the communication cost, unbalanced data distribution and equipment reliability in large-scale distribution are main factors for optimization. In addition, the data is divided by user ID or device ID, so the data dimension is distributed horizontally, and the data privacy in the decentralized collaborative learning setting is considered.
The application modes of federal learning are generally divided into three categories: horizontal federal learning, vertical federal learning, and migratory federal learning. The characteristic of horizontal federal learning is that the businesses (feature matrices) of all organizations are similar, but the samples (client IDs) are different or have smaller intersection, namely the quantity of the abundant samples under the condition of sharing the same feature matrix (feature vector space), for example, the transaction data of the respective clients of two city businesses, and the quantity of the samples can be increased by horizontal data combination; longitudinal federal learning is opposite to transverse learning, and is characterized in that a large number of samples (client IDs) are repeated in a training data set, but user behavior characteristics are less overlapped, such as transaction data and navigation data of banks, and the server can increase the dimensionality of the user characteristics by utilizing the data combination; federal migration learning applies to situations where the two data sets differ not only in sample but also in feature space, such as certain banks of china and amazon in the united states. In the embodiment of the invention, joint modeling is carried out based on the longitudinal federal logistic regression learning technology, and a more efficient monitoring model for suspicious transactions of illegal fund transfer is established by effectively combining local special data of each participant.
In the above application context, an embodiment of the present invention provides a suspicious transaction monitoring method based on longitudinal federal logistic regression, and fig. 1 is a flowchart of the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention, as shown in fig. 1, the method includes the following steps:
s101, obtaining modeling demand information of each financial institution system, wherein the modeling demand information is demand information of an illegal fund transfer suspicious transaction monitoring model established for each financial institution system based on a longitudinal federal learning algorithm.
Common illegal fund transfer suspicious transactions widely relate to various fields such as banks, insurance, securities and the like. Thus, the financial institution systems may be, but are not limited to, business systems of various banking institutions, insurance companies, and securities companies.
The longitudinal federated learning algorithm is beneficial to establishing cooperation among financial institutions, and a stronger model is established together by using respective unique data.
And S102, determining a tagged financial institution system and a non-tagged financial institution system according to the modeling demand information, wherein the tagged financial institution system is a financial institution system with tagged data, and the non-tagged financial institution system is a financial institution system without tagged data.
It should be noted that, in the embodiment of the present invention, the tagged financial institution system is a main participant for modeling based on the federal learning algorithm; the non-tagged financial institution system is a secondary participant modeled based on a federated learning algorithm.
And S103, calculating model parameters of the illegal fund transfer suspicious transaction monitoring model by the non-tag financial institution system based on locally stored user data, and uploading the model parameters to the tagged financial institution system.
It should be noted that the non-tag financial institution system may perform local machine learning based on respective locally stored data to obtain model parameters for modeling, and upload the model parameters to the tagged financial institution system, so that the tagged financial institution system performs aggregation and summation on the model parameters of the non-tag financial institution systems to construct a more powerful model.
S104, the tagged financial institution systems aggregate the model parameters of the untagged financial institution systems by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to an aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting preset conditions is obtained;
it should be noted that, if the model parameters of the illegal fund transfer suspicious transaction monitoring model are updated according to the aggregation result, and the obtained model does not satisfy the preset condition, S103 and S104 are executed again until the illegal fund transfer suspicious transaction monitoring model satisfying the preset condition (for example, the model accuracy is higher than the preset accuracy threshold or the model error rate is lower than the preset error rate threshold) is obtained.
And S105, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
After each financial institution system obtains the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions based on the federal learning algorithm, the illegal fund transfer suspicious transaction in each financial institution system can be monitored according to the model.
As can be seen from the above, in the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiments of the present invention, after the modeling demand information of the suspicious transaction monitoring model for illegal fund transfer established by each financial institution system based on the longitudinal federal learning algorithm is acquired, the tagged financial institution system having tagged data and the untagged financial institution system not having tagged data are determined according to the modeling demand information, so that the untagged financial institution system calculates model parameters of the suspicious transaction monitoring model for illegal fund transfer based on locally stored user data, and uploads the model parameters to the tagged financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and finally, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
By the suspicious transaction monitoring method based on longitudinal federal logistic regression provided by the embodiment of the invention, under the condition that each financial institution does not share data, a data isolated island is broken, the joint modeling of the suspicious transaction monitoring model of illegal fund transfer is completed, and the monitoring of illegal fund transfer transactions in each financial institution system is realized.
