CN114386727A - Risk identification method, device, equipment and storage medium - Google Patents
Risk identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN114386727A CN114386727A CN202011118242.2A CN202011118242A CN114386727A CN 114386727 A CN114386727 A CN 114386727A CN 202011118242 A CN202011118242 A CN 202011118242A CN 114386727 A CN114386727 A CN 114386727A
- Authority
- CN
- China
- Prior art keywords
- risk
- transaction
- entity
- conduction path
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The application relates to a risk identification method, a risk identification device, risk identification equipment and a storage medium, which belong to the technical field of computers, wherein the method comprises the following steps: acquiring transaction data of a target entity, and constructing a transaction relationship map of the target entity by taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, wherein each edge in the transaction relationship map corresponds to a risk conduction path; for each risk conduction path, acquiring each transaction between two nodes corresponding to the risk conduction path from transaction data, calculating a time attenuation coefficient of each transaction, and determining a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction; determining a target risk score for the target entity based on the weight factor for each risk conductive pathway. According to the method and the system, the influence of different associated entities on the target entity and the timeliness of transaction data are integrated, and risk identification can be performed more accurately.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a risk identification method, apparatus, device, and storage medium.
Background
At present, when entities such as enterprises, institutions and merchants are subjected to risk identification, due to the fact that data which can be obtained are limited, the concentration and stability of upstream and downstream transactions among the entities are mainly used as evaluation indexes. The more centralized the upstream and downstream transaction proportion of the entity, the higher the operational risk of the entity; the more upstream and downstream the entity changes, the lower the stability and the higher the operational risk of the entity.
But the upstream and downstream transactions among the entities are time-sensitive, and the concentration or stability can only represent the concentration or stability in a specific time period in the past; also, for a particular entity, each entity associated with the particular entity has a different impact on its business risk. Therefore, the risk identification is carried out only according to the concentration and the stability, the accuracy is low, and the real risk condition of the entity cannot be reflected.
Disclosure of Invention
The application provides a risk identification method, a risk identification device, risk identification equipment and a storage medium, which can synthesize the influence of different associated entities on a target entity and the timeliness of transaction data, and can more accurately identify the risk of the target entity.
In one aspect, the present application provides a risk identification method, including:
acquiring transaction data of a target entity, and constructing a transaction relationship map of the target entity by respectively taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, wherein each edge in the transaction relationship map corresponds to a risk conduction path;
for each risk conduction path, acquiring each transaction between two nodes corresponding to the risk conduction path from the transaction data, calculating a time attenuation coefficient of each transaction, and determining a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction;
determining a target risk score for the target entity based on the weight factor for each of the risk conductive pathways.
Another aspect provides a risk identification apparatus, the apparatus comprising:
the transaction relationship graph comprises a conducting path construction module, a risk conducting path construction module and a risk conducting path construction module, wherein the conducting path construction module is used for acquiring transaction data of a target entity, and constructing a transaction relationship graph of the target entity by taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, and each edge in the transaction relationship graph corresponds to one risk conducting path;
a path weight determination module, configured to, for each risk conduction path, obtain, from the transaction data, each transaction between two nodes corresponding to the risk conduction path, calculate a time attenuation coefficient of each transaction, and determine a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction;
a risk identification module to determine a target risk score for the target entity based on the weighting factor for each of the risk conductive pathways.
Another aspect provides a risk identification device, which includes a processor and a memory, where at least one instruction or at least one program is stored in the memory, and the at least one instruction or at least one program is loaded by the processor and executes the risk identification method as described above.
Another aspect provides a computer storage medium having at least one instruction or at least one program stored therein, the at least one instruction or the at least one program being loaded and executed by a processor to implement the risk identification method as described above.
The risk identification method, the risk identification device, the risk identification equipment and the risk identification storage medium have the following beneficial effects:
determining the upstream and downstream relationship between the target entity and each associated entity as a risk conduction path, and respectively identifying the risk conduction of each associated entity to the target entity; by introducing a time attenuation coefficient into each transaction, the weight factor of each risk conduction path is more reasonable, and the accuracy of risk identification based on the weight factor of each risk conduction path is higher.
Drawings
In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a schematic diagram of a risk identification system according to an embodiment of the present application.
Fig. 2 is a schematic flowchart of a risk identification method according to an embodiment of the present application.
Fig. 3 is a schematic structural diagram of a transaction relationship map provided in an embodiment of the present application.
Fig. 4 is a graph illustrating a time decay function provided in an embodiment of the present application.
Fig. 5 is a schematic flowchart of another risk identification method according to an embodiment of the present application.
Fig. 6 is a schematic flowchart of another risk identification method according to an embodiment of the present application.
Fig. 7 is a schematic flowchart of determining a target risk score according to an embodiment of the present disclosure.
Fig. 8 is an example of a transaction relationship graph provided by an embodiment of the present application.
Fig. 9 is an example of initial risk scores for nodes in a transaction relationship graph provided by an embodiment of the present application.
Fig. 10 is an example of separately calculating each risk conductive path weight value corresponding to an upstream entity according to an embodiment of the present application.
Fig. 11 is an example of separately calculating each risk conductive path weight value corresponding to a downstream entity according to an embodiment of the present application.
Fig. 12 is an example of calculating the weight values of the risk conductive paths as a whole upstream and downstream according to the embodiment of the present application.
Fig. 13 is a schematic structural diagram of a risk identification device according to an embodiment of the present application.
Fig. 14 is a schematic structural diagram of a path weight determining module in the risk identification device according to the embodiment of the present application.
Fig. 15 is a schematic structural diagram of another risk identification device provided in an embodiment of the present application.
Fig. 16 is a schematic structural diagram of another risk identification device provided in an embodiment of the present application.
Fig. 17 is a schematic structural diagram of a second risk identification unit in the risk identification device according to the embodiment of the present application.
