CN115600226B - Warehouse bill mortgage data processing method and device - Google Patents
Warehouse bill mortgage data processing method and device Download PDFInfo
- Publication number
- CN115600226B CN115600226B CN202211255061.3A CN202211255061A CN115600226B CN 115600226 B CN115600226 B CN 115600226B CN 202211255061 A CN202211255061 A CN 202211255061A CN 115600226 B CN115600226 B CN 115600226B
- Authority
- CN
- China
- Prior art keywords
- ciphertext
- mortgage
- matrix
- data
- relation
- 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.)
- Active
Links
- 238000003672 processing method Methods 0.000 title description 6
- 238000000034 method Methods 0.000 claims abstract description 123
- 238000011156 evaluation Methods 0.000 claims abstract description 64
- 230000008569 process Effects 0.000 claims abstract description 61
- 238000012545 processing Methods 0.000 claims abstract description 47
- 238000013210 evaluation model Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 67
- 239000013598 vector Substances 0.000 claims description 60
- 238000013528 artificial neural network Methods 0.000 claims description 52
- 238000012549 training Methods 0.000 claims description 27
- 238000013507 mapping Methods 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 22
- 238000006243 chemical reaction Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000011161 development Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 238000011160 research Methods 0.000 claims description 5
- 230000004931 aggregating effect Effects 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 230000007306 turnover Effects 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 238000013075 data extraction Methods 0.000 claims description 3
- 230000008520 organization Effects 0.000 claims description 3
- 238000010998 test method Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 claims description 2
- 230000008859 change Effects 0.000 claims description 2
- 238000009795 derivation Methods 0.000 claims 1
- 238000010801 machine learning Methods 0.000 abstract description 5
- 238000012502 risk assessment Methods 0.000 abstract description 5
- 238000007726 management method Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000013500 data storage Methods 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 4
- 230000000644 propagated effect Effects 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 241000579895 Chlorostilbon Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013529 biological neural network Methods 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 210000003169 central nervous system Anatomy 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000010976 emerald Substances 0.000 description 1
- 229910052876 emerald Inorganic materials 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- ZLIBICFPKPWGIZ-UHFFFAOYSA-N pyrimethanil Chemical compound CC1=CC(C)=NC(NC=2C=CC=CC=2)=N1 ZLIBICFPKPWGIZ-UHFFFAOYSA-N 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000010979 ruby Substances 0.000 description 1
- 229910001750 ruby Inorganic materials 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Finance (AREA)
- Computer Security & Cryptography (AREA)
- Game Theory and Decision Science (AREA)
- Computer Hardware Design (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Educational Administration (AREA)
- Bioethics (AREA)
- Accounting & Taxation (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of financial data processing, in particular to a method and a device for processing warehouse bill mortgage data; receiving an electronic warehouse bill mortgage application request; determining a target electronic warehouse list to be mortgage according to the first label of the mortgage application request; evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result; generating a signed mortgage protocol file corresponding to the target electronic warehouse bill, and sending the signed mortgage protocol file to an account of a corresponding user through encryption; according to the technical scheme, risk assessment of the financial information related to the consignment mortgage is achieved based on the knowledge graph and the machine learning, and the assessed financial data is encrypted, decrypted, sent and stored in a homomorphic encryption mode based on the business process, so that the risk of information leakage caused by external attack is reduced, and the informatization level in the consignment mortgage business process is improved as a whole.
Description
Technical Field
The application relates to the technical field of financial data processing, in particular to a method and a device for processing warehouse sheet mortgage data.
Background
The warehouse entry is one type mortgage established with warehouse entry as a target. As a novel service item, the system provides a wide stage for expanding the service item and developing various operations for the warehouse enterprises, and is particularly widely applied to the transformation process from the traditional warehouse enterprises to the modern logistics enterprises.
The prior art has at least the following problems:
In the existing warehouse bill mortgage method, risks generated in the mortgage process cannot be identified easily and cannot be evaluated, fund risks exist in the mortgage process, and business development of warehouse bill mortgages is not facilitated. In addition, because of the related financial sensitive information such as the mortgage, the preservation and transmission process secret mode of the information is more traditional in the prior art, and the information is more easily attacked by other people, so that serious financial risks and losses are caused by the tampering of the information.
Disclosure of Invention
In order to solve the technical problems, the application provides a processing method of the warehouse note data, which reduces the problems of inaccurate risk identification caused by inaccurate evaluation and lower information leakage and falsification caused by lower encryption mode level in information transmission and storage by evaluating characteristic information and associated information in the warehouse note data.
In order to achieve the above purpose, the technical scheme adopted by the embodiment of the application is as follows:
In a first aspect, a method for processing a warehouse entry mortgage data is applied to a server, and includes: receiving an electronic warehouse bill mortgage application request; determining a target electronic warehouse list to be mortgage according to the first label of the mortgage application request; extracting characteristic information in the target electronic warehouse bill, and evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result; sending approval mortgage result information based on the evaluation result, generating a signed mortgage protocol file corresponding to the target electronic warehouse bill, and sending the signed mortgage protocol file to an account of a corresponding user through encryption processing; extracting characteristic information in the target electronic warehouse bill, and evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result, wherein the method comprises the following steps: processing the characteristic information to be processed based on a preset knowledge graph template to obtain a target knowledge graph containing the information to be processed; learning and representing the target knowledge graph to obtain initial data; and inputting the data to be processed into the trained BP neural network to obtain expected output.
In a first implementation manner of the first aspect, learning and representing the target knowledge-graph includes: extracting entities, relations and relation planes of a triplet structure in the target knowledge graph, carrying out vector initialization, constructing a negative triplet based on the positive triplet, training the positive triplet and the negative triplet, updating the entities, the relations and the relation plane vectors according to the decreasing direction of the loss function by adopting a random gradient decreasing optimization method, and stopping iteration until the loss function tends to be stable, so as to obtain vector representations of the optimized entities, relations and relation planes.
With reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, the BP neural network includes an input layer, a hidden layer and an output layer, where the number of the input layers is the same as the number of the evaluation indexes; and the evaluation index is correspondingly set with the entity characteristics of the target electronic warehouse.
In a third implementation manner of the first aspect, the sending the signed mortgage protocol file to the account of the corresponding user through encryption processing includes: generating a random key, generating corresponding encryption noise for the random key, and acquiring a sequential ciphertext; sequentially adjusting the first sub-ciphertext in the sequential ciphertext based on the mapping relation and generating a second ciphertext; encrypting the second ciphertext based on the mapping relation to obtain a second sub ciphertext; and encrypting the second ciphertext and the second sub-ciphertext as final ciphertexts.
With reference to the third possible implementation manner of the first aspect, in a fourth possible implementation manner, the method further includes performing decryption processing based on the second sub-ciphertext and the second sub-ciphertext includes: decrypting the second sub-ciphertext to obtain a first mapping relation; processing the second ciphertext based on the mapping relation to obtain the sequential ciphertext; summing the sequential ciphertext based on the random key to obtain a set of information about the random key; deriving plain text based on the set.
With reference to the third possible implementation manner of the first aspect, in a fifth possible implementation manner, the sequential ciphertext includes a plurality of sub-ciphertexts, sequentially adjusting a first sub-ciphertext in the sequential ciphertext based on a mapping relationship and generating a second sub-ciphertext includes: randomly encrypting the multiple sub-ciphertexts in the sequence cipher text based on a mapping relation to obtain multiple first sub-ciphertexts; the plurality of first sub-ciphertexts are placed at the tail part of the sequence ciphertext arrangement sequence, and a plurality of second ciphertexts are finally obtained
With reference to the third possible implementation manner of the first aspect, in a sixth possible implementation manner, performing encryption processing on the second ciphertext based on a mapping relationship to obtain a second sub-ciphertext includes: and encrypting the second ciphertext based on the AES encryption algorithm and the mapping relation to obtain a second sub ciphertext.