In an embodiment, as shown in fig. 2, after determining the tagged financial institution system and the untagged financial institution system according to the modeling demand information, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention may further construct a longitudinal federal learned data sample by the following steps:
s201, determining user information with intersection among financial institution systems;
s202, according to the user information with intersection among the financial institution systems, establishing data samples of the illegal fund transfer suspicious transaction monitoring models established by the financial institution systems based on the longitudinal federal learning algorithm.
Further, when determining that there is user information of intersection between the financial institution systems, the method can be implemented by the following steps: encrypting the unique identity of each user in each financial institution system; and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
In an embodiment, as shown in fig. 3, before the non-tagged financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data and uploads the model parameters to the tagged financial institution system, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention may further include the following steps:
s301, the system of the tagged financial institution sends key information to the system of the untagged financial institution, wherein the key information is used for the system of the untagged financial institution to encrypt data to be exchanged with the system of the tagged financial institution.
Further, as shown in fig. 3, before the non-tagged financial institution system calculates the model parameters of the illegal fund transfer suspicious transaction monitoring model based on the locally stored user data and uploads the model parameters to the tagged financial institution system, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention may further include the following steps:
and S302, encrypting the model parameters to be uploaded by the non-tag financial institution system according to the key information.
In an embodiment, before the non-tag financial institution system calculates the model parameters of the illegal money transfer suspicious transaction monitoring model based on the locally stored user data, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention may further include the following steps: according to the modeling demand information of each financial institution system and locally stored user data, determining initial parameters and a characteristic vector matrix of the illegal fund transfer suspicious transaction monitoring model in each financial institution system.
Further, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided in the embodiment of the present invention may further include the following steps: determining a positive sample label and a negative sample label, wherein the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually approved as non-suspicious transaction information.
In the embodiment of the invention, based on the penetrating type pedestrian supervision requirement and the mature federal learning technology, by researching the suspicious transaction monitoring system and the data condition of the existing financial institution and combining various scenes, the monitoring of the illegal fund transfer suspicious transaction based on longitudinal federal learning is realized.
Fig. 4 is a flowchart of a specific implementation of a suspicious transaction monitoring method based on longitudinal federal logistic regression provided in an embodiment of the present invention, and as shown in fig. 4, the method specifically includes:
s401, determining the model requirements of each participant: each participant (client) proposes a joint modeling assumption according to own requirements, establishes joint modeling requirements after multiple parties reach an agreement, and defines the monitoring scene of the model for banks or other financial institutions with suspicious transaction monitoring requirements.
S402, determining a tag data holder and a non-tag data holder under different requirements: because the method is based on longitudinal federal learning, the characteristic of the technology determines that the multi-party enterprise of the combined modeling has small feature overlapping degree but high user overlapping degree, so the multi-party enterprise basically requires that the users of the multi-party enterprise are in the same region (country) as much as possible, and each enterprise is distributed in a plurality of industries; therefore, the modeling requirements of each participant are different, the patent stands at a financial institution with a bank as a main point, and a suspicious transaction monitoring method based on longitudinal federal learning is designed. Therefore, the step is to determine the label of the bank, where the suspicious report of the bank reported in the bank history is obtained as the positive sample label, and the report that is manually checked as non-suspicious is obtained as the negative sample label.
S403, determining a tag data holder and a non-tag data holder under different requirements; because the longitudinal federated learning technology needs to ensure that the sample IDs are consistent, namely, the users have intersection, the unique identifiers (such as identity numbers) of the users (samples) of all the participants need to be aligned, and because the step of changing relates to the user privacy data, the step of changing needs to finish the selection of the intersection users by encrypting the unique identifiers of the users.
S404, determining the feature vectors and the initial parameters of each participant model: determining initial parameters and characteristic vector matrixes of respective models according to own modeling requirements and own data conditions of each participant
S405, the label data direction sends modeling key information to the label-free data side: the tagged data party distributes modeling key information to other parties, and the key information is used for encrypting data needing to be exchanged.
S406, model local calculation: and each participant carries out parameter calculation based on the local data and uploads the calculation result to the tagged data holder.
