CN113240520A - Method for realizing rural finance three-resource supervision by using algorithm link based on data model - Google Patents
Method for realizing rural finance three-resource supervision by using algorithm link based on data model Download PDFInfo
- Publication number
- CN113240520A CN113240520A CN202110258472.7A CN202110258472A CN113240520A CN 113240520 A CN113240520 A CN 113240520A CN 202110258472 A CN202110258472 A CN 202110258472A CN 113240520 A CN113240520 A CN 113240520A
- Authority
- CN
- China
- Prior art keywords
- data
- financial
- node
- link
- supervision
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
- G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/64—Protecting data integrity, e.g. using checksums, certificates or signatures
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- 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
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Computer Security & Cryptography (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Technology Law (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a method for realizing rural finance three-resource supervision based on an algorithm link of a data model. The problems that the number of data required to be collected is large, the attribute dimensionality is wide, the data utilization rate is low, and the financial supervision difficulty is high are solved; the invention comprises the following steps: s1: collecting and summarizing fund inflow or fund outflow information of each distributed node required by rural financial supervision to form a financial link; s2: performing dimension algorithm analysis by using the link node data block, performing a training early warning model by using a logistic regression algorithm, and supervising node transaction in the financial link; s3: and inputting information of different characteristic dimensions required to be inquired by each node into the early warning model, judging whether the financial node has abnormal movement, and performing abnormal movement supervision. The online operation traceability is guaranteed through the financial link relation, risk management is conducted on the whole transaction process, the operation cost is reduced, the transaction efficiency is improved, and faults and attacks are effectively prevented.
Description
Technical Field
The invention relates to the field of rural financial supervision, in particular to a method for realizing rural financial three-resource supervision by an algorithm link based on a data model.
Background
At present, the rural financial fund service has the phenomenon of diversification, the supervision difficulty of the rural financial fund is high, and the informatization and internet technology application is relatively weak. For the individual fund payment and income functions of the current common rural intelligent platform and E-commerce platform, supervision means and measures are lacked, and the requirement of village-level administrative units on fund regulation and control cannot be met.
In the existing supervision mode, data acquisition and business document recording are required to be realized. The method has the advantages of more discrete small-granularity data, wide attribute dimension, poor data utilization rate, low value and no help for financial fund supervision and statistics.
For example, a "CN 107563936A" disclosed in chinese patent literature, whose publication number is a rural asset integrated management platform, includes an asset management data center construction module, a collective three-asset supervision service subsystem module and a property right transaction management service subsystem module; the asset management data center construction module is used for realizing the acquisition of the rural asset basic information; through the arrangement of the whole comprehensive management platform, the asset management data center construction module, the collective three-asset supervision service subsystem module and the property right transaction management service subsystem module realize an integrated processing platform. The scheme has the advantages of more discrete data acquisition, wide attribute dimensionality, low data utilization rate and high financial supervision difficulty.
Disclosure of Invention
The invention mainly solves the problems of more data, wide attribute dimension, poor data utilization rate and great financial supervision difficulty which are required to be acquired by rural financial supervision; the method for realizing rural finance three-resource supervision based on the algorithm link of the data model is characterized in that data collected by each node are processed according to characteristic dimensions and then are led into a database to form the node model, financial links are sequentially formed, the utilization rate of discrete data is improved, operation traceability is guaranteed through link relations, settlement efficiency is improved, and operation cost is reduced.
The technical problem of the invention is mainly solved by the following technical scheme:
a method for realizing rural finance three-resource supervision based on an algorithm link of a data model comprises the following steps:
s1: collecting and summarizing fund inflow or fund outflow information of each distributed node required by rural finance three-fund supervision, and respectively storing the information in each node to form a financial link;
s2: performing dimension algorithm analysis by using the link node data block, performing a training early warning model by using a logistic regression algorithm, and supervising node transaction in the financial link;
s3: and inputting information of different characteristic dimensions required to be inquired by each node into the early warning model, judging whether the financial node has abnormal movement, and performing abnormal movement supervision.
And the online operation traceability is ensured through the financial link relationship so as to carry out risk management on the whole transaction process. The data link property extends the traceable depth of financial supervision, and once abnormal data are monitored, data tracing can be carried out.