With reference to the third possible implementation manner of the first aspect, in a seventh possible implementation manner, before generating the random key, the method further includes dividing the mortgage protocol file to obtain a plurality of data blocks, and performing independent encryption processing on the plurality of data blocks, which specifically includes: comparing the mortgage protocol file with a preset data threshold value to obtain a comparison result, and dividing the mortgage protocol file into a plurality of data blocks with approximate sizes based on the comparison result; distributing the data blocks to a plurality of corresponding encryption units, and encrypting the data blocks based on the encryption units to obtain a plurality of encrypted data; and aggregating the plurality of encrypted data to obtain spliced target encrypted data.
With reference to the seventh possible implementation manner of the first aspect, in an eighth possible implementation manner, aggregating a plurality of encrypted data to obtain spliced encrypted data includes: acquiring text offset of encrypted data of a plurality of data blocks, and sorting based on the offset to obtain the encrypted data of the plurality of sorted data blocks; and splicing the encrypted data of the data blocks based on the sequence to form the target encrypted data.
In a second aspect, a warehouse entry mortgage data processing apparatus includes: the request receiving module is used for receiving an electronic warehouse bill mortgage application request; the target electronic warehouse list determining module is used for determining a target electronic warehouse list to be mortgage according to the first label of the mortgage application request; the evaluation module is used for evaluating the target electronic warehouse bill; and the encryption module is used for carrying out encryption processing on the target electronic bill in the transmission process.
In a third aspect, a terminal device is provided comprising a memory, a processor and a computer program stored in the memory and executable on the processor.
In a fourth aspect, a computer-readable storage medium storing a computer program is provided.
According to the technical scheme provided by the embodiment of the application, the risk assessment of the financial information related to the consignment mortgage is realized based on the knowledge graph and the machine learning, and the information is encrypted, decrypted, sent and stored based on the business process by homomorphic encryption, so that the risk of information leakage caused by external attack is reduced, and the informatization level in the consignment mortgage business process is integrally improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
The methods, systems, and/or programs in the accompanying drawings will be described further in terms of exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein the exemplary numbers represent like mechanisms throughout the various views of the drawings.
Fig. 1 is a schematic structural diagram of a terminal device according to an embodiment of the present application.
FIG. 2 is a flow chart of a data processing method according to some embodiments of the application.
Fig. 3 is a schematic block diagram of an apparatus according to an embodiment of the present application.
Detailed Description
In order to better understand the above technical solutions, the following detailed description of the technical solutions of the present application is made by using the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limiting the technical solutions of the present application, and the technical features of the embodiments and the embodiments of the present application may be combined with each other without conflict.
In the following detailed description, numerous specific details are set forth by way of examples in order to provide a thorough understanding of the relevant teachings. It will be apparent, however, to one skilled in the art that the application can be practiced without these details. In other instances, well known methods, procedures, systems, components, and/or circuits have been described at a relatively high-level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.
The present application uses a flowchart to illustrate the execution of a system according to an embodiment of the present application. It should be clearly understood that the execution of the flowcharts may be performed out of order. Rather, these implementations may be performed in reverse order or concurrently. Additionally, at least one other execution may be added to the flowchart. One or more of the executions may be deleted from the flowchart.
Before describing embodiments of the present invention in further detail, the terms and terminology involved in the embodiments of the present invention will be described, and the terms and terminology involved in the embodiments of the present invention will be used in the following explanation.
(1) In response to a condition or state that is used to represent the condition or state upon which the performed operation depends, the performed operation or operations may be in real-time or with a set delay when the condition or state upon which it depends is satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(2) Based on the conditions or states that are used to represent the operations that are being performed, one or more of the operations that are being performed may be in real-time or with a set delay when the conditions or states that are being relied upon are satisfied; without being specifically described, there is no limitation in the execution sequence of the plurality of operations performed.
(3) BP neural network, BP neural network is an artificial neural network (ARTIFICIAL NEURAL NETWORK, ANN), abbreviated as neural network or neural-like network, in the field of machine learning and cognitive science, is a mathematical model or computational model that mimics the structure and function of biological neural network (the central nervous system of animals, particularly the brain) for estimating or approximating functions.
(4) The knowledge graph structurally describes concepts and interrelationships thereof in a physical world in a symbol form, and the basic knowledge composition form is a triplet of < entities, relations and entities >, and the entities are mutually connected through the relations to form a complex net-shaped knowledge structure.
The technical scheme provided by the embodiment of the application has the main application scene that the consignment mortgage becomes an important means for melting funds between enterprises and banks abroad and is also an important component of value-added service of the warehousing industry in the business process of the consignment mortgage financial products. In China, the warehouse list mortgage is taken as an emerging service project, has less experience in reality, and meanwhile, because the warehouse list mortgage business involves a series of problems such as law, management system, information security and the like, various risks and disputes can be generated. At present, two main factors restricting the development of the warehouse sheet mortgage are information security and risk assessment, and the two practical problems are mainly caused by the fact that the assessment means and the potential risk pre-assessment difficulty are large and the information security management of the financial information is imperfect. And aiming at the realization of risk assessment, the evaluation of the display target object and the associated information is realized by mainly utilizing the evaluation indexes and through the multiple evaluation dimensions of the multiple evaluation indexes, so that the pre-judgment of the risk is realized. However, in the current use process of quantitative evaluation, the problems of excessively complicated evaluation indexes, excessively biased metering indexes, simple processing of metering indexes, difficult data acquisition, insufficient interpretation of quantitative evaluation and the like are increasingly prominent.
The present method is aimed at the selection of the evaluation method, and the H index, the RBF (Radial Basis Function ) of the main component, the gray system theory, the BP neural network, the gray correlation analysis method, the balance score card and the optimal weight set are applied, and good effects are obtained. In addition, the data envelope analysis (DataEnvelopmentAnalysis, DEA) method is widely adopted in non-profit sector evaluation, and the research method constructs an evaluation model from the perspective of 'input-output'. However, a single research method may encounter some limitations in the application process, and has the problems of low integrity and low accuracy of the evaluation result.
And, because the financial data is sensitive data, and more business sensitive information is involved in the business process, the security setting requirement for the financial data on the information in the transmission process and the storage process is higher. The conventional encryption method for financial data is mostly based on the conventional encryption method, the encryption stability is low, the encryption effect of the ciphertext can be reduced by obtaining the encryption logic under multiple statistics, so that a large amount of newer encryption is required for the ciphertext, the encryption mode reduces the data processing efficiency, improves the data cost, and is not beneficial to corresponding encryption processing under mature business logic.
Therefore, based on the above technical background, a method needs to be provided for two main links in the process of the warehouse sheet mortgage service to process data and information, so as to improve the identification of risks and the capture of risk information in the warehouse sheet mortgage service, and encrypt data containing sensitive information in the business process.
Based on the above technical background, the embodiment of the present application provides a terminal device 100, where the terminal device includes a memory 110, a processor 120, and a computer program stored in the memory and capable of running on the processor, where the processor executes a method for processing the job ticket mortgage data, evaluates the job ticket mortgage information, and encrypts the data in the process and the data to be stored after completion. In this embodiment, the terminal device communicates with the user end, and issues the obtained final data file to the corresponding user end, so as to send the mortgage protocol file on hardware. The method is based on network implementation aiming at the information sending mode, and an association relation between the user terminal and the terminal equipment is required to be established before the terminal equipment is applied, and the association between the terminal equipment and the user terminal can be realized through a registration mode. The terminal device can be aimed at a plurality of user terminals or one user terminal.