S407, the tagged data party aggregates the multi-party calculation results: and the tagged data party aggregates the calculation results of the plurality of non-tagged data parties to obtain the sum, and the final output is obtained by utilizing a logistic regression equation.
S408, updating the data side model with the label: and the data side with the label updates the model parameters according to the aggregation result.
S409, evaluating whether the updated model meets the expected demand.
If yes, executing S410 to exit the joint modeling process; if the loop S405 to S408 is not satisfied.
And S410, finishing training.
Therefore, the suspicious transaction monitoring method based on longitudinal federal logistic regression provided by the embodiment of the invention can realize but is limited to the following technical effects: the method is suitable for development of suspicious monitoring models in most illegal fund transfer scenes, efficiently completes modeling tasks, reduces development amount and ensures model effect; on the premise of ensuring data safety and user privacy, the value of mining multi-party data is fully utilized, so that the characteristic dimension of the suspicious transaction monitoring model for illegal fund transfer is enriched, the performance bottleneck of the past monitoring model is broken through, and the win-win situation is realized; and the problem of data island is broken, the low data quality of each participating enterprise is improved, and the efficiency performance of the machine learning model is greatly improved.
Based on the same inventive concept, the embodiment of the present invention further provides a suspicious transaction monitoring device based on longitudinal federal logistic regression, as in the following embodiments. Because the problem solving principle of the device is similar to that of the suspicious transaction monitoring method based on the longitudinal federal logistic regression, the implementation of the device can refer to the implementation of the suspicious transaction monitoring method based on the longitudinal federal logistic regression, and repeated parts are not repeated.
Fig. 5 is a schematic diagram of a suspicious transaction monitoring device based on longitudinal federal logistic regression according to an embodiment of the present invention, as shown in fig. 5, the device includes: the system comprises a modeling demand acquisition module 501, a federal learning participant determination module 502, a model partial calculation module 503, a partial calculation result aggregation module 504 and a transaction monitoring module 505.
The modeling demand acquisition module 501 is used for acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm; the federal learning participant determining module 502 is used for determining a tagged financial institution system and a non-tagged financial institution system according to the modeling demand information, wherein the tagged financial institution system is a financial institution system with tagged data, and the non-tagged financial institution system is a financial institution system without tagged data; the model local calculation module 503 is used for calculating the model parameters of the illegal fund transfer suspicious transaction monitoring model by the non-tag financial institution system based on the locally stored user data, and uploading the model parameters to the tag financial institution system; the local calculation result aggregation module 504 is configured to aggregate the model parameters of the non-tagged financial institution systems by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and the transaction monitoring module 505 is configured to monitor the illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset condition.
It should be noted here that the modeling requirement obtaining module 501, the federal learning participant determining module 502, the model partial calculation module 503, the partial calculation result aggregation module 504, and the transaction monitoring module 505 correspond to S101 to S105 in the method embodiment, and the modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules described above as part of an apparatus may be implemented in a computer system such as a set of computer-executable instructions.
As can be seen from the above, the suspicious transaction monitoring device based on longitudinal federal logistic regression provided in the embodiment of the present invention determines the tagged financial institution system having the tagged data and the untagged financial institution system not having the tagged data according to the modeling demand information after acquiring the modeling demand information of the suspicious transaction monitoring model for illegal fund transfer established by each financial institution system based on the longitudinal federal learning algorithm, so that the untagged financial institution system calculates the model parameters of the suspicious transaction monitoring model for illegal fund transfer based on the locally stored user data, and uploads the model parameters to the tagged financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and finally, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
By the suspicious transaction monitoring device based on longitudinal federal logistic regression provided by the embodiment of the invention, under the condition that each financial institution does not share data, a data isolated island is broken, the joint modeling of the suspicious transaction monitoring model of illegal fund transfer is completed, and the monitoring of illegal fund transfer transactions in each financial institution system is realized.
In an embodiment, as shown in fig. 6, the suspicious transaction monitoring apparatus based on longitudinal federal logistic regression provided in an embodiment of the present invention further includes: an intersection user information determination module 506 and a longitudinal federal learning sample determination module 507.