Preferably, the step S1 includes the following steps:
s11: processing the collected information data from each node into a numerical type, classifying and importing the numerical type information data into a database according to the characteristic dimension of the numerical type information data to form a data warehouse;
s12: different characteristic dimensions are analyzed to combine different node models;
s13: encrypting the node model data by using an asymmetric encryption algorithm to form a node data block;
s14: repeating the steps S11-S13 for the subsequent nodes of the data in sequence until all node data blocks are formed;
s15: and initiating data, namely taking the first node data block as the starting point of the link, and sequentially connecting all the node data blocks by using the data structure of the linked list to form a unidirectional data chain so as to form the financial link.
The financial link can improve payment, transaction and settlement efficiency. The process that the transaction is confirmed is the process of clearing, collection and auditing, and the financial link uses the data modular link to track, all transactions are displayed on the platform in real time, and the efficiency is greatly improved. Financial links can increase efficiency to the minute level, which can reduce settlement risk by 99%, effectively reducing capital costs and systemic risks. Financial links can reduce operating costs. The operation of each business system and background in the countryside often faces long flow and multiple links, which causes high cost of internal accounting and time expenditure and brings risk to capital. The financial link can simplify and automate a lengthy financial service process, reduce interaction between a foreground and a background, save a large amount of manpower and material resources, and has important significance for optimizing a financial platform business process and improving the competitiveness of a financial institution.
Preferably, the step S1 further includes the following steps:
s16: introducing a country financial fund supervision service, perfecting a fund transaction system and establishing a system for perfecting financial fund supervision;
s17: and establishing a monitoring node by utilizing the transparency and traceability of node data, monitoring the data and operation processing, and establishing a wind control analysis model of the big data model.
Preferably, more than three features are defined as high-dimensional, and less than three features are defined as low-dimensional; analyzing different characteristic dimensions comprises analyzing main components of high-dimensional data by selecting proper target dimensions, and reducing the dimensions of the data in a weight reduction mode; and averaging and weighting the low-dimensional data to improve the dimensionality.
Preferably, the rural financial link platform is converted into a financial asset exchange to trade the digital assets by changing the levy, the equity, the contract and the bill representing the financing requirement into unique and uncopyable digital assets; the finished financial resource supervision system comprises modules of business management, agricultural and economic reports, production and improvement data, bank and agricultural direct connection, early warning supervision, data analysis, loan application, transaction disclosure and the like.
Preferably, the wind control analysis model of the big data model comprises a data acquisition and analysis unit, a third-party data access unit, a house loan unit, a car loan unit and a credit card data unit; the data acquisition unit comprises portrait acquisition, voice acquisition, fingerprint acquisition, iris acquisition and basic information input.
Preferably, the step S2 includes the following steps:
s21: collecting data generated in a financial link in a mode of customer field service input or network browser customer input, wherein the data comprises data of customer names, ages, birth addresses, household registers, contact ways, bank accounts and fund transactions;
s22: analyzing data, and analyzing the existence condition of the bidirectional active edge in the fund transaction network by using the screened data through a plurality of screening modes, including age screening, household registration screening or fund transaction amount screening in recent years;
s23, training algorithm, obtaining the best classification regression coefficient by combining logistic regression algorithm training algorithm with Sigmoid function training;
s24, testing the algorithm, inputting the testing data to test the effectiveness and the matching degree of the algorithm, and outputting the completed early warning model when the effectiveness and the matching degree are both larger than the preset threshold value; otherwise, the process returns to step S23.
Preferably, the logistic regression algorithm repeats the following steps R times with regression coefficients initialized to 1: calculating gradient using step gradient of the whole data set, updating a vector of a regression coefficient by using a Sigmoid function, and returning the regression coefficient;
the Sigmoid function is:
wherein z is the input of the Sigmoid function; vector x is the input data to the classifier and vector w is the regression coefficients.
The Sigmoid function is also called Logistic function, and the value range is (0,1) and is used for two classifications.
Preferably, the step S3 includes the following steps:
s31: inputting data to be queried and converting the data into corresponding structured numerical values;
s32: and carrying out regression calculation on the input numerical value based on the regression coefficient of the trained early warning model, judging that the inquired node has abnormal motion when the output result of the model is greater than a set threshold value, and otherwise, judging that the node has no abnormal motion.
Preferably, the threshold is obtained from historical transaction data, and the average of the historical transaction data is taken.
The invention has the beneficial effects that:
1. and the scattered rural financial data is integrated into a financial link, so that the utilization value and the utilization rate of the data are improved.
