CN116485512A - Bank data analysis method and system based on reinforcement learning - Google Patents
Bank data analysis method and system based on reinforcement learning Download PDFInfo
- Publication number
- CN116485512A CN116485512A CN202310404487.9A CN202310404487A CN116485512A CN 116485512 A CN116485512 A CN 116485512A CN 202310404487 A CN202310404487 A CN 202310404487A CN 116485512 A CN116485512 A CN 116485512A
- Authority
- CN
- China
- Prior art keywords
- information
- analysis
- analysis object
- reinforcement learning
- credit
- 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.)
- Withdrawn
Links
- 230000002787 reinforcement Effects 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000007405 data analysis Methods 0.000 title claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims abstract description 286
- 238000012360 testing method Methods 0.000 claims abstract description 138
- 238000005065 mining Methods 0.000 claims abstract description 41
- 238000007418 data mining Methods 0.000 claims abstract description 15
- 230000006870 function Effects 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 230000003213 activating effect Effects 0.000 claims description 2
- 238000004891 communication Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 229910052744 lithium Inorganic materials 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000002893 slag Substances 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010304 firing Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
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
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention relates to the field of data analysis, and discloses a bank data analysis method based on reinforcement learning, which comprises the following steps: acquiring bank data, identifying an analysis object in the bank data, and calculating credit rating of the analysis object; extracting key information in an analysis object, and performing data mining on the analysis object by utilizing an expansion network layer in the reinforcement learning model to obtain mining information; combining the credit rating and mining information to construct an analysis strategy of an analysis object; according to an analysis strategy, testing an analysis object by using a testing layer in the reinforcement learning model to obtain testing feedback information; calculating a test feedback information score by using a scoring layer in the reinforcement learning model; when the information score is smaller than a preset threshold value, returning to a data decision step of constructing bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information; and when the information score is not smaller than a preset threshold value, generating an analysis report of the analysis object. The invention can improve the practicability of bank data analysis.
Description
Technical Field
The invention relates to the field of data analysis, in particular to a bank data analysis method and device based on reinforcement learning.
Background
In the past decades, the Chinese banking industry has rapidly developed, the electronic informatization of banks has been popularized, a large amount of basic data is accumulated, a large amount of newly-added data is generated by banks every day, the client information data also grows in an explosive trend, and the competitors of commercial banks have been expanded to industries with the advantage of all-round information of clients, such as securities, insurance, funds, internet finance companies and the like. Meanwhile, as the opening degree of the financial market is increasingly improved, competition among large banks in China becomes extremely intense, so that the analysis of data becomes very important.
At present, the analysis of the bank data is mainly based on the analysis of a large amount of data, and the comparison analysis is performed by acquiring a large amount of data to obtain a result, however, the analysis of the large data cannot test certain types of data in the bank data, so that the data decision in the analysis of the bank data is often not high in practicality.
Disclosure of Invention
In order to solve the technical problems, the invention provides a bank data analysis method based on reinforcement learning, which can improve the practicability of bank data analysis.
In a first aspect, the present invention provides a reinforcement learning-based bank data analysis method, including:
acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, and calculating the credit rating of the analysis object;
extracting key information in the analysis object, and performing data mining on the analysis object by utilizing an expansion network layer in a reinforcement learning model according to the key information to obtain mining information;
constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
according to the analysis strategy, testing the analysis object by using a testing layer in the reinforcement learning model to obtain testing feedback information;
calculating information scores of the test feedback information by using a scoring layer in the reinforcement learning model;
returning to the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold;
and generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
In a possible implementation manner of the first aspect, the acquiring bank data to be analyzed, and identifying an analysis object in the bank data to be analyzed includes:
classifying the bank data to be analyzed to obtain classified bank data, and extracting classification fields of each classification in the classified bank data;
and inquiring the field attribute of the classification field, and identifying the analysis object in the bank data to be analyzed according to the field attribute.