The intersection user information determining module 506 is configured to determine user information with an intersection between the financial institution systems; and the longitudinal federal learning sample determination module 507 is used for constructing data samples of the suspicious transaction monitoring model for illegal fund transfer established by each financial institution system based on a longitudinal federal learning algorithm according to the user information with intersection among the financial institution systems.
It should be noted here that the intersection user information determining module 506 and the vertical federal learning sample determining module 507 correspond to S201 to S202 in the method embodiment, and the modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules described above as part of an apparatus may be implemented in a computer system such as a set of computer-executable instructions.
In one embodiment, the intersection user information determining module 506 is further configured to: encrypting the unique identity of each user in each financial institution system; and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
In an embodiment, as shown in fig. 6, the suspicious transaction monitoring apparatus based on longitudinal federal logistic regression provided in an embodiment of the present invention further includes: and a key information issuing module 508, configured to send key information to the untagged financial institution system by the tagged financial institution system, where the key information is used for the untagged financial institution system to encrypt data to be exchanged with the tagged financial institution system.
Further, based on the above embodiment, as shown in fig. 6, the suspicious transaction monitoring device based on longitudinal federal logistic regression provided in the embodiment of the present invention further includes: and an encryption module 509, configured to encrypt the model parameter to be uploaded according to the key information by the non-tag financial institution system.
It should be noted here that the key information issuing module 508 and the encryption module 509 correspond to S301 to S302 in the method embodiment, and the modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the method embodiment. It should be noted that the modules described above as part of an apparatus may be implemented in a computer system such as a set of computer-executable instructions.
In an embodiment, as shown in fig. 6, the suspicious transaction monitoring apparatus based on longitudinal federal logistic regression provided in an embodiment of the present invention further includes: and the local model determining module 510 is configured to determine initial parameters and a feature vector matrix of the suspicious transaction monitoring model for illegal fund transfer in each financial institution system according to the modeling requirement information of each financial institution system and locally stored user data.
In an embodiment, as shown in fig. 6, the suspicious transaction monitoring apparatus based on longitudinal federal logistic regression provided in an embodiment of the present invention further includes: the positive and negative sample label determining module 511 is configured to determine a positive sample label and a negative sample label, where the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually approved as non-suspicious transaction information.
Based on the same inventive concept, the embodiment of the present invention further provides an embodiment of an electronic device for implementing all or part of the contents of the above suspicious transaction monitoring method based on longitudinal federal logistic regression. The electronic device specifically comprises the following contents:
a processor (processor), a memory (memory), a communication Interface (Communications Interface), and a bus; the processor, the memory and the communication interface complete mutual communication through the bus; the communication interface is used for realizing information transmission between related devices; the electronic device may be a desktop computer, a tablet computer, a mobile terminal, and the like, but the embodiment is not limited thereto. In this embodiment, the electronic device may be implemented with reference to the embodiment of the method for monitoring suspicious transactions based on longitudinal federal logistic regression and the embodiment of the device for monitoring suspicious transactions based on longitudinal federal logistic regression, which are incorporated herein, and repeated details are not repeated herein.
Fig. 7 is a schematic diagram of a system configuration structure of an electronic device according to an embodiment of the present invention. As shown in fig. 7, the electronic device 70 may include a processor 701 and a memory 702; a memory 702 is coupled to the processor 701. Notably, this fig. 7 is exemplary; other types of structures may also be used in addition to or in place of the structure to implement telecommunications or other functions.
In one embodiment, the functionality implemented by the longitudinal federal logistic regression-based suspicious transaction monitoring method may be integrated into the processor 701. Wherein, the processor 701 may be configured to control as follows: acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm; according to the modeling demand information, determining a tagged financial institution system and a non-tagged financial institution system, wherein the tagged financial institution system is a financial institution system with tag data, and the non-tagged financial institution system is a financial institution system without tag data; the non-tag financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data, and uploads the model parameters to the tag financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
As can be seen from the above, after obtaining the modeling demand information of the illegal fund transfer suspicious transaction monitoring model established by each financial institution system based on the longitudinal federal learning algorithm, the electronic device provided in the embodiment of the present invention determines the tagged financial institution system with the tagged data and the untagged financial institution system without the tagged data according to the modeling demand information, so that the untagged financial institution system calculates the model parameters of the illegal fund transfer suspicious transaction monitoring model based on the locally stored user data, and uploads the model parameters to the tagged financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and finally, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
By the electronic equipment provided by the embodiment of the invention, under the condition that all financial institutions do not share data, the data isolated island is broken, the joint modeling of the illegal fund transfer suspicious transaction monitoring model is completed, and the monitoring of illegal fund transfer transactions in the financial institution systems is realized.