2. The online operation traceability is ensured through the financial link relationship so as to carry out risk management on the whole transaction process, the link property of the data extends the traceable depth of financial supervision, and once abnormal data is monitored, the data traceability can be carried out.
3. The financial link uses the data modular link to track, clear up in real time, and efficiency is greatly improved, and the financial link can improve efficiency to minute level, lets settlement risk reduce 99%, thereby effectively reduces capital cost and systematic risk.
4. The financial link is simplified, the lengthy financial service process is automated, the interaction between a foreground and a background is reduced, and a large amount of manpower and material resources are saved.
Detailed Description
The technical scheme of the invention is further specifically described by the following embodiments.
Example (b):
the method for realizing rural finance three-resource supervision based on the algorithm link of the data model comprises the following steps:
s1: and collecting and summarizing fund inflow or fund outflow information of each distributed node required by rural finance three-fund supervision, and respectively storing the collected information in each node to form a financial link.
The financial link refers to collection and summarization of discrete data and low-value data of each transaction node. The financial link comprises a country financial platform, the country financial platform comprises a core enterprise, a bank, a fund party and a warranty organization, a link node can be built in each participant, information flow, logistics and fund flow are integrated together, and financial services are engaged on the basis. The bank develops a financial intelligent insurance management business application system based on the support of fund supervision technology, and provides the financial intelligent insurance management business application system for member enterprises of all link finances.
In the embodiment, the financial link relates to the information related to the fund inflow or outflow required by rural financial regulation, and the information comprises a user information dimension, a payment information dimension, a physical network environment information dimension, a data transmission and processing node dimension and the like.
S11: and processing the information data collected and summarized from each node into a numerical type, classifying and importing the numerical type information data into a database according to the characteristic dimension to which the numerical type information data belongs, and collecting and summarizing the data to form a data warehouse.
The real, effective and comprehensive fund inflow or fund outflow information of the country finance is obtained by utilizing the systematic and structural business characteristics of the country finance supervision.
The dimension refers to the number of database tables or the number of screening features, and generally refers to the number of features without special description. In addition to the index, one feature is one-dimensional and the N features are N-dimensional.
In the present embodiment, 3 features or more are defined as the high dimension, and 3 features or less are defined as the low dimension.
The high latitude category refers to a fund transaction mode, for example, the first characteristic is the butt joint of a bank interface and a financial system interface, the second characteristic is the payment of an integrated third-party interface butt joint user, the third characteristic is the payment of a payer, and the fourth characteristic is the WeChat payment.
The low-dimensional category refers to selecting a place where the capital flows in and is characterized by regions; and (4) carrying out statistics on the total income flow of the regional funds, wherein the characteristics are funds.
S12: different node models are combined through analysis of different characteristic dimensions.
Performing dimensionality algorithm analysis, namely performing statistical analysis on main components of high-dimensionality data by selecting proper target dimensionality, and reducing the dimensionality of the data in a weight reduction mode; averaging, weighting, etc. is performed on the low dimensional data to promote dimensionality.
And organizing node models of different topics through analysis of the characteristics. For example, when counting the data pattern with village as dimension, the formula is: village payment node mode data = village payment data (main table) + fund inflow (dimension) + fund inflow direction (dimension) + fund transaction mode (dimension) + docking mode (dimension).
By forming the data node without combining the dimension data and the payment data of the main table, the village and the village can be directly separated, and the village under the unified organization can be counted.
S13: and encrypting the node model data by using an asymmetric encryption algorithm to form a node data block.
Each node is encrypted by a public key and a private key of an asymmetric encryption algorithm, so that the security is high.
S14: steps S11-S13 are repeated for subsequent nodes of data in turn until all node data chunks are formed. And sequentially carrying out acquisition and summarization, dimensional algorithm analysis and standard data model (data structure, data operation and data constraint) abstraction on subsequent nodes of the data, and generating a node data block through an encryption algorithm. .
S15: and initiating data, namely taking the first node data block as the starting point of the link, and sequentially connecting all the node data blocks by using the data structure of the linked list to form a unidirectional data chain so as to form the financial link.
And establishing a multilink model based on one data center according to each node data link, and training the model by using a logistic regression algorithm. In this embodiment, the data involved is encrypted to generate a fund flow node data block linked to the end of the financial link model.
The distributed nodes summarize the end-to-end information data through the deployed information flow to form an access link transparently, and the fund inflow or fund outflow of all rural financial links is uniformly managed through a financial supervision system.