In this embodiment, the terminal may be a server, and includes a memory, a processor, and a communication unit for the physical structure of the server. The memory, the processor and the communication unit are electrically connected with each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory is used for storing specific information and programs, and the communication unit is used for sending the processed information to the corresponding user side.
In this embodiment, the storage module is divided into two storage areas, where one storage area is a program storage unit and the other storage area is a data storage unit. The program storage unit is equivalent to a firmware area, the read-write authority of the area is set to be in a read-only mode, and the data stored in the area can not be erased and changed. And the data in the data storage unit can be erased or read and written, and when the capacity of the data storage area is full, the newly written data can cover the earliest historical data.
The Memory may be, but is not limited to, random access Memory (RandomAccess Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-OnlyMemory, EPROM), electrically erasable Read Only Memory (Ele ultrasound ric Erasable Programmable Read-Only Memory, EEPROM), etc.
The processor may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also Digital Signal Processors (DSPs)), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The embodiment provides a warehouse bill mortgage data processing method, which comprises the following steps:
And S210, receiving an electronic warehouse sheet mortgage application request.
In this embodiment, the request is issued based on the client, and when the request is issued, the data related to the request and the corresponding file are issued together.
And S220, determining a target electronic warehouse to be mortgage according to the first label of the mortgage application request.
In this embodiment, the first tag is a registration code of the user end, where one user end corresponds to one user, and unique codes are configured for independent users, and identity identification and corresponding storage space determination are performed through the unique codes, that is, each independent user end corresponds to an independent storage space.
And S230, extracting characteristic information in the target electronic warehouse bill, and evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result.
In this embodiment, the obtaining of the evaluation result is mainly implemented through knowledge graph construction and corresponding machine learning, and specifically includes the following steps:
s231, processing the feature information to be processed based on a preset knowledge-graph template to obtain a target knowledge-graph containing the information to be processed.
In this embodiment, the preset knowledge-graph template is constructed by a mode of early-stage model construction, and in this embodiment, an evaluation system is set in the knowledge-graph template, and a specific scoring weight is set in the scoring system in the knowledge-graph template.
In this embodiment, the knowledge graph may be represented by g= (E, R), where E and R respectively represent multiple types of entities and a relationship set in the knowledge graph, the knowledge graph is composed of a number of RDF triples, and the representation is represented by t= < Vh, R, vt >, where Vh represents a head entity, vt represents a tail entity, and R represents a relationship between the head entity and the tail entity.
The construction of the knowledge graph comprises the steps of knowledge extraction, knowledge fusion, knowledge processing and knowledge reasoning of source data.
With respect to the interrelationship between the individual elements in the evaluation system in the present embodiment, a hierarchical structure is provided. According to the research requirement, a three-layer index structure is adopted, namely a total target layer, a sub-target layer and an index layer. The specific index structure of the third layer is provided with an evaluation index aiming at the application scene of the current model, specifically: operating profit margins, total asset return rates, flow rates, snap rates, asset loading rates, total asset turnover rates, net profit growth rates, leading credit information, mortgage rendering forces, mortgage price stability, mortgage legitimacy, mortgage loss resistance, insurance amounts, insurance years, commodity price fluctuations, exchange rate fluctuations, related economic policies, regional development, professional numbers, information management systems, organizational structures, warehouse entry and exit keeping systems, cargo assessment systems. And according to the characteristics of scientific research activities, the selected index elements are subjected to homogeneous classification to form 10 secondary indexes, wherein the secondary indexes are respectively: mortgage infrastructure, applicant team members, credit levels, product conditions, economic policy situation, international monetary policies, insurance capabilities, business conditions, management levels, loss bearing capabilities. Secondly, classifying and profiling 10 secondary indexes according to scientific metering principles and evaluation indexes to form 4 primary indexes, wherein the primary indexes are respectively as follows: support index, input index, output index, benefit index.
For the application scenario of this embodiment, the knowledge graph is an evaluation knowledge graph, and the structure of the knowledge graph provided for this embodiment is: the entity type set is X= { mortgage basic condition, applicant team member, credit level, product condition, insurance capability, operation condition and management level }, then attribute characteristics of each entity are constructed, the mortgage basic condition= { mortgage change force, mortgage price stability, mortgage legality and mortgage loss resistance capability }, the applicant team member= { professional technician number and organization structure }, credit level= { leading credit information }, product condition= { commodity price fluctuation }, insurance capability= { insurance amount, insurance period, insurance company }, operation condition= { operation profit rate, total asset return rate, flow rate, quick action rate, asset load rate, total asset turnover rate and net profit growth rate }, and management level = { information management system and in-out warehouse keeping system }.
The corresponding target knowledge graph is obtained through the knowledge graph template provided by the embodiment and is used for subsequent data processing and specific evaluation results.
And S232, learning and representing the target knowledge graph to obtain initial data.
In this embodiment, the structure of the knowledge graph obtained in step S231 is a triplet structure, wherein the entities and relationships of the red for the knowledge graph are represented by discrete symbols, and the data processing for the BP neural network used subsequently is based on continuous values, and cannot be directly applied to the knowledge graph, so that the data structure of the knowledge graph needs to be processed to obtain a data structure that can be used for processing the BP neural network. At present, common one-hot coding (one-hot Representation) can well represent different entities as mutually different and mutually independent vectors, but completely ignores semantic relations between the entities. In practical application, a plurality of entities in the knowledge graph have rich semantic relations, and the single-hot coding expression mode can not meet the requirement of expressing the semantic relations among the entities. In order to better represent semantic information of knowledge, a representation learning technique is used herein to project entities and relationships in a knowledge graph to a low-dimensional vector space, and each entity and relationship is learned to obtain a knowledge representation of a continuous vector. Based on the obtained knowledge representation of the continuous vector, the semantic relation between the entities can be measured by utilizing numerical operation, so that new knowledge can be effectively inferred, and the new knowledge can be input into the BP neural network as priori knowledge to be used for restricting and supervising the training process of the model.
In this embodiment, for the learning representation model TransH, representation learning can be performed on entities and relationships based on a triplet structure, which defines a real value vector for each entity and each relationship, and defines a hyperplane for each relationship, and represents the real value vector with a normal vector wr, and the specific process includes the following steps:
For a triplet (h, r, t), transH first projects the head entity vector h and the tail entity vector t along the normal wr into the hyperplane corresponding to the relation r, denoted by hr, tr:
In this embodiment, hr+r≡tr is satisfied for the triplet (h, r, t), otherwise tr should be distant from hr+r, where the scoring function is expressed as:
dist(h,r,t)＝||hr+r-tr||。
therefore TransH defines a loss function as:
In this embodiment, h 'and t' represent the head and tail entity vectors in the negative triplet; s represents a positive case triplet set; s' represents a negative case triplet set; gamma is a constant representing the spacing between positive and negative samples; [ x ] + represents max (0, x).
In this embodiment, the entities and relationships in the knowledge graph are representation-learned by TransH to obtain their vector representations. As the vector is trained once to contain rich semantic information and is used as the input of the deep learning model, the dependence of the BP neural network on a large amount of annotation data can be reduced, the learning process of the deep learning model can be guided and restrained, the classification effect is improved, and good interpretability is shown.
And S234, inputting the data to be processed into the trained BP neural network to obtain expected output.