Fig. 18 is a hardware structural diagram of an apparatus for implementing the method provided by the embodiment of the present application.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, the present application will be further described in detail with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The existing risk identification scheme mainly takes the concentration and stability in up-down transactions between entities as identification indexes, and the more concentrated the transaction proportion is, the higher the operating risk of the enterprise is; the more annual changes (changes per year) of the first N customers of the enterprise, the lower the operational stability and the higher the operational risk of the enterprise. For example, the following table shows the transaction proportion of the downstream customers of the enterprise in 2019 counted by the invoice data, the concentration ratio is used as an identification index, and the transaction proportion of the first 5 customers occupies 80% of the whole transaction of the enterprise, and if the transaction proportion is concentrated, the operation risk of the enterprise is considered to be high.
Table 1: transaction proportion of downstream customers of a certain enterprise
| Name of customer | Transaction amount (ten thousand) | Transaction proportion |
| Guangdong division, Inc. of China | 240 | 57.55％ |
| Guangzhou scientific Co Ltd | 52 | 12.47％ |
| Engineering, Inc., Guangdong province | 42 | 10.07％ |
| Guangzhou city company Limited for Metal product | 34 | 8.15％ |
| Southern Guangdong-construction Ltd | 19 | 4.56％ |
| Guangdong communication equipment Co., Ltd | 11 | 2.64％ |
| Guangdong:communicationengineering Co Ltd | 6 | 1.44％ |
| Shaoguan division, Inc. of China | 4 | 0.96％ |
| Dongguan division, Inc. of China | 3 | 0.72％ |
| Qingyuan division of China | 3 | 0.72％ |
| |
3 | 0.72％ |
| Total of | 416 | 100％ |
However, the transactions between the entities and the downstream are time-sensitive, for example, the above table only shows the data of 2019, if the data of the previous two years or the previous three years are counted, the transaction proportion will change, and the risk identification result will be different. Moreover, it is common that trading is concentrated on large well-operated and stable enterprises, which is beneficial to operation, rather than increasing operational risk. In addition, the relationship between the upstream and downstream enterprises is not considered, and each enterprise associated with an enterprise has a different influence on its operational risk, for example, the transaction ratio of "guangdong" communications engineering limited company "is only" 1.44% ", but actually the company has a large influence on the operational risk of the whole enterprise. Therefore, the risk is identified only according to the concentration ratio and the stability, and certain one-sidedness is achieved.
According to the method, firstly, the incidence relation between the target entity and each incidence entity is extracted from the transaction data of the target entity, and a transaction relation map is constructed based on the incidence relation. Each edge in the transaction relationship map corresponds to the incidence relationship between the target entity and a certain incidence entity, so that the operation risk influence of each incidence entity identification on the target entity can be identified. Then, aiming at all transactions between two nodes corresponding to each edge in the transaction relation graph, determining the weight value of each edge according to the time attenuation coefficient of each transaction so as to eliminate the timeliness influence of the transactions and further more accurately identify the operation risk of the target entity.
Referring to fig. 1, a schematic diagram of a risk identification system according to an embodiment of the present application is shown. As shown in fig. 1, the risk identification system may include at least a client 01 and a server 02.
Specifically, the client 01 may include a smart phone, a desktop computer, a tablet computer, a notebook computer, a digital assistant, a smart wearable device, and other types of devices, and may also include software running in the devices, such as web pages provided by some service providers to users, and applications provided by the service providers to users. Specifically, the client 01 may be configured to present a target entity, so that a user triggers risk identification of the target entity.
Specifically, the server 02 may include a server operating independently, or a distributed server, or a server cluster composed of a plurality of servers. The server 02 may comprise a network communication unit, a processor and a memory, etc. Specifically, the server 02 may be used for risk identification of a target entity.
A risk identification method of the present application is described below. Fig. 2 is a schematic flow chart of a risk identification method provided in an embodiment of the present application, and the present specification provides the method operation steps as described in the embodiment or the flow chart, but more or less operation steps may be included based on conventional or non-inventive labor. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. In practice, the system or server product may be implemented in a sequential or parallel manner (e.g., parallel processor or multi-threaded environment) according to the embodiments or methods shown in the figures. Specifically, as shown in fig. 2, the method may include:
s201, acquiring transaction data of a target entity, and constructing a transaction relationship map of the target entity by respectively taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, wherein each edge in the transaction relationship map corresponds to a risk conduction path.
In the embodiment of the present application, the category to which the entity (including the target entity and the associated entity) belongs may include a business entity, and an organization entity. The transaction data may be invoice data or bill data and the like reflecting transactions between the target entity and the upstream and downstream entities, and the transaction data at least includes transaction amount and transaction time corresponding to each transaction between the target entity and each associated entity. The associated entity refers to an entity having a transaction relationship with a target entity, and the associated entity may be an upstream entity of the target entity or a downstream entity of the target entity. The upstream entity refers to an entity that provides services for the target entity, such as a provider of the target entity; a downstream entity refers to an entity that is served by a target entity, e.g., a customer of the target entity.
After the server acquires the transaction data of the target entity, the upstream and downstream relations between the target entity and each associated entity in the transaction data are extracted, and a transaction relation map is established based on the upstream and downstream relations, wherein the structure of the established transaction relation map can be shown in fig. 3. In fig. 3, the target entity 1 has 5 associated entities, each entity corresponds to one node, and the upstream and downstream relationship between the target entity and each associated entity forms an edge, which includes 6 nodes and 5 edges. All edges have a common node, namely a target entity, each edge is a unidirectional edge, an edge pointed to the target entity by the associated entity is defined as an incoming edge, and an edge pointed to the target entity by the target entity is defined as an outgoing edge.
All incoming edges in the transaction relationship graph have the same directional attribute, all outgoing edges have the same directional attribute, and each edge corresponds to one risk conduction path, so that the directional attribute of each risk conduction path indicates that the associated entity corresponding to the risk conduction path is an upstream entity or a downstream entity. For example, in fig. 3, the associated entity 2 corresponding to the risk conduction path 7 (i.e., edge 7) is an upstream entity, and the associated entity 5 corresponding to the risk conduction path 10 (i.e., edge 10) is a downstream entity.