The information of the fund inflow and outflow is uploaded to the distributed nodes, and the financial links are enabled to provide services such as data authentication, source tracing and the like by utilizing the traceable characteristic of the financial links.
S16: and a system for introducing the countryside financial fund supervision service, perfecting a fund transaction system and finishing financial fund supervision.
The trading of digital assets is performed by converting the rural financial link platform into a financial asset exchange by converting the levy, equity, contract, and instrument representing the financing requirements into unique, non-duplicable digital assets.
The finished financial resource supervision system comprises modules of business management, agricultural and economic reports, production and improvement data, bank and agricultural direct connection, early warning supervision, data analysis, loan application, transaction disclosure and the like.
S17: and establishing a monitoring node by utilizing the transparency and traceability of node data, monitoring the data and operation processing, and establishing a wind control analysis model of the big data model.
The wind control analysis model of the big data model comprises a data acquisition and analysis unit, a third-party data access unit, a house credit unit, a vehicle credit unit and a credit card data unit. The data acquisition unit comprises portrait acquisition, voice acquisition, fingerprint acquisition, iris acquisition and basic information input.
A global mutual trust mechanism is exerted through the architecture of the financial link, a strong P2P trust relationship is formed, and a mutual trust mechanism is established. The common trust mechanism uploads contract appointments to form a block chain by introducing an intelligent contract generated by a financial link, so that the block chain is automatically triggered and operated, and introduces technical trust to make up for the possibility of unexpected processes and subjective default in performance, thereby ensuring the effect of designing wind control of a financial scheme and guaranteeing the safety of financing.
S2: and analyzing the dimension algorithm by using the link node data block, training an early warning model by using a logistic regression algorithm, and supervising the node transaction in the financial link.
S21: and collecting data generated in the financial link by means of customer field business entry or web browser customer entry, wherein the data comprises data of customer name, age, birth address, household registration, contact way, bank account and fund transaction.
Since the distance calculation is required, the data type is required to be numerical. The data in this embodiment is in a structured data format.
S22: and analyzing the data through a plurality of screening modes, including age screening, household screening or recent fund transaction amount screening, and analyzing the existence condition of the bidirectional active edge in the fund transaction network by using the screened data.
For example, direct transactions between account pairs reflect the intimacy of the accounts, and frequent fund flow between multiple accounts reflects the fund flow pattern of an organization, so that the existence of bidirectional active edges in a fund transaction network is analyzed.
And S23, training the training algorithm by combining the logistic regression algorithm training algorithm with the Sigmoid function to obtain the optimal classification regression coefficient.
The logistic regression algorithm repeats the following steps R times with regression coefficients initialized to 1:
and calculating the gradient of the whole data set by using the step gradient, updating the vector of the regression coefficient by using a Sigmoid function, and returning the regression coefficient.
The Sigmoid function is:
wherein z is the input of the Sigmoid function;
the vector x is composed of elements x0、x1、x2…xnVector x is the input data of the classifier;
the vector w is composed of elements w0、w1、w2…wnVector (w) is a regression systemAnd (4) counting.
The Sigmoid function is also called a Logistic function, has a value range of (0,1), is used for two classifications, and expresses that the z value is obtained by multiplying corresponding elements of two numerical vectors and then adding all the multiplied elements.
In order to find the optimal parameters, the principle of optimization theory needs to be used. In the present embodiment, a gradient ascent method is used.
S24, testing the algorithm, inputting the testing data to test the effectiveness and the matching degree of the algorithm, and outputting the completed early warning model when the effectiveness and the matching degree are both larger than the preset threshold value; otherwise, the process returns to step S23.
In the present embodiment, the threshold value of the efficiency is 95%, and the threshold value of the goodness of fit is 90%.
S3: and inputting information of different characteristic dimensions required to be inquired by each node into the early warning model, judging whether the financial node has abnormal movement, and performing abnormal movement supervision.
S31: the required query data is input and converted into corresponding structured numerical values.
S32: and carrying out regression calculation on the input numerical value based on the regression coefficient of the trained early warning model, judging that the inquired node has abnormal motion when the output result of the model is greater than a set threshold value, and otherwise, judging that the node has no abnormal motion. The threshold is obtained from historical transaction data, averaged over historical transaction data.
And binding the primary key information of the corresponding village in the data tables corresponding to different dimensions in the engineering, so that each data table is associated with the village. In this embodiment, if the program determines that the specification meets the requirement, a payment process can be initiated, and after the payment process is initiated, the process is recorded in the whole payment process through the theme table, so that the link tracking is facilitated.