In another embodiment, the suspicious transaction monitoring device based on longitudinal federal logistic regression may be configured separately from the processor 701, for example, the suspicious transaction monitoring device based on longitudinal federal logistic regression may be configured as a chip connected to the processor 701, and the function of the suspicious transaction monitoring method based on longitudinal federal logistic regression is implemented by the control of the processor.
As shown in fig. 7, the electronic device 70 may further include: a communication module 703, an input unit 704, an audio processing unit 705, a display 706, and a power supply 707. It is noted that the electronic device 70 does not necessarily include all of the components shown in fig. 7; furthermore, the electronic device 70 may also comprise components not shown in fig. 7, which can be referred to in the prior art.
As shown in fig. 7, the processor 701, which is sometimes referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, and the processor 701 receives input and controls the operation of the various components of the electronic device 70.
The memory 702 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information relating to the failure may be stored, and a program for executing the information may be stored. And the processor 701 may execute the program stored in the memory 702 to realize information storage or processing, or the like.
The input unit 704 provides input to the processor 701. The input unit 704 is, for example, a key or a touch input device. The power supply 707 is used to supply power to the electronic device 70. The display 706 is used for displaying display objects such as images and characters. The display may be, for example, an LCD display, but is not limited thereto.
The memory 702 may be a solid state memory such as Read Only Memory (ROM), Random Access Memory (RAM), a SIM card, or the like. There may also be a memory that holds information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes called an EPROM or the like. The memory 702 may also be some other type of device. Memory 702 includes a buffer memory 7021 (sometimes referred to as a buffer). The memory 702 may include an application/function storage portion 7022, the application/function storage portion 7022 being used to store application programs and function programs or procedures for performing operations of the electronic device 70 by the processor 701.
The memory 702 may also include a data store 7023, the data store 7023 being for storing data, such as contacts, digital data, pictures, sounds, and/or any other data used by the electronic device. The driver storage portion 7024 of the memory 702 may include various drivers of the electronic device for communication functions and/or for performing other functions of the electronic device (e.g., messaging application, address book application, etc.).
The communication module 703 is a transmitter/receiver that transmits and receives signals via the antenna 708. A communication module (transmitter/receiver) 703 is coupled to the processor 701 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 703, such as a cellular network module, a bluetooth module, and/or a wireless local area network module, may be provided in the same electronic device. The communication module (transmitter/receiver) 703 is also coupled to a speaker 709 and a microphone 710 via an audio processing unit 705 to provide audio output via the speaker 709 and receive audio input from the microphone 710 to implement general telecommunication functions. The audio processing unit 705 may include any suitable buffers, decoders, amplifiers and so forth. Additionally, an audio processing unit 705 is also coupled to the processor 701 to enable recording of sound locally through a microphone 710 and to enable playing of locally stored sound through a speaker 709.
An embodiment of the present invention further provides a computer-readable storage medium for implementing all the steps in the method for monitoring suspicious transactions based on longitudinal federated logistic regression in the foregoing embodiments, where the computer-readable storage medium stores thereon a computer program, and when the computer program is executed by a processor, the computer program implements all the steps of the method for monitoring suspicious transactions based on longitudinal federated logistic regression in the foregoing embodiments, for example, when the processor executes the computer program, the processor implements the following steps: acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm; according to the modeling demand information, determining a tagged financial institution system and a non-tagged financial institution system, wherein the tagged financial institution system is a financial institution system with tag data, and the non-tagged financial institution system is a financial institution system without tag data; the non-tag financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data, and uploads the model parameters to the tag financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
As can be seen from the above, the computer-readable storage medium provided in the embodiment of the present invention, after obtaining the modeling demand information of the illegal fund transfer suspicious transaction monitoring model established by each financial institution system based on the longitudinal federal learning algorithm, determines the tagged financial institution system having the tag data and the non-tagged financial institution system not having the tag data according to the modeling demand information, so that the non-tagged financial institution system calculates the model parameters of the illegal fund transfer suspicious transaction monitoring model based on the locally stored user data, and uploads the model parameters to the tagged financial institution system; the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained; and finally, monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
By the computer-readable storage medium provided by the embodiment of the invention, under the condition that all financial institutions do not share data, the joint modeling of the illegal fund transfer suspicious transaction monitoring model is completed by breaking a data island, and the monitoring of illegal fund transfer transactions in all financial institution systems is realized.