In a possible implementation manner of the first aspect, the calculating the credit rating of the analysis object includes:
inquiring a historical credit record of the object to be analyzed, and determining the current credit state of the object to be analyzed according to the historical credit record;
identifying credit characteristics of the analysis object according to the current credit state, and calculating the credit rating of the analysis object according to the credit characteristics by combining the following formula:
wherein P represents a credit rating, E represents a credit feature weight matrix of the analysis object, n represents the number of levels of the credit rating, a m Represents the mth credit state, m, of credit object a i The ith feature, c, represented in the mth credit state h Represents the h record in the credit record c, a b The b-th credit representing credit object aState, b j The jth feature, c, represented in the b-th credit state k Representing the kth record in credit record c.
In a possible implementation manner of the first aspect, the extracting key information in the analysis object includes:
acquiring the total information of the analysis object, and constructing an information matrix of the total information;
converting the full-quantity information into digital information by using the information matrix, and calculating the information weight of the digital information by using the following formula:
where q represents the information weight, log (σ) -1 Represents a matrix function, m represents the whole amount information of the analysis object, k a A-th information in the key information;
and extracting digital information corresponding to the information weight greater than a preset threshold value to obtain key information.
In a possible implementation manner of the first aspect, according to the key information, performing data mining on the analysis object by using an extended network layer in a reinforcement learning model to obtain mining information, including:
inquiring information characteristics of the key information, and identifying information rules of the key information according to the information characteristics;
and searching the associated information of the key information according to the information rule, and screening the related information to obtain mining information.
In a possible implementation manner of the first aspect, the constructing, by using a decision layer network in the reinforcement learning model, an analysis policy of the analysis object in combination with the credit rating and the mining information includes:
determining the current credit state of the analysis object according to the credit rating;
and normalizing the mining information to obtain normalized information, and constructing an analysis strategy of the analysis object by combining the credit state and the normalized information.
In a possible implementation manner of the first aspect, according to the analysis strategy, the testing the analysis object by using a testing layer in the reinforcement learning model to obtain test feedback information includes:
constructing a test environment of the analysis object according to the analysis strategy, and arranging an analysis framework of the analysis object in the test environment;
configuring a test device of the analysis object in the analysis framework, and generating a test code of the analysis object in the test;
and running the test code in the test device to obtain test feedback information.
In a possible implementation manner of the first aspect, the calculating, by using a scoring layer in the reinforcement learning model, an information score of the test feedback information includes:
Inquiring the test state of the analysis object corresponding to the test feedback information and the feedback value of the test feedback information;
according to the test state and the feedback value, calculating an information score of the test feedback information by using the following formula:
wherein F represents information scoring, a represents the test state of the analysis object, b represents the feedback value of the test feedback information, rank represents the scoring function, s i The ith information representing the test feedback information.
In a second aspect, the present invention provides a reinforcement learning-based banking data analysis system, the apparatus comprising:
the credit rating calculation module is used for acquiring the bank data to be analyzed, identifying the analysis object in the bank data to be analyzed and calculating the credit rating of the analysis object;
the information mining module is used for extracting key information in the analysis object, and carrying out data mining on the analysis object by utilizing an expansion network layer in the reinforcement learning model according to the key information to obtain mining information;
an analysis strategy construction module for constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
The analysis object testing module is used for testing the analysis object by utilizing a testing layer in the reinforcement learning model according to the analysis strategy to obtain testing feedback information;
the information score calculating module is used for calculating the information score of the test feedback information by utilizing a score layer in the reinforcement learning model;
the step of returning the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold value;
and the analysis report generation module is used for generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
Compared with the prior art, the technical principle and beneficial effect of this scheme lie in:
according to the scheme, firstly, the bank data to be analyzed is obtained, the analysis objects in the bank data to be analyzed are identified, target data can be screened out from a large amount of data, the data processing amount is reduced, the data analysis efficiency is improved, the credit rating of the analysis objects is calculated, the analysis objects can be initially classified, and further different strategies are implemented according to different classification results; secondly, key information in the analysis object can be extracted to quickly find key information characteristics of the analysis object and certain rules existing in the information, and according to the key information, an expansion network layer in a reinforcement learning model is utilized to conduct data mining on the analysis object, so that the information surface of the analysis object can be expanded by mining information, and further the analysis result of the analysis object is more reasonable and the reliability is higher; constructing analysis strategies of the analysis objects by utilizing a decision layer network in the reinforcement learning model by combining the credit rating and the mining information, making corresponding data analysis strategies for different analysis objects, testing the analysis objects by utilizing a test layer in the reinforcement learning model according to the analysis strategies, obtaining test feedback information, and knowing the results of different tests of the analysis objects by different analysis strategies, thereby obtaining the analysis strategies suitable for the analysis objects according to the test results; further, according to the embodiment of the invention, the grading layer in the reinforcement learning model is utilized to calculate the information grading of the test feedback information, so that the grading of the test feedback information can be known, the adaptability of the analysis strategy to the object to be analyzed is judged according to the grading of the test feedback information, and when the information grading is not smaller than the preset threshold value, the analysis report of the analysis object is generated, so that the comprehensive information of the analysis object can be known in detail, and reliable basis can be provided for subsequent strategy analysis. Therefore, the bank data analysis method based on reinforcement learning provided by the embodiment of the invention can improve the practicability of bank data analysis.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a schematic flow chart of a reinforcement learning-based bank data analysis method according to an embodiment of the present invention;
FIG. 2 is a schematic block diagram of a reinforcement learning-based bank data analysis system according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an internal structure of an electronic device for implementing a reinforcement learning-based bank data analysis method according to an embodiment of the present invention.
Detailed Description
It should be understood that the detailed description is presented by way of example only and is not intended to limit the invention.
The embodiment of the invention provides a reinforcement learning-based bank data analysis method, wherein an execution subject of the reinforcement learning-based bank data analysis method comprises, but is not limited to, at least one of a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the invention. In other words, the reinforcement learning-based bank data analysis method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a bank data analysis method based on reinforcement learning according to an embodiment of the invention is shown. The reinforcement learning-based bank data analysis method depicted in fig. 1 includes the following steps S1 to S7:
s1, acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, and calculating credit rating of the analysis object.
According to the embodiment of the invention, the bank data to be analyzed is obtained, the analysis object in the bank data to be analyzed is identified, the target data can be screened out from a large amount of data, the data processing amount is reduced, and the data analysis efficiency is improved. Wherein, the analysis object refers to some kind of data in the bank data, such as user data, user credit data, bank historical activity data and the like.
As one embodiment of the present invention, the acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, includes: classifying the bank data to be analyzed to obtain classified bank data, extracting classified fields of all classifications in the classified bank data, inquiring field attributes of the classified fields, and identifying analysis objects in the bank data to be analyzed according to the field attributes. The field is a record of each line of data and contains all information of each line of data, and the field attribute is that the data table head has a representative function and has specificity and uniqueness.
Optionally, the classified bank data is obtained by classifying the bank data to be analyzed through a classification function, the classification function is generated by java language, classified fields of each classification in the classified bank data are extracted through a c-language generated field extraction script, field attributes of the classified fields are queried through reading source codes generated by the fields, and analysis objects in the bank data to be analyzed are identified through reading field tags of the fields.
Furthermore, the embodiment of the invention can carry out preliminary classification on the analysis object by calculating the credit rating of the analysis object, and further implement different strategies according to different classification results.
The credit rating refers to comprehensive credit rating according to bank historical data and other channel data.
As one embodiment of the present invention, the calculating the credit rating of the analysis object includes: inquiring a historical credit record of the object to be analyzed, determining the current credit state of the object to be analyzed according to the historical credit record, identifying the credit characteristics of the object to be analyzed according to the current credit state, and calculating the credit rating of the object to be analyzed according to the credit characteristics in combination with the following formula:
Wherein P represents a credit rating, E represents a credit feature weight matrix of the analysis object, n represents the number of levels of the credit rating, a m Represents the mth credit state, m, of credit object a i The ith feature, c, represented in the mth credit state h Represents the h record in the credit record c, a b B represents the b-th credit status of credit object a, b j The jth feature, c, represented in the b-th credit state k Representing the kth record in credit record c.
S2, extracting key information in the analysis object, and performing data mining on the analysis object by using an expansion network layer in the reinforcement learning model according to the key information to obtain mining information.