The scheme of the embodiment ensures that online operation can be traced through the financial link relationship so as to carry out risk management on the whole transaction process. The data link property extends the traceable depth of financial supervision, and once abnormal data are monitored, data tracing can be carried out.
The financial link can improve the payment, transaction and settlement efficiency; the process that the transaction is confirmed is the process of clearing, collection and auditing, and the financial link uses the data modular link to track, shows all transactions on the platform in real time, and real-time clearing improves efficiency greatly. The financial link can improve the efficiency to the minute level, and the settlement risk is reduced by 99%, so that the capital cost and the systematic risk are effectively reduced.
Financial links can reduce operating costs. The financial link can simplify and automate the lengthy financial service flow between each business system of the countryside and the background work, reduce the interaction between the foreground and the background, and save a large amount of manpower and material resources.
The financial link can effectively prevent failures and attacks. The traditional financial model takes financial institutions such as exchanges or banks as centers, and once the centers break down or are attacked, the whole network can be broken down and the transaction can be suspended. The financial link technology passes through the chain nodes and the data module table, the whole operation cannot be influenced when any part of the chain nodes goes wrong, and each chain node is stored on the corresponding data module table. Therefore, the built-in business continuity of the financial link technology has extremely high reliability and fault tolerance.
The financial link can promote the automation level. Since all documents or assets can be embodied in the form of codes or ledgers, automated transactions are possible over the financial link by setting up the data processing program for the financial link.
The financial link can meet regulatory and auditing requirements. The record stored on the financial link technology has the characteristics of transparency, traceability and uniqueness. Any records, once written to the financial link, are permanently stored. Any transaction between two parties may be tracked and queried.
It should be understood that the examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention may be made by those skilled in the art after reading the teaching of the present invention, and such equivalents may fall within the scope of the present invention as defined in the appended claims.
Claims (10)
1. A method for realizing rural finance three-resource supervision based on an algorithm link of a data model is characterized by comprising the following steps:
s1: collecting and summarizing fund inflow or fund outflow information of each distributed node required by rural finance three-fund supervision, and respectively storing the information in each node to form a financial link;
s2: performing dimension algorithm analysis by using the link node data block, performing a training early warning model by using a logistic regression algorithm, and supervising node transaction in the financial link;
s3: and inputting information of different characteristic dimensions required to be inquired by each node into the early warning model, judging whether the financial node has abnormal movement, and performing abnormal movement supervision.
2. The method for realizing rural financial triage supervision based on the algorithmic link of the data model as claimed in claim 1, wherein the step S1 comprises the steps of:
s11: processing the collected information data from each node into a numerical type, classifying and importing the numerical type information data into a database according to the characteristic dimension of the numerical type information data to form a data warehouse; the feature dimension is the number of database tables or the number of screening features;
s12: different characteristic dimensions are analyzed to combine different node models;
s13: encrypting the node model data by using an asymmetric encryption algorithm to form a node data block;
s14: repeating the steps S11-S13 for the subsequent nodes of the data in sequence until all node data blocks are formed;
s15: and initiating data, namely taking the first node data block as the starting point of the link, and sequentially connecting all the node data blocks by using the data structure of the linked list to form a unidirectional data chain so as to form the financial link.
3. The method for realizing rural financial triage supervision based on the data model algorithm link according to claim 1 or 2, wherein the step S1 further comprises the steps of:
s16: introducing a country financial fund supervision service, perfecting a fund transaction system and establishing a system for perfecting financial fund supervision;
s17: and establishing a monitoring node by utilizing the transparency and traceability of node data, monitoring the data and operation processing, and establishing a wind control analysis model of the big data model.
4. The method for realizing rural finance three-resource supervision based on the algorithm link of the data model according to claim 2, wherein more than three characteristics are defined as high dimensionality, and less than three characteristics are defined as low dimensionality; analyzing different characteristic dimensions comprises analyzing main components of high-dimensional data by selecting proper target dimensions, and reducing the dimensions of the data in a weight reduction mode; and averaging and weighting the low-dimensional data to improve the dimensionality.