After the server completes the construction of the transaction relationship graph, the risk conduction result of the target entity from the upstream entity and the downstream entity can be calculated by using a graph calculation mode of a pagerank algorithm.
S202, aiming at each risk conduction path, acquiring each transaction between two nodes corresponding to the risk conduction path from the transaction data, calculating the time attenuation coefficient of each transaction, and determining the weight factor of the risk conduction path according to the time attenuation coefficient of each transaction.
In the transaction relationship map, two nodes corresponding to each risk conduction path are a target entity and an associated entity, and for each transaction between the target entity and each associated entity, a time attenuation coefficient of the transaction can be calculated according to the transaction time of the transaction. Specifically, the calculating the time decay coefficient of each transaction may include: calculating a time interval between the transaction time corresponding to the transaction and the current time; and according to the time interval, obtaining the time attenuation coefficient of the transaction according to a pre-established time attenuation function.
In the embodiment of the present application, the time decay function indicates the decay caused by the time interval, and the time decay function f (t) can be represented by the following expression:
where k is the decay parameter and t is the time interval between the current time and the transaction time. In particular implementations, t may be in units of quarters, or months or years, i.e., time intervals may characterize quarter differences, month differences, or year differences, etc. It will be appreciated that the time interval t may also be other values of reaction time difference, e.g. number of days in between, etc.
The time attenuation function can be used for showing that the upstream and downstream relation represented by the transaction data is time-efficient, the strength of ineffectiveness influence can be set by adjusting the value of the attenuation parameter k, and when k is fixed, the time attenuation coefficient of each transaction is determined. For example, if the time interval is in quarters, when k is 8, it indicates that transactions over 8 quarters have negligible impact on the risk identification result; when k is 16, it indicates that transactions over 16 quarters have negligible impact on risk identification results. As shown in fig. 4, it is a time decay function f (t) over a time interval t when k is equal to 8.
Accordingly, the following table shows the values of the time attenuation factor f (t) for the time interval t in quarterly units. As can be seen from table 2, the time decay coefficient indicates that the upstream and downstream relationship represented by the invoice data is seasonal, and the closer to the current season, the stronger the upstream and downstream relationship.
Table 2: corresponding relation between time interval and time attenuation coefficient
| Difference of quarter | Time interval t | Time attenuation coefficient F (t) |
| |
0 | 1.00 |
| Last 1 |
1 | 0.93 |
| Last 2 |
2 | 0.87 |
| Last 3 |
3 | 0.79 |
| Last 4 |
4 | 0.70 |
| Last 5 |
5 | 0.61 |
| Last 6 |
6 | 0.50 |
| Last 7 seasons | 7 | 0.30 |
| Last 8 |
8 | 0 |
After determining the time attenuation coefficient of each transaction, the server may determine a weight factor of the risk conduction path according to the time attenuation coefficients of all transactions between two nodes corresponding to the risk conduction path, where the weight factor characterizes a parameter that affects the weight of the risk conduction path in risk identification. Specifically, the determining a weight factor of the risk conduction path according to a time attenuation coefficient of each transaction may include: for each transaction, determining the product of the time attenuation coefficient of the transaction and a transaction attribute value corresponding to the transaction as a weight factor of the transaction, wherein the transaction attribute value comprises a transaction amount or a transaction importance degree; and superposing the weight factors of all the transactions to obtain the weight factor of the risk conduction path.
In the embodiment of the present application, the mathematical relationship for calculating the weight factor of each risk conduction path may be represented as:
wherein n represents n transactions between two nodes corresponding to risk conduction paths, viTransaction attribute value representing the ith transaction, F (t)iRepresenting the time decay factor of the ith transaction.
In the embodiment of the application, the transaction attribute value may be a transaction amount or a transaction importance, and the transaction importance characterizes a value of the transaction importance. In the specific implementation, the importance degree of each transaction between a certain associated entity and a target entity may be the same or different, and the description is not limited specifically.
S203, determining a target risk score of the target entity based on the weight factor of each risk conduction path.
To differentiate the impact of different associated entities on target entity risk conductance, as shown in fig. 5, before determining a target risk score for the target entity based on the weight factors of the various risk conductance pathways, the method further comprises:
s501, obtaining an initial risk score of each node in the transaction relationship graph.
Wherein, the initial risk score of each node is determined in advance according to the operation information (such as financial statement) of the node, and is a preliminary identification of the operation risk of the node. In fact, the operation information of each node is not all acquirable, and when the operation information of a node cannot be acquired, the initial risk score of the node may be set as a default value, or the initial risk score of the node may be set as a risk average value of a corresponding industry or region.
In one particular embodiment, the initial risk scores may be ranked, with different ranks corresponding to different risks. For example, if the initial risk score is divided into 5 levels, and the value is 0, 1, 2, 3 or 4, and the risk is gradually increased from 0 to 4, 0 indicates that the node is in good operation (the operation risk is small), and 4 indicates that the node is in difficult operation (the risk of closing is large).
Accordingly, with continued reference to fig. 5, the determining a target risk score for the target entity based on the weight factor for each of the risk conductive pathways may include:
s502, determining a target risk score of the target entity based on the initial risk scores of the two nodes corresponding to each risk conduction path and the weighting factors of the risk conduction paths.
When the server identifies the risk of the target entity, the server can have a grouping mode and a non-grouping mode. In the non-grouping mode, the transaction relationship map is taken as a whole, and whether the associated entity is an upstream entity or a downstream entity is not distinguished; in the grouping mode, the transaction relationship maps are grouped, with all upstream entities as a group and all downstream entities as a group.