The embodiment of the invention can extract the key information in the analysis object through the key information, wherein the key information is such as identity information, transaction records, default records and the like.
As one embodiment of the present invention, the extracting key information in the analysis object includes: acquiring the total information of the analysis object, constructing an information matrix of the total information, converting the total information into digital information by using the information matrix, and calculating the information weight of the digital information by using the following formula:
Where q represents the information weight, log (σ) -1 Represents a matrix function, m represents the whole amount information of the analysis object, k a Represents the a-th information in the key information,
and extracting digital information corresponding to the information weight greater than a preset threshold value to obtain key information. Wherein the information matrix refers to a set, and the digital information refers to an information form represented by numbers.
Optionally, the whole information of the analysis object is obtained by reading a database corresponding to the analysis object, an information matrix of the whole information is constructed by a matrix constructor, the whole information is converted into digital information by using the information matrix, the digital information is coded and converted by a coding tool in the information matrix, the information weights of different information in the digital information are calculated by an aph method, and the digital information with the information weight greater than a preset threshold value is extracted by an information extraction script.
Furthermore, according to the embodiment of the invention, the data mining is carried out on the analysis object by utilizing the expansion network layer in the reinforcement learning model according to the key information, so that the information surface of the analysis object can be expanded by the mining information, and further, the analysis result of the analysis object is more reasonable and the reliability is higher.
According to the key information, the data mining of the analysis object is performed by using an extended network layer in a reinforcement learning model to obtain mining information, which comprises the following steps: inquiring information characteristics of the key information, identifying information rules of the key information according to the information characteristics, searching associated information of the key information according to the information rules, and screening the related information to obtain mining information. The information features refer to features of information such as identity information, historical consumption records, consumption preference and the like, and the information rules refer to specific rules existing in certain types of information such as weekly consumption in specific markets, deposit and withdrawal habits and the like.
Optionally, the information features of the key information are extracted through a feature extraction function, the feature extraction function is generated by Java, the information rule of the key information is identified by reading the information features of the key information, the associated information of the key information is retrieved through a big data technology, the related information is screened, the obtained mining information is realized through an information screening script, and the information screening script is generated by python language.
S3, combining the credit rating and the mining information, and constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model.
According to the embodiment of the invention, through combining the credit rating and the mining information, the analysis strategy of the analysis object is constructed by utilizing the decision layer network in the reinforcement learning model, so that corresponding data analysis strategies can be formulated for different analysis objects.
As one embodiment of the present invention, the constructing an analysis policy of the analysis object using a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information includes: and determining the current credit state of the analysis object according to the credit rating, normalizing the mining information to obtain normalized information, and constructing an analysis strategy of the analysis object by combining the credit state and the normalized information. The credit status refers to the status of the analysis object credit information, such as normal, attention, secondary, etc.
Optionally, the current credit state of the analysis object is determined by identifying credit information corresponding to the credit level, normalizing the mined information to obtain normalized information, and implementing the normalized information by a normalization function, wherein the normalization function is generated by python language, and an analysis strategy of the analysis object is constructed by a strategy analysis script.
And S4, testing the analysis object by utilizing a testing layer in the reinforcement learning model according to the analysis strategy to obtain test feedback information.
According to the embodiment of the invention, the analysis object is tested by utilizing the test layer in the reinforcement learning model according to the analysis strategy, so that the test feedback information is obtained, the results of different tests of different analysis strategies on the analysis object can be known, and further, the analysis strategy suitable for the analysis object is obtained according to the test results.
As an embodiment of the present invention, according to the analysis strategy, the testing the analysis object by using the testing layer in the reinforcement learning model to obtain the test feedback information includes: and constructing a test environment of the analysis object according to the analysis strategy, arranging an analysis framework of the analysis object in the test environment, configuring a test device of the analysis object in the analysis framework, generating a test code of the analysis object in the test, and running the test code in the test device to obtain test feedback information.
The test environment refers to hardware, software, network conditions and the like necessary for data testing, the analysis framework refers to overall layout or hierarchy of data analysis such as data input, data testing, test result return and the like, and the test device refers to a tool for data testing such as a test script generated through java language.