5. The method for realizing rural financial three-fund supervision based on the algorithmic link of the data model according to claim 3, wherein the trading of the digital assets is performed by converting the rural financial link platform into a financial asset exchange by changing the land acquisition, the equity, the contract and the bill representing the financing requirement into unique and uncopyable digital assets; the finished financial resource supervision system comprises modules of business management, agricultural and economic reports, production and improvement data, bank and agricultural direct connection, early warning supervision, data analysis, loan application, transaction disclosure and the like.
6. The method for realizing rural financial three-fund supervision based on the data model algorithm link according to claim 3, wherein the wind control analysis model of the big data model comprises a data acquisition and analysis unit, a third-party data access unit, a house loan unit, a car loan unit and a credit card data unit; the data acquisition unit comprises portrait acquisition, voice acquisition, fingerprint acquisition, iris acquisition and basic information input.
7. The method for realizing rural financial triage supervision based on the data model algorithm link according to claim 1 or 2, wherein the step S2 comprises the following steps:
s21: collecting data generated in a financial link in a mode of customer field service input or network browser customer input, wherein the data comprises data of customer names, ages, birth addresses, household registers, contact ways, bank accounts and fund transactions;
s22: analyzing data, and analyzing the existence condition of the bidirectional active edge in the fund transaction network by using the screened data through a plurality of screening modes, including age screening, household registration screening or fund transaction amount screening in recent years;
s23: training algorithm, namely training by combining a logistic regression algorithm training algorithm with a Sigmoid function to obtain an optimal classification regression coefficient;
s24: the testing algorithm is used for inputting testing data to test the effectiveness and the matching degree of the algorithm, and outputting a finished early warning model when the effectiveness and the matching degree are both greater than a preset threshold value; otherwise, the process returns to step S23.
8. The method as claimed in claim 7, wherein the logistic regression algorithm is initialized with regression coefficients to 1 and the following steps are repeated R times: calculating gradient using step gradient of the whole data set, updating a vector of a regression coefficient by using a Sigmoid function, and returning the regression coefficient;
the Sigmoid function is:
z＝w0x0+w1x1+w2x2+...+wnxn
wherein z is the input of the Sigmoid function; vector x is the input data to the classifier and vector w is the regression coefficients.
9. The method for realizing rural financial triage supervision based on the algorithmic link of the data model as claimed in claim 1, wherein the step S3 comprises the steps of:
s31: inputting data to be queried and converting the data into corresponding structured numerical values;
s32: and carrying out regression calculation on the input numerical value based on the regression coefficient of the trained early warning model, judging that the inquired node has abnormal motion when the output result of the model is greater than a set threshold value, and otherwise, judging that the node has no abnormal motion.
10. The method as claimed in claim 9, wherein the threshold is obtained from historical transaction data, and the historical transaction data is averaged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110258472.7A CN113240520A (en) | 2021-03-09 | 2021-03-09 | Method for realizing rural finance three-resource supervision by using algorithm link based on data model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110258472.7A CN113240520A (en) | 2021-03-09 | 2021-03-09 | Method for realizing rural finance three-resource supervision by using algorithm link based on data model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113240520A true CN113240520A (en) | 2021-08-10 |
Family
ID=77130240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110258472.7A Pending CN113240520A (en) | 2021-03-09 | 2021-03-09 | Method for realizing rural finance three-resource supervision by using algorithm link based on data model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113240520A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115062042A (en) * | 2022-06-15 | 2022-09-16 | 五矿国际信托有限公司 | Intelligent query and supervision method for interest of non-standard trust assets principal |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107563936A (en) * | 2017-09-04 | 2018-01-09 | 湖北地信科技集团股份有限公司 | A kind of rural area assets comprehensive management platform |
| CN109377334A (en) * | 2018-09-27 | 2019-02-22 | 大连亿辉科技有限公司 | Rural collective fund management system |
| CN109726848A (en) * | 2018-11-20 | 2019-05-07 | 江苏智途科技股份有限公司 | A kind of wisdom agricultural big data service platform |