In some embodiments, if the grouping mode is adopted, as shown in fig. 6, before determining the target risk score of the target entity based on the initial risk scores of the two nodes corresponding to each risk conduction path and the weighting factors of the risk conduction paths, the method may further include:
s601, in the transaction relationship graph, determining risk conduction paths with the same direction attribute as a risk conduction set, wherein the direction attribute indicates that an associated entity corresponding to the risk conduction path is an upstream entity or a downstream entity.
For example, referring to fig. 3, the associated entity 2 corresponding to the risk conduction path 7 is an upstream entity, the associated entity 3 corresponding to the risk conduction path 8 is an upstream entity, the associated entity 4 corresponding to the risk conduction path 9 is an upstream entity, the associated entity 5 corresponding to the risk conduction path 10 is a downstream entity, and the associated entity 3 corresponding to the risk conduction path 11 is a downstream entity. Thus, risk conduction path 7, risk conduction path 8, and risk conduction path 9 have the same directional property as one risk conduction set; the risk conduction path 10 and the risk conduction path 11 have the same directional property as one risk conduction set.
Accordingly, continuing with fig. 6, determining a target risk score for the target entity based on the initial risk scores of the two nodes corresponding to each of the risk conduction paths and the weight factors for the risk conduction paths may include:
s602, for each risk conduction set, calculating a weight value of each risk conduction path according to the weight factor of each risk conduction path in the risk conduction set, and determining a target risk score of the target entity according to the initial risk score of the two nodes corresponding to each risk conduction path and the weight value of the risk conduction path.
Specifically, as shown in fig. 7, the calculating a weight value of each risk conduction path according to a weight factor of each risk conduction path in the risk conduction set may include:
s6021, adding the weight factors of the risk conduction pathways in the risk conduction set to obtain a weight factor of the risk conduction set.
And S6022, based on the weight factor of the risk conduction set, carrying out normalization processing on the weight factor of each risk conduction path to obtain a weight value of each risk conduction path.
In the embodiment of the present application, the weighted value v of the ith risk conduction pathiCan be represented by the following formula:
wherein m represents the total number of at-risk conduction pathways, skA weight factor representing the k-th risky conduction path.
Continuing with fig. 7, said determining a target risk score for said target entity based on the initial risk scores of the two nodes corresponding to each of said risk conduction paths and the weight values of said risk conduction paths may comprise:
s6023, for each risk conduction path, calculating an initial risk score of the associated entity corresponding to the risk conduction path, and multiplying the initial risk score by the weight value of the risk conduction path to obtain a risk conduction value of the risk conduction path.
And S6024, superposing the risk conduction values of the risk conduction paths to obtain a risk deviation value of the target entity.
And S6025, determining the sum of the initial risk score and the risk offset value of the target entity as a target risk score corresponding to the target entity.
In the embodiment of the present application, the target risk score may be calculated by the following equation:
wherein r isjInitial risk score, w, representing the associated entity corresponding to the jth risk conduction pathwayjWeight value, r, representing the jth risk conduction pathj*wjA risk conduction value representing the jth risk conduction path, m represents the total number of risk conduction paths,
a risk offset value, r, representing the target entity0Representing an initial risk score for the target entity.
In specific implementation, the target risk score r may be rounded to obtain an initial risk score r0And the corresponding grade is obtained, so that whether the operation risk is increased or not is determined by the target entity through the upstream and downstream transaction relation on the basis of the initial risk score, namely:
wherein [ ] is the rounding operation.
It should be noted that, in the above formula for calculating the target risk score, only the time attenuation coefficient is introduced into the risk conduction path, and in some embodiments, the time attenuation coefficient may also be introduced into the initial risk score, so that the formula for calculating the target risk score may be updated as follows:
wherein q is r0The time attenuation coefficient of (2) is between 0 and 1. In specific implementation, can be according to r0The time interval between the calculated time of (a) and the current time is determined in the manner of f (t) above.
In other embodiments, if the non-grouping mode is adopted, the transaction relationship map may be used as a total risk conduction set, and for the total risk conduction set, another target risk score corresponding to the target entity may be determined in the same manner as in steps S6021 to S6025.
By employing a non-grouping mode, the total risk conductance results of the target entity from upstream and downstream entities can be calculated. By adopting the grouping mode, the risk conduction results of the target entity from the upstream entity and the downstream entity can be respectively calculated, and the risk identification of the target entity is facilitated.
It should be noted that, in the grouping mode or the non-grouping mode, the calculation manner of the target risk score determined by each risk conduction set is the same, and the differences between the risk conduction sets are not distinguished. In order to distinguish that different risk conduction sets have different influences on the operation risk of the target entity, for example, the operation influences of suppliers and customers on enterprises may be different, when the server finally confirms the risk of the target entity by using each target risk score, an influence factor can be added to each target risk score for embodying, and the value of the influence factor is between 0 and 1.
In the operation of the entity, the invoice data is the issued and received service certificate, is the original basis for accounting, is also the important basis for law enforcement check of auditing agencies, tax agencies and the like, and has high reliability. The risk identification method according to the embodiment of the present application will be described in detail below with transaction data as invoice data. As shown in Table 3, it is the invoice data portion content for company A and the upstream company (seller) and downstream company (buyer).
Table 3: invoice data partial content of company A and related companies upstream and downstream
| Seller side | Purchaser of goods | Time of invoice | Amount of invoice |
| Company B | Company A | 2020.01.01 | 80W |
| C Corp Ltd | Company A | 2020.04.01 | 70W |
| Company D | Company A | 2019.12.01 | 40W |
| Company D | Company A | 2020.01.01 | 20W |
| ··· | ··· | ··· | ··· |
| Company A | Company E | 2020.03.01 | 60W |
| Company A | Company F | 2020.04.01 | 30W |
| Company A | Company F | 2020.07.01 | 20W |
| ··· | ··· | ··· | ··· |
Based on the invoice data, a transaction relationship map indicating upstream and downstream relationships between companies can be formed, as shown in fig. 8. In fig. 8, each node corresponds to a company, and the edges are upstream or downstream relationships formed by invoicing between companies, and each edge may be an incoming edge or an outgoing edge. Wherein, the edge entry means that company a serves as a client, and other suppliers supply company a, such as edge 7, edge 8, and edge 9 of company a in fig. 8; out-of-edge refers to company a as the provider that supplies other enterprises, such as the out-of-edge of company a in fig. 8, which is edge 10 and edge 11.