In this embodiment, the BP neural network is a neural network obtained by training, and training for the BP neural network includes the following processes:
based on the constructed knowledge-graph template in step S231, and an evaluation system corresponding to the knowledge-graph template is constructed.
In this embodiment, the structure of the knowledge graph template is not described, but the construction of the evaluation system is described, and the rating system in this embodiment is a mathematical model, and the specific construction method includes: an evaluation system is constructed based on the first-level, second-level and third-level evaluation indexes, and specifically comprises the following steps:
Establishing an importance influence relation among layers, namely setting m factors to have relative importance to a certain factor of the previous layer, and comparing and judging an ith factor (i=1, 2,3, …, m) with a jth factor (j=1, 2,3, … m) according to a specific scale rule, wherein the relative importance degree is aij. The m-order matrix thus constructed can be used to solve for the priority of each factor with respect to certain criteria, called the judgment matrix, noted as
A＝(aij)m*m。
The key point of constructing the judgment matrix is to design a specific scale method for comparing and judging the relative importance degree of two factors in pairs, so that the relative importance degree of any two factors has a certain quantity standard, in the embodiment, the judgment matrix is established by adopting a 1-9 scale method for processing and a 1-9 scale method: a= (a ij)m*m.
Specifically, j and m in the scale a jm represent the j-th and m-th factors, and if a jm =3, it indicates that the j-th factor is slightly more important than the m-th factor with respect to the criterion Cr, and the remaining scales are similar.
According to the above processing, the expert interview method and the analytic hierarchy process are comprehensively applied, and a judgment matrix is respectively required to be constructed for the primary index and the 10 secondary indexes, specifically as follows:
wherein the matrix for the 10 secondary indexes is respectively:
The mortgage basic condition matrix is as follows: the applicant team member matrix is: The credit level matrix is: the product condition matrix is: The economic policy situational matrix is: the operation condition matrix is as follows: the international currency policy matrix is: The management level matrix is:
the loss-bearing capacity matrix is:
The insurance capability matrix is:
Based on the above matrix, the maximum feature root λmax of each judgment matrix and the corresponding feature vector W thereof, that is AiW =λ maxW, are obtained, and the specific process of the process comprises:
The components (W1, W2, …, wm) of the feature vectors W, W corresponding to the respective judgment matrices are calculated as the relative importance levels, i.e., weight coefficients, corresponding to the m elements.
Among them, there are two methods of sum-product and root method for calculating the weight coefficient, and the sum-product method is adopted in the present embodiment, so only the sum-product method will be described:
Normalizing each column of the judgment matrix: For the judgment matrix normalized by columns, and then by rows Summing, obtaining normalized vector of the sum result,Then w= [ W1, W2, W3, … Wm ] is the required.
According to the calculation formula of the maximum characteristic root: Where (AWi) represents the ith component of vector AW.
According to the formula, the weight of each corresponding secondary index and the maximum characteristic value of the corresponding index can be obtained, and the characteristic vector W and the single-order matrix P of each index can be obtained.
The eigenvector matrix is as follows:
W＝(0.0231,0.0165,0.0331,0.1521,0.0616,0.2090,0.0816,0.1109,0.2680,0.0440)T。
The single rank matrix for the secondary hierarchy is as follows:
The mortgage basic condition matrix is as follows: p 2＝(0.8000,0.2000)T;
The applicant team member matrix is: p 3＝(0.8000,0.2000)T;
The credit level matrix is: p 4＝(0.2000,0.8000)T;
the product condition matrix is: p 5＝(0.1782,0.7514,0.0704)T;
the economic policy situational matrix is: p 6＝(0.7500,0.2500)T;
The operation condition matrix is as follows: p 7＝(0.4402,0.0920,0.1470,0.0368,0.2350,0.0491)T;
The international currency policy matrix is: p 8＝(0.2000,0.8000)T;
the management level matrix is: p 9＝(0.1667,0.8333)T;
the loss-bearing capacity matrix is: p 10＝(0.1667,0.8333)T;
The insurance capability matrix is: p 11＝(0.1667,0.8333)T.
The corresponding BP neural network is constructed based on the evaluation system, in the embodiment, the structure aiming at the BP neural network comprises an input layer, a hidden layer and an output layer, wherein the layer number structure of the output layer is 1, and the setting aiming at the hidden layer is obtained by any one of the following formulas:
where k is the number of hidden layer nodes, n is the number of output layer nodes, l is the number of output layer nodes, and a is any constant from 1 to 10.
And the acquisition of the number of nodes of the hidden layer is determined based on the number of three-level evaluation indexes in the evaluation system. In the present embodiment, the number of the index for the secondary rank is 10, and the number of the nodes for the output layer is 1 according to the structure of the BP neural network, so the number of the nodes for the hidden layer can be obtained as 8 for the above formula.
The transfer function of the hidden layer setting in the BP neural network in this embodiment is a Sigmoid function.
The basic model structure of the BP neural network provided in this embodiment is provided above, and the optimal target BP neural network is obtained by training the basic model structure, and specific training process is as follows:
And inputting the training sample, namely the acquired knowledge-graph template, into the BP neural network for training, and correcting the trained BP neural network through the knowledge-graph template acquired by testing the sample to obtain a final evaluation model.
In this embodiment, the training process for the BP neural network includes two parts, the first part is a conversion of a data structure of the knowledge-graph template, and the second part is a specific process for training the BP neural network based on data after the conversion of the data structure.
Wherein the conversion of the data structure for the knowledge-graph template comprises: extracting entities, relations and relation planes of a triplet structure in the knowledge graph, carrying out vector initialization, constructing a negative triplet based on the positive triplet, training the positive triplet and the negative triplet, updating the entities, the relations and the relation plane vectors according to the decreasing direction of the loss function by adopting a random gradient decreasing optimization method, and stopping iteration until the loss function tends to be stable, so as to obtain vector representations of the entities, the relations and the relation planes after optimization. In this embodiment, the negative triplet is obtained by randomly replacing the head entity or the tail entity in the positive triplet.
The data extraction and labeling are required for the acquired data of the structural conversion, and specifically:
After data conversion, each entity in the knowledge graph has an entity vector corresponding to the entity vector, and each relation has a relation vector and a hyperplane corresponding to the relation vector. The present embodiment uses a following formula to extract the projection vector set Er of all entity vectors in the "scoring" relational hyperplane. Wherein the formula is as follows: Wherein, eri E Er represents a projection vector of an entity vector in a hyperplane corresponding to a scoring relation; ei represents the entity vector; wr represents the normal vector of the hyperplane to which the "scoring" relationship corresponds.
In this embodiment, the evaluation index in the evaluation system is determined, the entity is labeled, and the labeled data is input as training to the corresponding input layer.
The specific training for the converted structure and the labeled data belongs to the training method for the BP neural network, which may be the existing training process, and will not be described in detail in this embodiment.
And corresponding tests are required to be carried out aiming at the BP neural network after training, the test method in this embodiment includes: and obtaining target expected output based on an evaluation system, obtaining real-time expected output based on the BP neural network, comparing the relation between the target expected output and the BP neural network, and determining whether the trained BP neural network is completed based on a set threshold value.
The method comprises the following steps: obtaining expected output based on an evaluation system, wherein the expected output is a target expected output of the BP neural network; based on the test sample, the test sample is used as an input to the iterated BP neural network, and real-time expected output is obtained; comparing the real-time desired output with the target desired output; and adjusting the BP neural network by using a loss function based on the comparison result.
In this embodiment, the process of obtaining the desired output for the evaluation system is described in step S231, and the corresponding output result is obtained by constructing a matrix of the two-stage classification, that is, the maximum feature and feature vector of the judgment matrix.