| CN109767327A (en) * | 2018-12-20 | 2019-05-17 | 平安科技（深圳）有限公司 | Customer information acquisition and its application method based on anti money washing |
| CN110288223A (en) * | 2019-06-20 | 2019-09-27 | 徐州独角兽信息科技有限公司 | A kind of money of rural collective three supervising platform |
| CN111523988A (en) * | 2020-03-20 | 2020-08-11 | 福建省农村信用社联合社 | Method for realizing safe storage and whole-course trace retention of rural three-information by using block chain |
-
2021
- 2021-03-09 CN CN202110258472.7A patent/CN113240520A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107563936A (en) * | 2017-09-04 | 2018-01-09 | 湖北地信科技集团股份有限公司 | A kind of rural area assets comprehensive management platform |
| CN109377334A (en) * | 2018-09-27 | 2019-02-22 | 大连亿辉科技有限公司 | Rural collective fund management system |
| CN109726848A (en) * | 2018-11-20 | 2019-05-07 | 江苏智途科技股份有限公司 | A kind of wisdom agricultural big data service platform |
| CN109767327A (en) * | 2018-12-20 | 2019-05-17 | 平安科技（深圳）有限公司 | Customer information acquisition and its application method based on anti money washing |
| CN110288223A (en) * | 2019-06-20 | 2019-09-27 | 徐州独角兽信息科技有限公司 | A kind of money of rural collective three supervising platform |
| CN111523988A (en) * | 2020-03-20 | 2020-08-11 | 福建省农村信用社联合社 | Method for realizing safe storage and whole-course trace retention of rural three-information by using block chain |
Non-Patent Citations (2)
| Title |
|---|
| 吕芳等: "金融网络频繁链路发现算法", 《网络与信息安全学报》 * |
| 王煦然等: "土地流转联盟链平台的架构设计", 《信息技术与网络安全》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115062042A (en) * | 2022-06-15 | 2022-09-16 | 五矿国际信托有限公司 | Intelligent query and supervision method for interest of non-standard trust assets principal |
| CN115062042B (en) * | 2022-06-15 | 2023-06-13 | 五矿国际信托有限公司 | Intelligent query and supervision method for principal interest of non-standard trusted asset |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Makarov et al. | Blockchain analysis of the bitcoin market | |
| US11532040B2 (en) | International cash management software using machine learning | |
| Zhang et al. | Credit risk prediction of SMEs in supply chain finance by fusing demographic and behavioral data | |
| Kuhn et al. | Learning from WorldCom: Implications for fraud detection through continuous assurance | |
| Koh et al. | A two-step method to construct credit scoring models with data mining techniques | |
| Deller et al. | Rural broadband speeds and business startup rates | |
| CN105308640A (en) | Methods and systems for automatically generating high quality adverse action notifications | |
| KR20180060044A (en) | Security System for Cloud Computing Service | |
| KR20210127512A (en) | System for small business credit loan based on artificial intelligence, and method thereof | |
| Van Thiel et al. | Artificial intelligence credit risk prediction: An empirical study of analytical artificial intelligence tools for credit risk prediction in a digital era | |
| US20220414762A1 (en) | Method of international cash management using machine learning | |
| CN112817939A (en) | Construction method of data wind control model and data wind control model | |
| CN102496126A (en) | Custody asset transaction data monitoring equipment | |
| US20070239587A1 (en) | System and Method For Dynamically Utilizing and Managing Financial, Operational, and Compliance Data | |
| Wang et al. | A multi-factor two-stage deep integration model for stock price prediction based on intelligent optimization and feature clustering | |
| CN113240520A (en) | Method for realizing rural finance three-resource supervision by using algorithm link based on data model | |
| Jiajia et al. | The effect of the disposal of non-performing loans on interbank liquidity risk in China: A cash flow network-based analysis | |
| Zhao et al. | Corporate social network, ownership structure, and corporate financing constraints: Evidence from China | |
| Wang et al. | Stock return prediction based on Bagging-decision tree | |
| CN111339102B (en) | Financial record information accurate and trusted storage method based on block chain | |
| CN109949145A (en) | A kind of daily account book algorithm based on retrial book keeping operation method | |
| dei Belliera et al. | Flood risk insurance: the Blockchain approach to a Bayesian adaptive design of the contract. | |
| Stezhkin | Management of banking network stability taking into account industry-specific risks | |
| Zhao | Financial Risk Evaluation of Digital Currency Based on CART Algorithm Blockchain | |
| Samakovitis et al. | Computer-aided financial fraud detection: promise and applicability in monitoring financial transaction fraud |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220324 Address after: 310051 room 2349, building 7, No. 1197, Bin'an Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou Zhengyun Data Technology Co.,Ltd. Applicant after: Hangzhou Xinzhongda Technology Co., Ltd Address before: 310000 room 905, building 1, No. 88, Xiangyuan Road, Gongshu District, Hangzhou City, Zhejiang Province Applicant before: Hangzhou Zhengyun Data Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right |