Although the present invention provides method steps as described in the examples or flowcharts, more or fewer steps may be included based on routine or non-inventive labor. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an actual apparatus or client product executes, it may execute sequentially or in parallel (e.g., in the context of parallel processors or multi-threaded processing) according to the embodiments or methods shown in the figures.
As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, apparatus (system) or computer program product. Accordingly, embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment. In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "upper", "lower", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience in describing the present invention and simplifying the description, but do not indicate or imply that the referred devices or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Unless expressly stated or limited otherwise, the terms "mounted," "connected," and "connected" are intended to be inclusive and mean, for example, that they may be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention is not limited to any single aspect, nor is it limited to any single embodiment, nor is it limited to any combination and/or permutation of these aspects and/or embodiments. Each aspect and/or embodiment of the invention can be used alone or in combination with one or more other aspects and/or embodiments.
The above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.
Claims (16)
1. A suspicious transaction monitoring method based on longitudinal federated logistic regression is characterized by comprising the following steps:
acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm;
according to the modeling demand information, determining a tagged financial institution system and a non-tagged financial institution system, wherein the tagged financial institution system is a financial institution system with tag data, and the non-tagged financial institution system is a financial institution system without tag data;
the non-tag financial institution system calculates model parameters of the illegal fund transfer suspicious transaction monitoring model based on locally stored user data, and uploads the model parameters to the tag financial institution system;
the tagged financial institution systems aggregate the model parameters of each non-tagged financial institution system by using a logistic regression equation, and update the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained;
and monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
2. The method of claim 1, wherein after determining a tagged financial institution system and an untagged financial institution system based on the modeling demand information, the method further comprises:
determining user information with intersection among financial institution systems;
and according to the user information with intersection among the financial institution systems, establishing data samples of the illegal fund transfer suspicious transaction monitoring models established by the financial institution systems based on a longitudinal federal learning algorithm.
3. The method of claim 2, wherein determining user information that intersections exist between financial institution systems comprises:
encrypting the unique identity of each user in each financial institution system;
and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
4. The method of claim 1, wherein before the untagged financial institution system calculates model parameters of the illegal funds transfer suspicious transaction monitoring model based on locally stored user data and uploading to the tagged financial institution system, the method further comprises:
the tagged financial institution system sends key information to the non-tagged financial institution system, wherein the key information is used for the non-tagged financial institution system to encrypt data to be exchanged with the tagged financial institution system.
5. The method of claim 4, wherein before the untagged financial institution system calculates model parameters of the illegal funds transfer suspicious transaction monitoring model based on locally stored user data and uploading to the tagged financial institution system, the method further comprises:
and the non-tag financial institution system encrypts the model parameters to be uploaded according to the key information.
6. The method of claim 1, wherein prior to the untagged financial institution system calculating the model parameters of the illegal funds transfer suspicious transaction monitoring model based on locally stored user data, the method further comprises:
according to the modeling demand information of each financial institution system and locally stored user data, determining initial parameters and a characteristic vector matrix of the illegal fund transfer suspicious transaction monitoring model in each financial institution system.
7. The method of claim 1, wherein the method further comprises:
determining a positive sample label and a negative sample label, wherein the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually checked as non-suspicious transaction information.
8. A suspicious transaction monitoring device based on longitudinal federated logistic regression is characterized by comprising:
the modeling demand acquisition module is used for acquiring modeling demand information of each financial institution system, wherein the modeling demand information is demand information for establishing an illegal fund transfer suspicious transaction monitoring model for each financial institution system based on a longitudinal federal learning algorithm;
the federal learning participant determining module is used for determining a tagged financial institution system and a non-tagged financial institution system according to the modeling demand information, wherein the tagged financial institution system is a financial institution system with tagged data, and the non-tagged financial institution system is a financial institution system without tagged data;
the model local calculation module is used for calculating model parameters of the illegal fund transfer suspicious transaction monitoring model by the non-tag financial institution system based on locally stored user data and uploading the model parameters to the tag financial institution system;
the local calculation result aggregation module is used for aggregating the model parameters of each non-tag financial institution system by using a logistic regression equation by the tagged financial institution system, and updating the model parameters of the illegal fund transfer suspicious transaction monitoring model according to the aggregation result until the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions is obtained;
and the transaction monitoring module is used for monitoring illegal fund transfer suspicious transactions in each financial institution system based on the illegal fund transfer suspicious transaction monitoring model meeting the preset conditions.