Optionally, the test environment of the analysis object is built through an lnmt tool, the analysis frame of the analysis object is arranged through a test frame generated by a pre-built test scheme structure liux tool, the test code of the analysis object is generated through v+, and the test code is operated through a code operation script pre-configured by the test device.
And S5, calculating information scores of the test feedback information by using a scoring layer in the reinforcement learning model.
According to the embodiment of the invention, the grading of the test feedback information can be known by calculating the information grading of the test feedback information by utilizing the grading layer in the reinforcement learning model, and the adaptability of the analysis strategy to the object to be analyzed is further judged according to the grading of the test feedback information.
As one embodiment of the present invention, the calculating the information score of the test feedback information using the scoring layer in the reinforcement learning model includes: inquiring the test state of the test feedback information corresponding to the analysis object and the feedback value of the test feedback information, and calculating the information score of the test feedback information according to the test state and the feedback value by using the following formula:
Wherein F represents information scoring, a represents the test state of the analysis object, b represents the feedback value of the test feedback information, rank represents the scoring function, s i The ith information representing the test feedback information.
And S6, returning to the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold.
According to the embodiment of the invention, the data decision step of constructing the bank data to be analyzed by combining the credit rating and the expansion information and utilizing the decision layer network in the reinforcement learning model can be carried out after the information score is smaller than the preset threshold value, so that the data decision can be continuously adjusted until the data decision that the test feedback information score is not smaller than the preset threshold value appears.
And S7, when the information score is not smaller than a preset threshold value, generating an analysis report of the analysis object.
According to the embodiment of the invention, when the information score is not smaller than the preset threshold value, the analysis report of the analysis object is generated, so that the comprehensive information of the analysis object can be known in detail, and a reliable basis can be provided for subsequent strategy analysis.
As one embodiment of the present invention, the generating the analysis report of the analysis object when the information score is not less than a preset threshold includes: and constructing an analysis report text of the analysis object, inserting a text control component into the report text, loading analysis information of the analysis object in the text control component, and activating the text control component to obtain an analysis report of the analysis object. Wherein the report text refers to an interface or window for loading data information in a computer, and the control component refers to a tool for packaging data.
Optionally, the analysis report text is constructed through an Excel text tool, the text control component is inserted through a component generation script of the sql language, analysis information of the analysis object is loaded through a data transmission interface corresponding to the analysis object, the text control component is activated through a component activation Han Shu, and the component activation function is generated by java.
According to the scheme, firstly, the bank data to be analyzed is obtained, the analysis objects in the bank data to be analyzed are identified, the target data can be screened out from a large amount of data, the data processing amount is reduced, the data analysis efficiency is improved, the credit rating of the analysis objects is calculated, the analysis objects can be initially classified, and further different strategies are implemented according to different classification results; secondly, key information in the analysis object can be extracted to quickly find key information characteristics of the analysis object and certain rules existing in the information, and according to the key information, an expansion network layer in a reinforcement learning model is utilized to conduct data mining on the analysis object, so that the information surface of the analysis object can be expanded by mining information, and further the analysis result of the analysis object is more reasonable and the reliability is higher; constructing analysis strategies of the analysis objects by utilizing a decision layer network in the reinforcement learning model by combining the credit rating and the mining information, making corresponding data analysis strategies for different analysis objects, testing the analysis objects by utilizing a test layer in the reinforcement learning model according to the analysis strategies, obtaining test feedback information, and knowing the results of different tests of the analysis objects by different analysis strategies, thereby obtaining the analysis strategies suitable for the analysis objects according to the test results; further, according to the embodiment of the invention, the grading layer in the reinforcement learning model is utilized to calculate the information grading of the test feedback information, so that the grading of the test feedback information can be known, the adaptability of the analysis strategy to the object to be analyzed is judged according to the grading of the test feedback information, and when the information grading is not smaller than the preset threshold value, the analysis report of the analysis object is generated, so that the comprehensive information of the analysis object can be known in detail, and reliable basis can be provided for subsequent strategy analysis. Therefore, the bank data analysis method based on reinforcement learning provided by the embodiment of the invention can improve the practicability of bank data analysis.