Based on the training sample as input to the iterated BP neural network, obtaining real-time expected output, comparing the real-time expected output with the target expected output, and adjusting the loss function based on a comparison result.
And processing the BP neural network obtained through training for the data to be processed with the labeling label to obtain expected output, wherein the expected output is a specific score.
And S240, sending approval mortgage result information based on the evaluation result, generating a signed mortgage protocol file corresponding to the target electronic warehouse bill, and sending the signed mortgage protocol file to an account of a corresponding user through encryption processing.
Based on the step S230, a specific evaluation result is obtained after the evaluation, wherein the evaluation result includes agreement or disagreement, and the business processing flow after agreement is a mortgage protocol file made for the corresponding evaluation result and the application filed by the applicant, wherein the mortgage protocol file relates to information of the corresponding applicant and specific financial data, and encryption processing is required for the data in the data transmission process and the data storage process, so that the risk of information leakage caused by external attack is reduced.
In this embodiment, the encryption method adopted is a homomorphic encryption method, and in the prior art, the homomorphic encryption method supports management of ciphertext data under privacy protection, so that various operations such as retrieval, calculation, statistics and the like can be directly performed on the ciphertext at a far end (if the built environment is a cloud server, the cloud server is directly cloud end), and a result is returned to a user in a ciphertext form. Compared with the traditional encryption algorithm, the encryption and decryption operations between the cloud end and the user are not needed frequently, and the expenditure of communication and calculation resources is reduced. The user privacy data is stored on the cloud server in a ciphertext mode, so that the user privacy data is prevented from being stolen and tampered by a third party, and the security guarantee is provided for the user to use the cloud server platform.
And the method for encrypting the data comprises the following processes:
S241, generating a random secret key, generating corresponding encryption noise for the random secret key, and acquiring a sequential ciphertext.
In this embodiment, the generation of the key is a plurality of processes, the initial key generated first is a random key, and in the subsequent process, the random key is subjected to multi-process conversion and deformation of the key structure. The generation process for the random secret key is as follows:
Let K i∈R,si e R be the basic structure of the key, the symmetric key be K (n) = [ (K 1,s1),…,(kn,sn) ], where n > 3, and the basic structure for the symmetric key meets the following requirements:
wherein the symmetric key is an initial basic key structure, and the processing for the structure key comprises:
The n-1 pair of encryption noise p= [ (r 1,p1),…,(rn-1,pn-1) ] corresponding to the initial key is configured, and in this embodiment, the security of the key is improved by adding encryption noise, where the nature of encryption noise is a floating point number set, that is, the security of the key is improved by adding floating point numbers.
The method is characterized in that the method is used for determining the sequential ciphertext based on encryption noise and a symmetric key, and comprises the following specific calculation processes of:
Wherein m is the plaintext, namely the plaintext of the data of the file to be encrypted.
And S242, sequentially adjusting the first sub-ciphertext in the sequential ciphertext based on the mapping relation and generating a second ciphertext.
In this embodiment, the sequential ciphertext generated in step S241 is secondarily encrypted based on the mapping encryption manner to form a corresponding first sub-ciphertext, and the sequence of the first sub-ciphertext is adjusted to form a second ciphertext, where the purpose of the sequence adjustment is to scramble the sequence of each key in the ciphertext, so that the external decryption cost of the key is increased, and the security of the ciphertext is improved. The specific process is that the ith sub-ciphertext i c in the sequential ciphertext is mapped into the jth sub-ciphertext of the disordered ciphertext according to a function, and is denoted as j dc, so d j =i.
Wherein, the mapping function in this embodiment is: n (i) =j, 1.ltoreq.i, j.ltoreq.n.
Where for i to be a natural number set, specifically i e [1,2, …, n ], the set of mapped results is defined as j, so ci and jdc satisfy:
wherein, C ' is sequential ciphertext, C ' is disordered ciphertext, wherein disordered ciphertext C ' is: c "= (C d1,…cdn), in this embodiment, the out-of-order ciphertext is the second ciphertext.
And S243, carrying out encryption processing on the second ciphertext based on the mapping relation to obtain a second sub ciphertext.
In this embodiment, the mapping relationship in this step is specifically that the second ciphertext, that is, the disordered ciphertext, is encrypted based on the AES encryption algorithm and the mapping relationship in the above step to obtain a second sub-ciphertext, where the encryption algorithm in this embodiment is the AES encryption algorithm, the mapping relationship is a mapping function in the above step, and the corresponding second sub-ciphertext is obtained after the processing in this process and is c n+1. The processing is to encrypt the array J by using the AES encryption algorithm, and generate the ciphertext c n+1 as a part of the ciphertext. The AES encryption algorithm belongs to an existing encryption algorithm, and is not described in detail in this embodiment.
And S244, encrypting the second ciphertext and the second sub-ciphertext serving as final ciphertext.
In this embodiment, the second ciphertext is an out-of-order ciphertext, that is, the second sub-ciphertext is used as encrypted data, and encrypting the second ciphertext corresponds to encrypting the plaintext m, so as to obtain a final ciphertext as follows: c= [ C d1,…,cdn,cn+1 ].
In this embodiment, steps S241 to 244 are encryption processes, and a decryption process is configured for the client, and the decryption process is described in detail below, which includes:
S245, decrypting the sub-ciphertext C n+1 based on an AES algorithm to obtain a mapping array J, and restoring the sub-ciphertext i c corresponding to the sequence ciphertext C', thereby obtaining the corresponding relation between i c and i k and between i s.
Step S246, determining a secret key set.
Wherein, the secret key set is:
And S247, calculating a plaintext m.
Wherein the expression of plaintext m is:
The deriving process for plaintext m may be based on the expression of sequential ciphertext, which may be the following process:
The process of encrypting and decrypting the data is performed with respect to step S240, and the encrypted data is formed. And if the data volume of the to-be-encrypted mortgage protocol file is large, the encryption cost is high only through real-time encryption processing, so that a corresponding encryption processing method is required to be provided for the encryption operation process of a large data volume.
S250, dividing the mortgage protocol file to obtain a plurality of data blocks, and independently encrypting the plurality of data blocks.
S251, comparing the mortgage protocol file with a preset data threshold value to obtain a comparison result, and dividing the mortgage protocol file into a plurality of data blocks with approximate sizes based on the comparison result.
And S252, distributing the data blocks to a plurality of corresponding encryption units, and encrypting the data blocks based on the encryption units to obtain a plurality of encrypted data.
And S253, aggregating the plurality of encrypted data to obtain spliced target encrypted data.
In this embodiment, this process includes the steps of:
and acquiring text offset of the encrypted data of the plurality of data blocks, and sorting based on the offset to obtain the encrypted data of the plurality of sorted data blocks.
And splicing the encrypted data of the data blocks based on the sequence to form the target encrypted data.
Referring to fig. 3, the present embodiment further provides a warehouse sheet mortgage data processing apparatus 300, including: the request receiving module 310 is configured to receive an electronic warehouse ticket mortgage application request. And the target electronic bill determining module 320 is configured to determine a target electronic bill to be mortgage according to the first tag of the mortgage application request. And the evaluation module 330 is used for evaluating the target electronic warehouse bill. And the encryption module 340 is configured to encrypt the target electronic bill in the transmission process.
According to the technical scheme provided by the embodiment, the risk assessment of the financial information related to the consignment mortgage is realized based on the knowledge graph and the machine learning, and the information is encrypted, decrypted, sent and stored in a homomorphic encryption mode based on the business process, so that the risk of information leakage caused by external attack is reduced, and the informatization level in the consignment mortgage business process is integrally improved.