9. The apparatus of claim 8, wherein the apparatus further comprises:
the intersection user information determining module is used for determining user information with intersection among the financial institution systems;
and the longitudinal federal learning sample determining module is used for constructing data samples of the suspicious transaction monitoring model for illegal fund transfer established by each financial institution system based on a longitudinal federal learning algorithm according to the user information with intersection among the financial institution systems.
10. The apparatus of claim 9, wherein the intersection user information determination module is further to:
encrypting the unique identity of each user in each financial institution system;
and determining user information with intersection among the financial institution systems based on the encrypted unique identity.
11. The apparatus of claim 8, wherein the apparatus further comprises:
and the key information issuing module is used for sending the key information to the non-tag financial institution system by the tag financial institution system, wherein the key information is used for encrypting data to be exchanged with the tag financial institution system by the non-tag financial institution system.
12. The apparatus of claim 11, wherein the apparatus further comprises:
and the encryption module is used for encrypting the model parameters to be uploaded by the non-tag financial institution system according to the key information.
13. The apparatus of claim 8, wherein the apparatus further comprises:
and the local model determining module is used for determining initial parameters and a characteristic vector matrix of the illegal fund transfer suspicious transaction monitoring model in each financial institution system according to the modeling demand information of each financial institution system and locally stored user data.
14. The apparatus of claim 8, wherein the apparatus further comprises:
the positive and negative sample label determining module is used for determining a positive sample label and a negative sample label, wherein the positive sample label is suspicious transaction information historically reported by each financial institution system; the negative sample label is manually checked as non-suspicious transaction information.
15. An electronic device comprising a memory, a processor and a computer program stored on the memory and operable on the processor, wherein the processor, when executing the computer program, implements the method for monitoring suspicious transactions based on longitudinal federal logistic regression according to any one of claims 1 to 7.
16. A computer-readable storage medium storing a computer program for executing the longitudinal federal logistic regression-based suspicious transaction monitoring method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110339256.5A CN113191072A (en) | 2021-03-30 | 2021-03-30 | Suspicious transaction monitoring method and device based on longitudinal federal logistic regression |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110339256.5A CN113191072A (en) | 2021-03-30 | 2021-03-30 | Suspicious transaction monitoring method and device based on longitudinal federal logistic regression |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113191072A true CN113191072A (en) | 2021-07-30 |
Family
ID=76974606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110339256.5A Pending CN113191072A (en) | 2021-03-30 | 2021-03-30 | Suspicious transaction monitoring method and device based on longitudinal federal logistic regression |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113191072A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417388A (en) * | 2022-01-25 | 2022-04-29 | 云南电网有限责任公司信息中心 | Power load prediction method, system, equipment and medium based on longitudinal federal learning |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948704A (en) * | 2019-03-20 | 2019-06-28 | ***股份有限公司 | A kind of transaction detection method and apparatus |
| CN110276618A (en) * | 2019-06-28 | 2019-09-24 | 第四范式（北京）技术有限公司 | The method and system for generating money laundering ancestor prediction model, predicting money laundering ancestor |
| CN110659989A (en) * | 2019-09-10 | 2020-01-07 | 马洪富 | Active exploration type compliance anti-money laundering method, device, system and storage medium |
| CN111383113A (en) * | 2020-03-06 | 2020-07-07 | 深圳前海微众银行股份有限公司 | Suspicious client prediction method, device, equipment and readable storage medium |
| CN111722043A (en) * | 2020-06-29 | 2020-09-29 | 南方电网科学研究院有限责任公司 | Power equipment fault detection method, device and system |