As shown in fig. 2, a functional block diagram of the reinforcement learning-based bank data analysis system of the present invention is shown.
The reinforcement learning-based bank data analysis system 200 of the present invention may be installed in an electronic device. Depending on the functions implemented, the reinforcement learning based banking data analysis system may include a credit rating calculation module 201, an information mining module 202, an analysis policy construction module 203, an analysis object test module 204, an information scoring calculation module 205, a step return module 206, and an analysis report generation module 207.
The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the embodiment of the present invention, the functions of each module/unit are as follows:
the credit rating calculation module 201 is configured to obtain bank data to be analyzed, identify an analysis object in the bank data to be analyzed, and calculate a credit rating of the analysis object;
the information mining module 202 is configured to extract key information in the analysis object, and perform data mining on the analysis object by using an extended network layer in a reinforcement learning model according to the key information to obtain mining information;
The analysis policy construction module 203 is configured to construct an analysis policy of the analysis object by using a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
the analysis object testing module 204 is configured to test the analysis object by using a testing layer in the reinforcement learning model according to the analysis strategy, so as to obtain test feedback information;
the information score calculating module 205 is configured to calculate an information score of the test feedback information by using a score layer in the reinforcement learning model;
the step return module 206 is configured to return to the data decision step of constructing the bank data to be analyzed by using the decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold;
the analysis report generating module 207 is configured to generate an analysis report of the analysis object when the information score is not less than a preset threshold.
In detail, the modules in the reinforcement learning-based bank data analysis system 200 in the embodiment of the present invention use the same technical means as the reinforcement learning-based bank data analysis method described in fig. 1 and can produce the same technical effects, and are not described herein.
Fig. 3 is a schematic structural diagram of an electronic device for implementing a reinforcement learning-based bank data analysis method according to the present invention.
The electronic device may include a processor 30, a memory 31, a communication bus 32, and a communication interface 33, and may also include a computer program, such as a fired lithium slag forging program, stored in the memory 31 and executable on the processor 30.
The processor 30 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing Unit, CPU), a microprocessor, a digital processing chip, a graphics processor, a combination of various control chips, and so on. The processor 30 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules (e.g., executing a firing lithium slag forging program, etc.) stored in the memory 31, and calling data stored in the memory 31.
The memory 31 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 31 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 31 may in other embodiments also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device. Further, the memory 31 may also include both an internal storage unit and an external storage device of the electronic device. The memory 31 may be used not only for storing application software installed in an electronic device and various data such as codes of a firing lithium slag forging program, etc., but also for temporarily storing data that has been output or is to be output.
The communication bus 32 may be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus, or an extended industry standard architecture (Extended Industry Standard Architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 31 and at least one processor 30 or the like.
The communication interface 33 is used for communication between the electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. Among other things, the display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 3 shows only an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 3 is not limiting of the electronic device and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.
For example, although not shown in the drawings, the electronic device may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 30 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited in scope by this configuration.
The roasted lithium slag forging program stored in the memory 31 of the electronic device is a combination of a plurality of computer programs, which when run in the processor 30, can implement the following method:
acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, and calculating the credit rating of the analysis object;
extracting key information in the analysis object, and performing data mining on the analysis object by utilizing an expansion network layer in a reinforcement learning model according to the key information to obtain mining information;
Constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
according to the analysis strategy, testing the analysis object by using a testing layer in the reinforcement learning model to obtain testing feedback information;
calculating information scores of the test feedback information by using a scoring layer in the reinforcement learning model;
returning to the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold;
and generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
In particular, the specific implementation method of the processor 30 on the computer program may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.
Further, the electronic device integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a non-volatile computer readable storage medium. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement the method of:
acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, and calculating the credit rating of the analysis object;
extracting key information in the analysis object, and performing data mining on the analysis object by utilizing an expansion network layer in a reinforcement learning model according to the key information to obtain mining information;
constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
according to the analysis strategy, testing the analysis object by using a testing layer in the reinforcement learning model to obtain testing feedback information;
calculating information scores of the test feedback information by using a scoring layer in the reinforcement learning model;
returning to the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold;
And generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
It should be noted that in this document, relational terms such as "first" and "second" and the like are 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. Moreover, 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. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A reinforcement learning-based bank data analysis method, the method comprising:
acquiring bank data to be analyzed, identifying an analysis object in the bank data to be analyzed, and calculating the credit rating of the analysis object;
extracting key information in the analysis object, and performing data mining on the analysis object by utilizing an expansion network layer in a reinforcement learning model according to the key information to obtain mining information;
constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
According to the analysis strategy, testing the analysis object by using a testing layer in the reinforcement learning model to obtain testing feedback information;
calculating information scores of the test feedback information by using a scoring layer in the reinforcement learning model;
returning to the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold;
and generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
2. The method of claim 1, wherein the acquiring the bank data to be analyzed, identifying the analysis object in the bank data to be analyzed, comprises:
classifying the bank data to be analyzed to obtain classified bank data, and extracting classification fields of each classification in the classified bank data;
and inquiring the field attribute of the classification field, and identifying the analysis object in the bank data to be analyzed according to the field attribute.
3. The method of claim 1, wherein said calculating a credit rating for said analysis object comprises:
Inquiring a historical credit record of the object to be analyzed, and determining the current credit state of the object to be analyzed according to the historical credit record;
identifying credit characteristics of the analysis object according to the current credit state, and calculating the credit rating of the analysis object according to the credit characteristics by combining the following formula:
wherein P represents a credit rating, E represents a credit feature weight matrix of the analysis object, n represents the number of levels of the credit rating, a m Represents the mth credit state, m, of credit object a i The ith feature, c, represented in the mth credit state h Represents the h record in the credit record c, a b B represents the b-th credit status of credit object a, b j The jth feature, c, represented in the b-th credit state k Representing the kth record in credit record c.
4. The method of claim 1, wherein the extracting key information in the analysis object comprises:
acquiring the total information of the analysis object, and constructing an information matrix of the total information;
converting the full-quantity information into digital information by using the information matrix, and calculating the information weight of the digital information by using the following formula:
Wherein q represents information weight, log sigma -1 Represents a matrix function, m represents the whole amount information of the analysis object, k a A-th information in the key information;
and extracting digital information corresponding to the information weight greater than a preset threshold value to obtain key information.
5. The method of claim 1, wherein the data mining the analysis object according to the key information by using an extended network layer in a reinforcement learning model to obtain mining information comprises:
inquiring information characteristics of the key information, and identifying information rules of the key information according to the information characteristics;
and searching the associated information of the key information according to the information rule, and screening the related information to obtain mining information.
6. The method of claim 1, wherein the constructing an analysis strategy for the analysis object using a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information comprises:
determining the current credit state of the analysis object according to the credit rating;
and normalizing the mining information to obtain normalized information, and constructing an analysis strategy of the analysis object by combining the credit state and the normalized information.
7. The method of claim 1, wherein the testing the analysis object with the test layer in the reinforcement learning model according to the analysis strategy to obtain test feedback information comprises:
constructing a test environment of the analysis object according to the analysis strategy, and arranging an analysis framework of the analysis object in the test environment;
configuring a test device of the analysis object in the analysis framework, and generating a test code of the analysis object in the test;
and running the test code in the test device to obtain test feedback information.
8. The method of claim 1, wherein calculating an information score for the test feedback information using a scoring layer in the reinforcement learning model comprises:
inquiring the test state of the analysis object corresponding to the test feedback information and the feedback value of the test feedback information;
according to the test state and the feedback value, calculating an information score of the test feedback information by using the following formula:
wherein F represents information scoring, a represents the test state of the analysis object, b represents the feedback value of the test feedback information, rank represents the scoring function, s i The ith information representing the test feedback information.
9. The method according to any one of claims 1 to 8, wherein generating an analysis report of the analysis object when the information score is not less than a preset threshold value comprises:
constructing an analysis report text of the analysis object, and inserting a text control component into the report text;
and loading analysis information of the analysis object in the text control component, and activating the text control component to obtain an analysis report of the analysis object.
10. A reinforcement learning based banking data analysis system, the apparatus comprising:
the credit rating calculation module is used for acquiring the bank data to be analyzed, identifying the analysis object in the bank data to be analyzed and calculating the credit rating of the analysis object;
the information mining module is used for extracting key information in the analysis object, and carrying out data mining on the analysis object by utilizing an expansion network layer in the reinforcement learning model according to the key information to obtain mining information;
an analysis strategy construction module for constructing an analysis strategy of the analysis object by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the mining information;
The analysis object testing module is used for testing the analysis object by utilizing a testing layer in the reinforcement learning model according to the analysis strategy to obtain testing feedback information;
the information score calculating module is used for calculating the information score of the test feedback information by utilizing a score layer in the reinforcement learning model;
the step of returning the data decision step of constructing the bank data to be analyzed by utilizing a decision layer network in the reinforcement learning model in combination with the credit rating and the expansion information when the information score is smaller than a preset threshold value;
and the analysis report generation module is used for generating an analysis report of the analysis object when the information score is not smaller than a preset threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310404487.9A CN116485512A (en) | 2023-04-17 | 2023-04-17 | Bank data analysis method and system based on reinforcement learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310404487.9A CN116485512A (en) | 2023-04-17 | 2023-04-17 | Bank data analysis method and system based on reinforcement learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116485512A true CN116485512A (en) | 2023-07-25 |
Family
ID=87213145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310404487.9A Withdrawn CN116485512A (en) | 2023-04-17 | 2023-04-17 | Bank data analysis method and system based on reinforcement learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116485512A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117332992A (en) * | 2023-11-24 | 2024-01-02 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
-
2023
- 2023-04-17 CN CN202310404487.9A patent/CN116485512A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117332992A (en) * | 2023-11-24 | 2024-01-02 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
| CN117332992B (en) * | 2023-11-24 | 2024-02-09 | 北京国联视讯信息技术股份有限公司 | Collaborative manufacturing method and system for industrial Internet |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112507936B (en) | Image information auditing method and device, electronic equipment and readable storage medium | |
| CN111523677B (en) | Method and device for realizing interpretation of prediction result of machine learning model | |
| CN113095076A (en) | Sensitive word recognition method and device, electronic equipment and storage medium | |
| CN113177700B (en) | Risk assessment method, system, electronic equipment and storage medium | |
| CN109299235A (en) | Knowledge base searching method, apparatus and computer readable storage medium | |
| CN113886708A (en) | Product recommendation method, device, equipment and storage medium based on user information | |
| CN113592605A (en) | Product recommendation method, device, equipment and storage medium based on similar products | |
| CN116485512A (en) | Bank data analysis method and system based on reinforcement learning | |
| CN112328657A (en) | Feature derivation method, feature derivation device, computer equipment and medium | |
| CN108664501B (en) | Advertisement auditing method and device and server | |
| CN114969484A (en) | Service data searching method, device, equipment and storage medium | |
| CN113362162A (en) | Wind control identification method and device based on network behavior data, electronic equipment and medium | |
| CN112579781A (en) | Text classification method and device, electronic equipment and medium | |
| CN113705201B (en) | Text-based event probability prediction evaluation algorithm, electronic device and storage medium | |
| CN116578696A (en) | Text abstract generation method, device, equipment and storage medium | |
| CN114676430A (en) | Malicious software identification method, device, equipment and computer readable storage medium | |
| CN115186188A (en) | Product recommendation method, device and equipment based on behavior analysis and storage medium | |
| CN113343102A (en) | Data recommendation method and device based on feature screening, electronic equipment and medium | |
| CN114219367A (en) | User scoring method, device, equipment and storage medium | |
| KR102101456B1 (en) | Method for reducing false positives for diagnosis of personal information exposure of text files and irregular image files | |
| CN113505117A (en) | Data quality evaluation method, device, equipment and medium based on data indexes | |
| CN113468421A (en) | Product recommendation method, device, equipment and medium based on vector matching technology | |
| CN114462411B (en) | Named entity recognition method, device, equipment and storage medium | |
| CN112949752B (en) | Training method and device of business prediction system | |
| CN116188049B (en) | Potential user mining method and device based on chain analysis |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230725 |