It is to be understood that the terminology which is not explained by terms of nouns in the foregoing description is not intended to be limiting, as those skilled in the art can make any arbitrary deduction from the foregoing disclosure.
The person skilled in the art can undoubtedly determine technical features/terms of some preset, reference, predetermined, set and preference labels, such as threshold values, threshold value intervals, threshold value ranges, etc., from the above disclosure. For some technical feature terms which are not explained, a person skilled in the art can reasonably and unambiguously derive based on the logical relation of the context, so that the technical scheme can be clearly and completely implemented. The prefixes of technical feature terms, such as "first", "second", "example", "target", etc., which are not explained, can be unambiguously deduced and determined from the context. Suffixes of technical feature terms, such as "set", "list", etc., which are not explained, can also be deduced and determined unambiguously from the context.
The foregoing disclosure of embodiments of the present application will be apparent to and complete in light of the foregoing disclosure to those skilled in the art. It should be appreciated that the development and analysis of technical terms not explained based on the above disclosure by those skilled in the art is based on the description of the present application, and thus the above is not an inventive judgment of the overall scheme.
While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements and adaptations of the application may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within the present disclosure, and therefore, such modifications, improvements, and adaptations are intended to be within the spirit and scope of the exemplary embodiments of the present disclosure.
Meanwhile, the present application uses specific terms to describe embodiments of the present application. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the application. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various portions of this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of at least one embodiment of the present application may be combined as suitable.
In addition, those skilled in the art will appreciate that the various aspects of the application are illustrated and described in the context of a number of patentable categories or conditions, including any novel and useful processes, machines, products, or materials, or any novel and useful improvements thereof. Accordingly, aspects of the application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "unit," component, "or" system. Furthermore, aspects of the application may be embodied as a computer product in at least one computer-readable medium, the product comprising computer-readable program code.
The computer readable signal medium may comprise a propagated data signal with computer program code embodied therein, for example, on a baseband or as part of a carrier wave. The propagated signal may take on a variety of forms, including electro-magnetic, optical, etc., or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer readable signal medium may be propagated through any suitable medium including radio, electrical, fiber optic, RF, or the like, or any combination of the foregoing.
Computer program code required for carrying out aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming such as Java, scala, smalltalk, eiffel, JADE, emerald, C ++, c#, vb net, python, etc., or similar conventional programming languages such as the "C" programming language, visual Basic, fortran2003, perl, COBOL 2002, php, abap, dynamic programming languages such as Python, ruby and Groovy or other programming languages. The programming code may execute entirely on the user's computer, or as a stand-alone software package, or partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service, such as software as a service (SaaS).
Furthermore, the order in which the processing elements and sequences are described, the use of numerical letters, or other designations are used is not intended to limit the order in which the processes and methods of the application are performed unless specifically recited in the claims. While in the foregoing disclosure there has been discussed, by way of various examples, some embodiments of the application which are presently considered to be useful, it is to be understood that this detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments of the application. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.
It should also be appreciated that in the foregoing description of at least one embodiment of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of at least one embodiment of the application. This method of disclosure does not imply that the subject application requires more features than are set forth in the claims. Indeed, less than all of the features of a single embodiment disclosed above.
Claims (3)
1. A method for processing warehouse-order mortgage data, which is applied to a server and comprises the following steps:
receiving an electronic warehouse bill mortgage application request;
Determining a target electronic warehouse list to be mortgage according to the first label of the mortgage application request;
extracting characteristic information in the target electronic warehouse bill, and evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result;
sending the approval mortgage result information based on the evaluation result, generating a signed mortgage protocol file corresponding to the target electronic warehouse bill, and sending the signed mortgage protocol file to an account of a corresponding user through encryption processing, wherein the method specifically comprises the following steps of: comparing the mortgage protocol file with a preset data threshold value to obtain a comparison result, and dividing the mortgage protocol file into a plurality of data blocks with approximate sizes based on the comparison result; distributing the data blocks to a plurality of corresponding encryption units, encrypting the data blocks based on the encryption units to obtain a plurality of encrypted data, and specifically comprising the following steps: generating a random key, generating corresponding encryption noise for the random key, and acquiring a sequential ciphertext, wherein the generation of the key is a plurality of processes, the initial key generated at first is the random key, and the random key is subjected to multi-process conversion and key structure deformation in the subsequent process, wherein the generation process of the random key is as follows: let K i∈R,Si e R be the infrastructure of the key, the symmetric key be K (n) = [ (K i,si),…,(kn,sn) ], where n > 3, and the infrastructure for the symmetric key meets the following requirements: Wherein the symmetric key is an initial key structure, and processing the structured key comprises: configuring n-1 pairs of encryption noise P= [ (r 1,p1),…,(rn-1,pn-1) ] corresponding to the initial secret key, determining a sequential ciphertext based on the encryption noise and the symmetric key, and specifically calculating the sequential ciphertext: Wherein m is a plaintext, namely the plaintext of the data of the file to be encrypted; sequentially adjusting the first sub-ciphertext in the sequential ciphertext based on the mapping relation and generating a second ciphertext, wherein the method specifically comprises the following steps of: randomly encrypting the multiple sub-ciphertexts in the sequence cipher text based on a mapping relation to obtain multiple first sub-ciphertexts; the plurality of first sub-ciphertexts are placed at the tail part of the sequence ciphertexts, and a plurality of second ciphertexts are finally obtained, and the method specifically comprises the steps of mapping an ith sub-ciphertext i c in the sequence ciphertexts into a jth sub-ciphertext of the disordered ciphertext according to a function, and recording the jth sub-ciphertext as j dc, so d j =1, wherein the mapping function is as follows: n (i) =j, 1+.i, j+.n, where for i being a natural set of numbers, specifically i e [1,2, …, n ], the combination of mapping results is defined as j, so c i and j dc satisfy: Wherein C ' is a sequential ciphertext and C ' is an out-of-order ciphertext, wherein the out-of-order ciphertext C ' is: c "= (C d1,…cdn), wherein the out-of-order ciphertext is the second ciphertext; encrypting the second ciphertext based on the mapping relation to obtain a second sub ciphertext, which specifically comprises: encrypting the second ciphertext based on an AES encryption algorithm and the mapping relation to obtain a second sub-ciphertext, encrypting the second ciphertext and the second sub-ciphertext as a final ciphertext, wherein the second ciphertext is an disordered ciphertext, encrypting the disordered ciphertext and the second sub-ciphertext serving as encrypted data, and encrypting the disordered ciphertext and the second sub-ciphertext to obtain a plaintext m, wherein the finally generated ciphertext is: c= [ C d1,…,cdn,cn+1 ], where the expression of plaintext m is: the derivation process for plaintext m is based on the expression of sequential ciphertext, and may be the following process: The method for aggregating the plurality of encrypted data to obtain the spliced target encrypted data specifically comprises the following steps: acquiring text offset of encrypted data of a plurality of data blocks, and sorting based on the offset to obtain the encrypted data of the plurality of sorted data blocks; splicing the encrypted data of a plurality of data blocks based on the sequence to form the target encrypted data;
extracting characteristic information in the target electronic warehouse bill, and evaluating the characteristic information based on a preset risk evaluation model to obtain an evaluation result, wherein the method comprises the following steps:
Processing the characteristic information to be processed based on a preset knowledge graph template to obtain a target knowledge graph containing the information to be processed;
Learning and representing the target knowledge graph to obtain initial data; an evaluation system is arranged in the knowledge graph template, and a specific scoring weight is arranged in the scoring system in the knowledge graph template; the evaluation system adopts a three-layer index structure, namely a total target, a sub-target layer and an index layer, wherein the specific index structure aiming at a third layer is set as follows: operating profit margin, total asset return rate, flow rate, snap action rate, asset load rate, total asset turnover rate, net profit growth rate, leading credit information, mortgage variable capacity, mortgage price stability, mortgage legitimacy, mortgage loss resistance capacity, insurance amount, insurance age, commodity price fluctuation, exchange rate fluctuation, related economic policies, regional development, professional staff number, information management system, organization architecture, warehouse in and out keeping system, cargo evaluation system; and according to the characteristics of scientific research activities, the selected index elements are subjected to homogeneous classification to form 10 secondary indexes, wherein the secondary indexes are respectively: mortgage base conditions, applicant team members, credit levels, product conditions, economic policy conditions, international monetary policies, insurance capabilities, business conditions, management levels, loss bearing capabilities; classifying and profiling the 10 secondary indexes to form 4 primary indexes, wherein the primary indexes are respectively as follows: support index, input index, output index, benefit index; the structure of the knowledge graph is as follows: the entity type set is X= { mortgage basic condition, applicant team member, credit level, product condition, insurance capability, operation condition and management level }, then attribute characteristics of each entity are constructed, the mortgage basic condition= { mortgage change force, mortgage price stability, mortgage legality and mortgage loss resistance capability }, the applicant team member= { professional technician number and organization structure }, credit level= { leading credit information }, product condition= { commodity price fluctuation }, insurance capability= { insurance amount, insurance period, insurance company }, operation condition= { operation profit rate, total asset return rate, flow rate, quick action rate, asset load rate, total asset turnover rate and net profit growth rate }, and management level = { information management system and in-out warehouse keeping system };
inputting the data to be processed into a trained BP neural network to obtain expected output; the BP neural network comprises an input layer, a hidden layer and an output layer, wherein the number of the input layer is the same as that of the evaluation indexes, and the hidden layer is set by any one of the following formulas: Wherein k is the number of hidden layer nodes, n is the number of output layer nodes, l is the number of output layer nodes, a is any constant of 1-10, and the acquisition of the number of hidden layer nodes is determined based on the number of three-level evaluation indexes in an evaluation system; the evaluation index is correspondingly set with the entity characteristics of the target electronic warehouse bill; the training process of the BP neural network comprises two parts, wherein the first part is the conversion of the data structure of the knowledge-graph template, and the second part is the specific process of training the BP neural network based on the data converted by the data structure; the conversion of the data structure for the knowledge-graph template comprises: extracting entities, relations and relation planes of a triplet structure in the knowledge graph to perform vector initialization, constructing a negative triplet based on a positive triplet, training the positive triplet and the negative triplet, updating the entities, the relations and the relation plane vectors according to a loss function reducing direction by adopting a random gradient descent optimization method, and stopping iteration until the loss function tends to be stable, so as to obtain vector representations of the optimized entities, relations and relation planes; The negative example triples are obtained by randomly replacing a head entity or a tail entity in the positive example triples; The data extraction and labeling are required for the acquired data of the structural conversion, and specifically: after data conversion, each entity in the target knowledge graph has an entity vector corresponding to the entity, each relation has a corresponding relation vector and a hyperplane, and the projection vector set Er of all entity vectors in the grading relation hyperplane is extracted by using the following formula, wherein the formula is as follows: e r＝Ei-wr T×Ei×wr, wherein Eri E Er represents a projection vector of an entity vector in a hyperplane corresponding to a scoring relation; Ei represents the entity vector; wr represents the normal vector of the hyperplane corresponding to the scoring relation; determining a third layer of evaluation index in the evaluation system, marking the entity, and inputting the marked data as training to a corresponding input layer; corresponding tests are carried out on the trained BP neural network, and the test method comprises the following steps: obtaining target expected output based on an evaluation system, obtaining real-time expected output based on the BP neural network, comparing the relation between the target expected output and the BP neural network, and determining whether the trained BP neural network is completed based on a set threshold; Training for BP neural networks involves the following process: the construction method for the evaluation system comprises the following steps: an evaluation system is constructed based on the first-layer, second-layer and third-layer evaluation indexes, and specifically comprises the following steps: establishing an importance influence relation among layers, namely setting that m factors have relative importance to a certain factor of the upper layer, comparing and judging an ith factor (i=1, 2,3, …, m) with a jth factor (j=1, 2,3, … m) according to a specific scale rule, wherein the relative importance degree is aij, and constructing an m-order matrix which can be used for solving the priority of each factor about certain criteria, namely judging the matrix, namely, the matrix is recorded as: a= (a ij)m*m, the judgment matrix is established by a 1-9 scale method, and specifically, j and m in a jm represent the j and m factors, and the judgment matrix is respectively constructed for a first-level index and 10 second-level indexes by using an expert interview method and an analytic hierarchy process, and the method is specifically as follows: wherein the matrix for the 10 secondary indexes is respectively: the mortgage basic condition matrix is as follows: the applicant team member matrix is: The credit level matrix is: the product condition matrix is: The economic policy situational matrix is: the operation condition matrix is as follows: the international currency policy matrix is: The management level matrix is: the loss-bearing capacity matrix is: The insurance capability matrix is: Based on the above matrices, the maximum feature root λ max and the corresponding feature vector W of each matrix are obtained, namely AiW =λ max, and the specific process of this process includes: the components (W1, W2, …, wm) of the feature vectors W, W corresponding to the respective judgment matrices are calculated as the relative importance levels corresponding to the m elements, namely, the weight coefficients, wherein the method for calculating the weight coefficients includes a sum-product method and a root method, and the sum-product method is adopted, so only the sum-product method will be described: normalizing each column of the judgment matrix: For the judgment matrix normalized by columns, and then by rows Summing, obtaining normalized vector of the sum result,Then w= [ W1, W2, W3, … Wm ] is the result, according to the calculation formula of the maximum feature pair: Wherein Awi denotes the ith component of the vector AW, the weight of each corresponding secondary index and the maximum eigenvalue of the corresponding index target can be obtained by calculation according to the above formula, and the eigenvector W and the single rank matrix P of each index can be obtained, wherein the eigenvector matrix is as follows: w= (0.0231,0.01165,0.0331,0.1521,0.0616,0.2090,0.0816,0.1109,0.2680,0.0440) T; the single rank matrix for the secondary hierarchy is as follows: the mortgage basic condition matrix is as follows: p 2＝(0.8000,0.2000)T, applicant team member matrix is: p 3＝(0.8000,0.2000)T, credit level matrix is: p 4＝(0.2000,0.8000)T, the product condition matrix is: p 5＝(0.1782,0.7514,0.0704)T, economic policy situational matrix is: p 6＝(0.7500,0.2500)T, the management situation matrix is: p 7＝(0.4402,0.0920,0.1470,0.0368,0.2350,0.0491)T, international monetary policy matrix is: p 8＝(0.2000,0.8000)T, management level matrix is: p 9＝(0.1667,0.8333)T, the loss-bearing capacity matrix is: p 10＝(0.1667,0.8333)T, the safety capability matrix is: p 11＝(0.1667,0.8333)T.
2. The method of claim 1, further comprising performing decryption processing based on the second sub-ciphertext and the second sub-ciphertext, comprising:
Decrypting the second sub-ciphertext to obtain a first mapping relation;
processing the second ciphertext based on the mapping relation to obtain the sequential ciphertext;
summing the sequential ciphertext based on the random key to obtain a set of information about the random key;
Deriving plain text based on the set.