| CN112101950A (en) * | 2020-09-27 | 2020-12-18 | 中国建设银行股份有限公司 | Suspicious transaction monitoring model feature extraction method and device |
| WO2021004551A1 (en) * | 2019-09-26 | 2021-01-14 | 深圳前海微众银行股份有限公司 | Method, apparatus, and device for optimization of vertically federated learning system, and a readable storage medium |
| CN112434818A (en) * | 2020-11-19 | 2021-03-02 | 脸萌有限公司 | Model construction method, device, medium and electronic equipment |
-
2021
- 2021-03-30 CN CN202110339256.5A patent/CN113191072A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109948704A (en) * | 2019-03-20 | 2019-06-28 | ***股份有限公司 | A kind of transaction detection method and apparatus |
| CN110276618A (en) * | 2019-06-28 | 2019-09-24 | 第四范式（北京）技术有限公司 | The method and system for generating money laundering ancestor prediction model, predicting money laundering ancestor |
| CN110659989A (en) * | 2019-09-10 | 2020-01-07 | 马洪富 | Active exploration type compliance anti-money laundering method, device, system and storage medium |
| WO2021004551A1 (en) * | 2019-09-26 | 2021-01-14 | 深圳前海微众银行股份有限公司 | Method, apparatus, and device for optimization of vertically federated learning system, and a readable storage medium |
| CN111383113A (en) * | 2020-03-06 | 2020-07-07 | 深圳前海微众银行股份有限公司 | Suspicious client prediction method, device, equipment and readable storage medium |
| CN111722043A (en) * | 2020-06-29 | 2020-09-29 | 南方电网科学研究院有限责任公司 | Power equipment fault detection method, device and system |
| CN112101950A (en) * | 2020-09-27 | 2020-12-18 | 中国建设银行股份有限公司 | Suspicious transaction monitoring model feature extraction method and device |
| CN112434818A (en) * | 2020-11-19 | 2021-03-02 | 脸萌有限公司 | Model construction method, device, medium and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114417388A (en) * | 2022-01-25 | 2022-04-29 | 云南电网有限责任公司信息中心 | Power load prediction method, system, equipment and medium based on longitudinal federal learning |
| CN114417388B (en) * | 2022-01-25 | 2022-08-26 | 云南电网有限责任公司信息中心 | Power load prediction method, system, equipment and medium based on longitudinal federal learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110229B (en) | Information recommendation method and device | |
| CN111311410B (en) | Community problem multi-department cooperative processing system and method based on block chain | |
| CN112003858B (en) | Block chain-based platform docking method, electronic device and storage medium | |
| CN112199709A (en) | Multi-party based privacy data joint training model method and device | |
| CN111563267A (en) | Method and device for processing federal characteristic engineering data | |
| CN113568973B (en) | Financial credit investigation data sharing method and device based on blockchain and federal learning | |
| WO2022156594A1 (en) | Federated model training method and apparatus, electronic device, computer program product, and computer-readable storage medium | |
| CN113034281A (en) | Service data processing method based on block chain, related system and node | |
| CN113409134A (en) | Enterprise financing trust method and device based on federal learning | |
| CN113051239A (en) | Data sharing method, use method of model applying data sharing method and related equipment | |
| CN112907243A (en) | Block chain transaction auditing method and device | |
| CN113836809A (en) | Cross-industry data joint modeling method and system based on block chain and federal learning | |
| CN113191072A (en) | Suspicious transaction monitoring method and device based on longitudinal federal logistic regression | |
| CN113792890B (en) | Model training method based on federal learning and related equipment | |
| CN112860805A (en) | Block chain data interaction method and system | |
| CN111680968B (en) | Building rights management system and method based on block chain | |
| CN112949866A (en) | Poisson regression model training method and device, electronic equipment and storage medium | |
| WO2023124219A1 (en) | Joint learning model iterative update method, apparatus, system, and storage medium | |
| CN111881166A (en) | Operation data processing method, device and system based on block chain | |
| CN112884562A (en) | Block chain-based mortgage processing method and device and readable storage medium | |
| CN113900828B (en) | Special processor for federal learning, federal learning processing chip and chip | |
| CN114723012A (en) | Computing method and device based on distributed training system | |
| Pandey et al. | Leveraging Blockchain Technology in Industry 4.0 and Industrial Internet of Things (IIoT) Scenarios | |
| CN114092226A (en) | Method and device for recommending foreign exchange products of bank outlets | |
| CN112417018A (en) | Data sharing method and device |
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 |