In the embodiment of the application, each edge is taken as a risk conduction path, and the risk conduction paths of the supplier and the client to the company A are calculated respectively. In fig. 8, company B, company C, and company D are all upstream companies (suppliers) of company a, and the risk conduction path 7, the risk conduction path 8, and the risk conduction path 9 have the same directional property (all being edge entries), forming a risk conduction set 1; company E and company F are downstream companies (clients) of company a, and the risk conduction path 10 and the risk conduction path 11 have the same directional property (both edge-out), forming a risk conduction set 2. Assuming that the current time is 2020-08-01 (quarter 3 of 2020), risk conduction calculation can be performed on the risk conduction set 1 and the risk conduction set 2 respectively by using the pagerank algorithm.
The server can preliminarily identify the operation risk of each company through the financial statement information of each company, and as shown in fig. 9, the initial risk score r of each company can be obtained0。
For risk conduction set 1, risk conduction path 7 corresponds to 1 invoice between company B and company a, invoice time 2020.01.01 (quarter 1 of 2020), invoice amount 80W, current time 2020-08-01, 2020.01.01 for the last 2 quarters, i.e. time interval 2. According to the time attenuation function F (t), the time attenuation coefficient of the invoice is 0.87. Accordingly, the weight factor of risk conduction path 7 is s7＝80*0.87＝69.6。
There are 2 invoices between company D and company a for risk conduction path 9. In the first invoice, the invoice time is 2019.12.01 (quarter 4 in 2019), the invoice amount is 40W, corresponding to the current time 2020-08-01, 2019.12.01 is the last 3 quarters, that is, the time interval is 3, and the time decay coefficient of the first invoice is 0.79 according to the time decay function F (t). In the second invoice, the invoice time is 2020.01.01 (season 1 in 2020), and the invoice amount is20W, corresponding to the current time 2020-08-01, 2020.01.01 being the last 2 quarters, i.e. the time interval is 2, the time decay factor of the first invoice is 0.87 as obtained from the time decay function f (t). Accordingly, the weight factor of the risk conduction path 9 is s9＝40*0.79+20*0.87＝31.6+17.4＝49。
The weight factor of each risk conduction path in the risk conduction set 1 is normalized, as shown in fig. 10, and the weight value of each risk conduction path can be obtained:
w7＝s7/(s7+s8+s9)＝69.6/(69.6+65.1+49)＝0.379
w8＝s8/(s7+s8+s9)＝65.1/(69.6+65.1+49)＝0.354
w9＝s9/(s7+s8+s9)＝49/(69.6+65.1+49)＝0.267
the upstream risk score for company a is the sum of the initial risk score for company a and the risk of upstream conducted, i.e.:
r＝0+2*0.379+1*0.354+0*0.267＝1.112
with an initial risk score of 0 for company a, an upstream increased risk to company a may be obtained from an enterprise upstream supply perspective.
For Risk conduction set 2, there are 1 invoice between company E and company A for risk conduction path 10, invoice time 2020.03.01 (quarter 1 of 2020), invoice amount 60W, current time 2020-08-01, and current time 2020.03.01 for the last 2 quarters, i.e. time interval 2. According to the time attenuation function F (t), the time attenuation coefficient of the invoice is 0.87, and correspondingly, the weight factor of the risk conduction path 10 is s10＝60*0.87＝52.2。
There are 2 invoices between company F and company a for risk conduction path 11. In the first invoice, the invoice time is 2020.04.01 (quarter 2 in 2020), the invoice amount is 30W, corresponding to the current time 2020-08-01, 2020.04.01 is the last quarter, that is, the time interval is 1, and the time of the first invoice can be obtained according to the time decay function F (t)The inter-attenuation coefficient was 0.97. In the second invoice, the invoice time is 2020.07.01 (quarter 3 in 2020), the invoice amount is 20W, corresponding to the current time 2020-08-01, 2020.07.01 is the current quarter, i.e. the time interval is 0, and the time decay coefficient of the second invoice is 1 according to the time decay function f (t). Accordingly, the weight factor of the risk conduction path 11 is s11＝30*0.97+20*1＝31.6+17.4＝49.1。
The weight factor of each risk conduction path in the risk conduction set 2 is normalized, as shown in fig. 11, and the weight value of each risk conduction path can be obtained:
w10＝s10/(s10+s11)＝52.2/(52.2+49.1)＝0.516
w11＝s11/(s10+s11)＝49.1/(52.2+49.1)＝0.484
for the downstream risk score of company a, it is the sum of the initial risk score of company a and the risk of downstream propagation, i.e.:
r＝0+0*0.516+1*0.484＝0.484
with the downstream risk propagation calculation, in the case where the initial risk score for company a is 0, the downstream increased risk to company a can be found.
If the transaction relationship graph in fig. 8 is taken as a whole, i.e., the upstream and downstream risk conduction is comprehensively evaluated, as shown in fig. 12, the weight value of each risk conduction path can be obtained:
w7＝s7/(s7+s8+s9+s10+s11)＝0.245
w8＝s8/(s7+s8+s9+s10+s11)＝0.228
w9＝s9/(s7+s8+s9+s10+s11)＝0.172
w10＝s10/(s7+s8+s9+s10+s11)＝0.183
w11＝s11/(s7+s8+s9+s10+s11)＝0.172
from the perspective of upstream supply and from the perspective of downstream sales, although the graph structure and calculation method are the same, the weight values of the risk conduction paths are different, and therefore better indexes can be provided for upstream and downstream analysis of target entities.
In the above embodiment, a transaction relationship graph of enterprise operation is first constructed through enterprise invoice data, nodes in the graph are enterprises, edges are upstream and downstream relationships among the enterprises, initial risk scores of the enterprises are attributes of the nodes, and weight values are attributes of the edges. After the atlas is constructed, risk conduction results of the enterprises from the supply end and the sales end are respectively calculated by using a graph calculation mode of a pagerank algorithm.
In practical application, the risk identification method can be used in a financial institution which can acquire entity invoice data and master the basic operation conditions of upstream and downstream entities in the invoice, and the entity risk conduction is calculated by constructing an upstream and downstream transaction relation map of the entity, so that the entity operation conditions are better identified.
An embodiment of the present application further provides a risk identification device, as shown in fig. 13, the device may include:
a conducting path constructing module 1310, configured to acquire transaction data of a target entity, and construct a transaction relationship map of the target entity by using the target entity and each associated entity in the transaction data as nodes and using upstream and downstream relationships between the target entity and each associated entity as edges, where each edge in the transaction relationship map corresponds to one risk conducting path;
a path weight determining module 1320, configured to, for each risk conduction path, obtain each transaction between two nodes corresponding to the risk conduction path from the transaction data, calculate a time attenuation coefficient of each transaction, and determine a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction;
a risk identification module 1330 configured to determine a target risk score for the target entity based on the weighting factor for each of the risk conduction pathways.
In this embodiment, as shown in fig. 14, the path weight determining module 1320 may include an attenuation coefficient determining unit 1321 and a weight factor determining unit 1322, where the attenuation coefficient determining unit 1321 is configured to calculate a time attenuation coefficient of each transaction, and the weight factor determining unit 1322 is configured to determine a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction.
Specifically, the attenuation coefficient determining unit 1321 may include:
a time interval calculating unit 13211, configured to calculate a time interval between a transaction time corresponding to the transaction and a current time;
a function calculating unit 13212 configured to obtain a time decay coefficient of the transaction according to a pre-established time decay function according to the time interval, wherein the time decay function indicates a decay caused by the time interval.
Specifically, the weighting factor determining unit 1322 may include:
the transaction weight calculation unit 13221 is configured to determine, for each transaction, a product of a time decay coefficient of the transaction and a transaction attribute value corresponding to the transaction as a weight factor of the transaction, where the transaction attribute value includes a transaction amount or a transaction importance;
a weight factor superposition unit 13222, configured to superpose the weight factors of all the transactions, so as to obtain a weight factor of the risk conduction path.
In some embodiments, as shown in fig. 15, the apparatus may further include:
the initial risk identification module 1340 is configured to obtain an initial risk score of each node in the transaction relationship graph.
Accordingly, the risk identification module 1330 may include:
a first risk identification unit 1331, configured to determine a target risk score of the target entity based on the initial risk scores of the two nodes corresponding to each risk conduction path and the weight factors of the risk conduction paths.
In some embodiments, as shown in fig. 16, the apparatus may further include:
a risk conduction set determining module 1350, configured to determine risk conduction paths with the same directional attribute as a risk conduction set in the transaction relationship graph, where the directional attribute indicates that the associated entity corresponding to the risk conduction path is an upstream entity or a downstream entity.
Accordingly, the first risk identification unit 1331 may include:
a second risk identification unit 13311, configured to, for each risk conduction set, calculate a weight value of each risk conduction path according to the weight factor of each risk conduction path in the risk conduction set, and determine a target risk score of the target entity according to the initial risk scores of the two nodes corresponding to each risk conduction path and the weight values of the risk conduction paths.
In this embodiment of the application, as shown in fig. 17, the second risk identifying unit 13311 may include:
a weight value determination unit 133111, configured to calculate a weight value of each risk conduction path according to a weight factor of each risk conduction path in the risk conduction set;
a target risk determining unit 133112, configured to determine a target risk score of the target entity according to the initial risk scores of the two nodes corresponding to each risk conduction path and the weight values of the risk conduction paths.
Specifically, the weight value determination unit 133111 may include:
a set weight factor determination unit 1331111, configured to add the weight factors of the risk conduction paths in the risk conduction set to obtain a weight factor of the risk conduction set;
a normalization processing unit 1331112, configured to perform normalization processing on the weight factor of each risk conduction path based on the weight factor of the risk conduction set, so as to obtain a weight value of each risk conduction path.
Specifically, the target risk determining unit 133112 may include:
a risk conductive value determination unit 1331121, configured to calculate, for each risk conductive path, an initial risk score of the associated entity corresponding to the risk conductive path, and multiply the initial risk score by a weight value of the risk conductive path to obtain a risk conductive value of the risk conductive path;
a risk deviation value determining unit 1331122, configured to superimpose the risk conduction values of the risk conduction paths to obtain a risk deviation value of the target entity;
a risk score determining unit 1331123, configured to determine a sum of the initial risk score and the risk offset value of the target entity as a target risk score corresponding to the target entity.
It should be noted that the device and method embodiments in the device embodiment described above are based on the same inventive concept.
The embodiment of the present application further provides a risk identification device, where the device includes a processor and a memory, where the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded by the processor and executes the risk identification method provided in the above method embodiment.
Further, fig. 18 shows a hardware structure diagram of an apparatus for implementing the method provided in the embodiment of the present application, and the apparatus may participate in constituting or containing the device or system provided in the embodiment of the present application. As shown in fig. 18, the device 18 may include one or more (shown in the figures as 1802a, 1802b, … …, 1802 n) processors 1802 (the processors 1802 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), memory 1804 for storing data, and a transmission device 1806 for communication functions. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the I/O interface), a network interface, a power source, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 18 is merely an illustration and is not intended to limit the structure of the electronic device. For example, device 18 may also include more or fewer components than shown in FIG. 18, or have a different configuration than shown in FIG. 18.
It should be noted that the one or more processors 1802 and/or other data processing circuitry described above may be referred to generally herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part in software, hardware, firmware, or any combination thereof. Further, the data processing circuitry may be a single, stand-alone processing module, or incorporated in whole or in part into any of the other elements in the device 18 (or mobile device).
The memory 1804 may be used for storing software programs and modules of application software, such as program instructions/data storage devices corresponding to the methods described in the embodiments of the present application, and the processor 1802 executes various functional applications and data processing by executing the software programs and modules stored in the memory 1804, so as to implement one of the risk identification methods described above. The memory 1804 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 1804 may further include memory located remotely from the processor 1802 that may be connected to the device 18 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 1806 is used for receiving or sending data via a network. Specific examples of such networks may include wireless networks provided by the communication provider of the device 180. In one example, the transmission device 1806 includes a network adapter (NIC) that can be connected to other network devices through a base station so as to communicate with the internet. In one example, the transmission device 1806 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the device 18 (or mobile device).
The embodiment of the present application further provides a computer storage medium, where at least one instruction or at least one program is stored in the computer storage medium, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the risk identification method provided by the above method embodiment.
Alternatively, in this embodiment, the computer storage medium may be located on at least one of a plurality of network servers of a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
Embodiments of the present application also provide a computer program product or computer program comprising computer instructions stored in a computer storage medium. The computer instructions are read from the computer storage medium by a processor of the risk identification device, and the computer instructions are executed by the processor to cause the risk identification device to perform the steps in the method embodiments described above.
According to the technical scheme provided by the embodiment, the upstream and downstream relation between the target entity and each associated entity is determined as a risk conduction path, and the risk conduction of each associated entity to the target entity can be respectively identified; by introducing a time attenuation coefficient into each transaction, the weight factor of each risk conduction path is more reasonable, and the accuracy of risk identification based on the weight factor of each risk conduction path is higher; performing risk conduction calculation based on the initial risk score of each node, and identifying whether the target entity has a condition of increasing the operation risk; the risk conduction calculation is carried out on the upstream entity and the downstream entity of the target entity respectively, and the current operation risk of the target entity can be accurately identified to the upstream entity and/or the downstream entity.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the device and electronic apparatus embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference may be made to some descriptions of the method embodiments for relevant points.
The foregoing description has disclosed fully embodiments of the present application. It should be noted that those skilled in the art can make modifications to the embodiments of the present application without departing from the scope of the claims of the present application. Accordingly, the scope of the claims of the present application is not to be limited to the particular embodiments described above.
Claims (10)
1. A method for risk identification, the method comprising:
acquiring transaction data of a target entity, and constructing a transaction relationship map of the target entity by respectively taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, wherein each edge in the transaction relationship map corresponds to a risk conduction path;
for each risk conduction path, acquiring each transaction between two nodes corresponding to the risk conduction path from the transaction data, calculating a time attenuation coefficient of each transaction, and determining a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction;
determining a target risk score for the target entity based on the weight factor for each of the risk conductive pathways.
2. The method of claim 1, wherein said calculating a time decay factor for each of said transactions comprises:
calculating a time interval between the transaction time corresponding to the transaction and the current time;
and obtaining a time attenuation coefficient of the transaction according to the time interval and a pre-established time attenuation function, wherein the time attenuation function indicates the attenuation caused by the time interval.
3. The method according to claim 1, wherein said determining a weight factor for said risk conduction path as a function of a time decay factor for each of said transactions comprises:
for each transaction, determining the product of the time attenuation coefficient of the transaction and a transaction attribute value corresponding to the transaction as a weight factor of the transaction, wherein the transaction attribute value comprises a transaction amount or a transaction importance degree;
and superposing the weight factors of all the transactions to obtain the weight factor of the risk conduction path.
4. The method of claim 1, wherein prior to said determining a target risk score for said target entity based on a weight factor for each of said risk conductive pathways, said method further comprises:
acquiring an initial risk score of each node in the transaction relationship map;
accordingly, the determining a target risk score for the target entity based on the weight factor for each of the risk conductive pathways comprises:
determining a target risk score for the target entity based on the initial risk scores of the two nodes corresponding to each of the risk conduction paths and the weight factors of the risk conduction paths.
5. The method according to claim 4, wherein prior to said determining a target risk score for said target entity based on an initial risk score for two nodes corresponding to each of said risk conduction paths and a weight factor for said risk conduction path, said method further comprises:
determining risk conduction paths with the same direction attribute as a risk conduction set in the transaction relationship graph, wherein the direction attribute indicates that an associated entity corresponding to the risk conduction paths is an upstream entity or a downstream entity;
correspondingly, the determining a target risk score of the target entity based on the initial risk scores of the two nodes corresponding to each risk conduction path and the weighting factors of the risk conduction paths comprises:
for each risk conduction set, calculating a weight value of each risk conduction path according to the weight factor of each risk conduction path in the risk conduction set, and determining a target risk score of the target entity according to the initial risk scores of the two nodes corresponding to each risk conduction path and the weight values of the risk conduction paths.
6. The method according to claim 5, wherein said calculating a weight value for each said risk conduction path according to a weight factor for each said risk conduction path in said risk conduction set comprises:
adding the weight factors of all the risk conduction paths in the risk conduction set to obtain the weight factor of the risk conduction set;
and based on the weight factors of the risk conduction set, carrying out normalization processing on the weight factors of each risk conduction path to obtain the weight value of each risk conduction path.
7. The method according to claim 5, wherein determining a target risk score for the target entity based on the initial risk scores of the two nodes corresponding to each risk conduction path and the weight values of the risk conduction paths comprises:
for each risk conduction path, calculating an initial risk score of the associated entity corresponding to the risk conduction path, and multiplying the initial risk score by the weight value of the risk conduction path to obtain a risk conduction value of the risk conduction path;
superposing the risk conduction values of the risk conduction paths to obtain a risk offset value of the target entity;
and determining the sum of the initial risk score and the risk offset value of the target entity as a target risk score corresponding to the target entity.
8. A risk identification device, the device comprising:
the transaction relationship graph comprises a conducting path construction module, a risk conducting path construction module and a risk conducting path construction module, wherein the conducting path construction module is used for acquiring transaction data of a target entity, and constructing a transaction relationship graph of the target entity by taking the target entity and each associated entity in the transaction data as nodes and taking an upstream-downstream relationship between the target entity and each associated entity as an edge, and each edge in the transaction relationship graph corresponds to one risk conducting path;
a path weight determination module, configured to, for each risk conduction path, obtain, from the transaction data, each transaction between two nodes corresponding to the risk conduction path, calculate a time attenuation coefficient of each transaction, and determine a weight factor of the risk conduction path according to the time attenuation coefficient of each transaction;
a risk identification module to determine a target risk score for the target entity based on the weighting factor for each of the risk conductive pathways.
9. A risk identification device, characterized in that it comprises a processor and a memory, in which at least one instruction or at least one program is stored, which is loaded by the processor and executes the risk identification method according to any of claims 1-7.
10. A computer storage medium having stored therein at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by a processor to implement the risk identification method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011118242.2A CN114386727A (en) | 2020-10-19 | 2020-10-19 | Risk identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011118242.2A CN114386727A (en) | 2020-10-19 | 2020-10-19 | Risk identification method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114386727A true CN114386727A (en) | 2022-04-22 |
Family
ID=81193713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011118242.2A Pending CN114386727A (en) | 2020-10-19 | 2020-10-19 | Risk identification method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114386727A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925217A (en) * | 2022-05-24 | 2022-08-19 | 中国电子科技集团公司第十研究所 | High-value path discovery method based on relational attribute weighting |
| CN115034520A (en) * | 2022-08-09 | 2022-09-09 | 太平金融科技服务(上海)有限公司深圳分公司 | Risk prediction method, device, equipment and storage medium |
| CN115170304A (en) * | 2022-06-22 | 2022-10-11 | 支付宝(杭州)信息技术有限公司 | Method and device for extracting risk feature description |
| CN115292424A (en) * | 2022-10-08 | 2022-11-04 | 凯美瑞德（苏州）信息科技股份有限公司 | Risk conduction analysis method, electronic equipment and storage medium |
| CN115660688A (en) * | 2022-10-24 | 2023-01-31 | 西南财经大学 | Financial transaction abnormity detection method and cross-region sustainable training method thereof |
-
2020
- 2020-10-19 CN CN202011118242.2A patent/CN114386727A/en active Pending
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114925217A (en) * | 2022-05-24 | 2022-08-19 | 中国电子科技集团公司第十研究所 | High-value path discovery method based on relational attribute weighting |
| CN114925217B (en) * | 2022-05-24 | 2023-05-02 | 中国电子科技集团公司第十研究所 | High-value path discovery method based on relation attribute weighting |
| CN115170304A (en) * | 2022-06-22 | 2022-10-11 | 支付宝(杭州)信息技术有限公司 | Method and device for extracting risk feature description |
| CN115034520A (en) * | 2022-08-09 | 2022-09-09 | 太平金融科技服务(上海)有限公司深圳分公司 | Risk prediction method, device, equipment and storage medium |
| CN115034520B (en) * | 2022-08-09 | 2023-01-10 | 太平金融科技服务(上海)有限公司深圳分公司 | Risk prediction method, device, equipment and storage medium |
| CN115292424A (en) * | 2022-10-08 | 2022-11-04 | 凯美瑞德（苏州）信息科技股份有限公司 | Risk conduction analysis method, electronic equipment and storage medium |
| CN115660688A (en) * | 2022-10-24 | 2023-01-31 | 西南财经大学 | Financial transaction abnormity detection method and cross-region sustainable training method thereof |
| CN115660688B (en) * | 2022-10-24 | 2024-04-30 | 西南财经大学 | Financial transaction anomaly detection method and cross-regional sustainable training method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114386727A (en) | Risk identification method, device, equipment and storage medium | |
| Podobnik et al. | Cross-correlations between volume change and price change | |
| Andersen et al. | Primal-dual simulation algorithm for pricing multidimensional American options | |
| US9020988B2 (en) | Database aggregation of purchase data | |
| CA2930158C (en) | Data conversion and distribution systems | |
| Coyle et al. | Cloud computing and national accounting | |
| Tsai | The cause and outcomes of the ripple effect: housing prices and transaction volume | |
| Ergün et al. | Effect of financial constraints on the growth of family and nonfamily firms in Turkey | |
| Zaigham et al. | Causal relation between stock market performance and firm investment in China: Mediating role of information asymmetry | |
| Maldonado et al. | No‐arbitrage matrices of exchange rates: Some characterizations | |
| Abendin et al. | Bilateral Trade in West Africa: Does Digitalization Matter? | |
| CN105303447A (en) | Method and device for carrying out credit rating through network information | |
| KR101927578B1 (en) | System for providing enterprise information and method | |
| Ho et al. | A network analysis of the cryptocurrency market | |
| Fiordelisi et al. | Efficiency in the factoring industry | |
| Chen et al. | Mutual fund performance evaluation–application of system BCC model | |
| CN112540959A (en) | Data processing method and device | |
| Syamsuar et al. | Empirical Review Of M-Banking User Satisfaction Using End User Computing Satisfaction (EUCS) | |
| Prior et al. | Costs, revenues and performance in Spanish banking: a comparative analysis of pre-and early crisis years | |
| Beare | Optimal measure preserving derivatives revisited | |
| Choe et al. | The k th default time distribution and basket default swap pricing | |
| EP2770473A1 (en) | Systems and methods for detecting market irregularities | |
| CN115357729A (en) | Method and device for constructing securities relation map and electronic equipment | |
| Mukhibad et al. | Equity-based financing and risk in Islamic banks: A cross-country analysis | |
| Denis | Approximate hedging of contingent claims under transaction costs for general pay-offs |
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 |