3. A warehouse entry mortgage data processing apparatus performing the method of claim 1, comprising:
the request receiving module is used for receiving an electronic warehouse bill mortgage application request;
the target electronic warehouse list determining module is used for determining a target electronic warehouse list to be mortgage according to the first label of the mortgage application request;
the evaluation module is used for evaluating the target electronic warehouse bill;
The encryption module is used for carrying out encryption processing on the target electronic warehouse bill in the transmission process; the BP neural network comprises an input layer, a hidden layer and an output layer, wherein the number of the input layer is the same as that of the evaluation indexes; the evaluation index is correspondingly set with the entity characteristics of the target electronic warehouse bill; the training process of the BP neural network comprises two parts, wherein the first part is the conversion of the data structure of the knowledge-graph template, and the second part is the specific process of training the BP neural network based on the data converted by the data structure; the conversion of the data structure for the knowledge-graph template comprises: extracting entities, relations and relation planes of a triplet structure in the knowledge graph to perform vector initialization, constructing a negative triplet based on the positive triplet, training the positive triplet and the negative triplet, updating the entities, the relations and the relation plane vectors according to a loss function reducing direction by adopting a random gradient descent optimization method, and stopping iteration until the loss function tends to be stable, so as to obtain vector representations of the optimized entities, relations and relation planes; the negative example triples are obtained by randomly replacing a head entity or a tail entity in the positive example triples; the data extraction and labeling are required for the acquired data of the structural conversion, and specifically: after data conversion, each entity in the target knowledge graph has an entity vector corresponding to the entity, each relation has a corresponding relation vector and hyperplane, and the projection vector set Er of all entity vectors in the scoring relation hyperplane is extracted by using the following formula, wherein the formula is as follows: Wherein, eri E Er represents a projection vector of an entity vector in a hyperplane corresponding to a scoring relation; ei represents the entity vector; wr represents the normal vector of the hyperplane corresponding to the scoring relation; determining a third layer of evaluation index in the evaluation system, marking the entity, and inputting the marked data as training to a corresponding input layer; corresponding tests are carried out on the trained BP neural network, and the test method comprises the following steps: and obtaining target expected output based on an evaluation system, obtaining real-time expected output based on the BP neural network, comparing the relation between the target expected output and the BP neural network, and determining whether the trained BP neural network is completed based on a set threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211255061.3A CN115600226B (en) | 2022-10-13 | 2022-10-13 | Warehouse bill mortgage data processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211255061.3A CN115600226B (en) | 2022-10-13 | 2022-10-13 | Warehouse bill mortgage data processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115600226A CN115600226A (en) | 2023-01-13 |
| CN115600226B true CN115600226B (en) | 2024-10-11 |
Family
ID=84847268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211255061.3A Active CN115600226B (en) | 2022-10-13 | 2022-10-13 | Warehouse bill mortgage data processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115600226B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116305225B (en) * | 2023-05-24 | 2023-08-18 | 山东梧桐树软件有限公司 | User data encryption protection method used in online payment process |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107317666A (en) * | 2017-05-25 | 2017-11-03 | 南京邮电大学 | A kind of parallel full homomorphism encipher-decipher method for supporting floating-point operation |
| CN111639193A (en) * | 2020-05-27 | 2020-09-08 | 上海明略人工智能(集团)有限公司 | Product risk assessment method and device, electronic equipment and storage medium |
| CN113570448A (en) * | 2020-11-10 | 2021-10-29 | 中储京科供应链管理有限公司 | Data processing method, equipment and medium in electronic warehouse elementary substance escort pledge |
| CN113590837A (en) * | 2021-07-29 | 2021-11-02 | 华中农业大学 | Deep learning-based food and health knowledge map construction method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101089890A (en) * | 2007-07-12 | 2007-12-19 | 中国工商银行股份有限公司 | Loan accept system based on network and its loan accept terminal |
| CN109685651A (en) * | 2019-01-11 | 2019-04-26 | 中信梧桐港供应链管理有限公司 | The credit risk monitoring method and device of current assets mortgage |
| CN110609903B (en) * | 2019-08-01 | 2022-11-11 | 华为技术有限公司 | Information presentation method and device |
| CN114445140A (en) * | 2022-01-29 | 2022-05-06 | 中国建设银行股份有限公司 | Value evaluation method, early warning method, device and equipment |
| CN114610899A (en) * | 2022-03-07 | 2022-06-10 | 支付宝(杭州)信息技术有限公司 | Representation learning method and system of knowledge graph |
-
2022
- 2022-10-13 CN CN202211255061.3A patent/CN115600226B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107317666A (en) * | 2017-05-25 | 2017-11-03 | 南京邮电大学 | A kind of parallel full homomorphism encipher-decipher method for supporting floating-point operation |
| CN111639193A (en) * | 2020-05-27 | 2020-09-08 | 上海明略人工智能(集团)有限公司 | Product risk assessment method and device, electronic equipment and storage medium |
| CN113570448A (en) * | 2020-11-10 | 2021-10-29 | 中储京科供应链管理有限公司 | Data processing method, equipment and medium in electronic warehouse elementary substance escort pledge |
| CN113590837A (en) * | 2021-07-29 | 2021-11-02 | 华中农业大学 | Deep learning-based food and health knowledge map construction method |
Non-Patent Citations (2)
| Title |
|---|
| 层次分析法和神经网络的信息系统风险评估;赵保华;微电子学与计算机;20151031;第32卷(第10期);163-166 * |
| 融合知识图谱与卷积神经网络的电压暂降分类识别;白浩等;电力信息与通信技术;20211130;第19卷(第11期);131-139 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115600226A (en) | 2023-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3985578A1 (en) | Method and system for automatically training machine learning model | |
| US20220075670A1 (en) | Systems and methods for replacing sensitive data | |
| Diana et al. | Minimax group fairness: Algorithms and experiments | |
| EP4222629A1 (en) | Predicting data tampering using augmented machine learning models | |
| US11893493B2 (en) | Clustering techniques for machine learning models | |
| US12050625B2 (en) | Systems and methods for classifying imbalanced data | |
| Rastegarpanah et al. | Auditing black-box prediction models for data minimization compliance | |
| CN112861003A (en) | User portrait construction method and system based on cloud edge collaboration | |
| CN113674087A (en) | Enterprise credit rating method, apparatus, electronic device and medium | |
| Wu et al. | Application analysis of credit scoring of financial institutions based on machine learning model | |
| CN115600226B (en) | Warehouse bill mortgage data processing method and device | |
| CN117235811A (en) | Internet finance-based payment big data analysis method and big data analysis system | |
| US11281728B2 (en) | Data generalization for predictive models | |
| Zhao | Comprehensive Budget Execution Performance Evaluation of Companies Incorporating EVA Unsupervised Learning Model | |
| CN116150663A (en) | Data classification method, device, computer equipment and storage medium | |
| Iwai et al. | Bayesian Network Oriented Transfer Learning Method for Credit Scoring Model | |
| CN111209403A (en) | Data processing method, device, medium and electronic equipment | |
| Peng | Research on credit risk identification of internet financial enterprises based on big data | |
| CN115409127A (en) | Training method and related device of contract default risk prediction model | |
| CN115018627A (en) | Credit risk evaluation method and device, storage medium and electronic equipment | |
| CN115018608A (en) | Risk prediction method and device and computer equipment | |
| Abassi et al. | Crowd label aggregation under a belief function framework | |
| Kong et al. | The risk prediction of mobile user tricking account overdraft limit based on fusion model of logistic and GBDT | |
| Kesari | Predicting Cybersecurity Incidents Through Mandatory Disclosure Regulation | |
| Piesio et al. | Applying machine learning to anomaly detection in car insurance sales |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |