CN113988638A - Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium - Google Patents
Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium Download PDFInfo
- Publication number
- CN113988638A CN113988638A CN202111271844.6A CN202111271844A CN113988638A CN 113988638 A CN113988638 A CN 113988638A CN 202111271844 A CN202111271844 A CN 202111271844A CN 113988638 A CN113988638 A CN 113988638A
- Authority
- CN
- China
- Prior art keywords
- relationship
- relation
- strength
- paths
- nodes
- 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
- G06F16/30—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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/08—Auctions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- 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
- G06Q40/06—Asset management; Financial planning or analysis
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Technology Law (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Educational Administration (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method, a device, a medium and a terminal for measuring and calculating the strength of a general incidence relation, wherein the method comprises the following steps: acquiring target enterprise information, and acquiring all relationship paths among enterprises according to the target enterprise information; establishing a network model for measuring and calculating the relation strength of each relation path, training, carrying out data coding on all relation paths according to the relation between nodes and the length of the relation paths to form a coding matrix, uniquely mapping each coding rule into a score value through the coding matrix, and taking the relation strength with the highest score value as the relation strength of a target enterprise to complete the measurement and calculation of the incidence relation strength between the target enterprises; the method can efficiently obtain all relation paths among companies, avoids the problems of low efficiency, missed judgment, instability and the like caused by an expert identification mode, improves the intensity calculation problem in an ultra-multipath scene, theoretically has the time complexity of matrix operation as constant time, and has the operation time not influenced by the number of paths.
Description
Technical Field
The invention relates to the field of computer data processing, in particular to a method and a device for measuring and calculating strength of a general association relationship, electronic equipment and a medium.
Background
The incidence relation of the enterprise has great effects on enterprise risk transmission and enterprise pedigree identification, is beneficial to evaluating risks such as capital flow direction monitoring, is beneficial to excavating a risk subgraph distribution mode which is difficult to identify by manpower, and analyzing the conduction mode of the risks in a market relation network, for example, in the bidding activities of the enterprise, the entering enterprise needs to be checked in a cross bidding manner, the incidence relation among the entering enterprises is prevented, the fairness of the bidding activities is influenced, the interests of the enterprise are damaged, and the violation of relevant supervision and management regulations and even the violation of relevant laws are avoided.
At present, the existing solutions mainly depend on expert experience determination, that is, some industrial and commercial information of a company is obtained as much as possible according to information such as the name of the company, and a possible association relation is found by means of accumulated expert experience, but in actual work, the following disadvantages exist in the process of troubleshooting by depending on the expert experience: firstly, the possible association relationship between companies is complicated, and it is difficult to find out the actual relationship path. Second, the association relationship composition and path length conditions between companies are different, and the relationship strength level between companies cannot be determined for the found relationship path. And thirdly, the traditional mode depends on expert experience, information is collected manually for scouting, the operation efficiency is low, and the efficient development of bidding activities is influenced. Fourthly, the implicit relationship is difficult to find by depending on the method.
Disclosure of Invention
In view of the above-mentioned shortcomings of the prior art, the present invention provides a method, an apparatus, a medium and a terminal for measuring and calculating the strength of a general association relationship, so as to solve the above technical problems.
The invention provides a general incidence relation strength measuring and calculating method, which comprises the following steps:
acquiring target enterprise information, wherein the target enterprise information at least comprises: nodes and relationships between nodes;
acquiring all relation paths among enterprises according to the target enterprise information;
establishing a network model for measuring and calculating the relation strength of each relation path, and training the network model according to the relation among the nodes, the length of the relation path and a coding matrix, wherein the training comprises the steps of carrying out data coding on all the relation paths according to the relation among the nodes and the length of the relation path to form a coding matrix, and uniquely mapping each coding rule into a score value through the coding matrix;
and obtaining the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the calculation of the relation strength among the target enterprises.
In an embodiment of the present invention, the performing data encoding on all the relationship paths according to the relationship between the nodes and the lengths of the relationship paths to form an encoding matrix, and mapping each encoding rule uniquely as a score value through the encoding matrix includes:
acquiring the relationship among the nodes, the length of the relationship path and the corresponding relationship strength, and performing data coding on all the relationship paths through the one-hot code to acquire a coding matrix;
training a fully-connected neural network through the coding matrix to obtain a coefficient matrix and an intercept vector of the fully-connected neural network;
and (3) uniquely mapping each coding rule into a score value by using the trained fully-connected neural network, and calculating the score values of the relationship paths of all the target enterprises at one time.
In an embodiment of the present invention, the nodes include companies, legal persons, and high governments, and the relationships among the nodes include investment relationships, job relationships, legal persons relationships, and branch office relationships;
taking the relationship among the nodes as edges among the nodes to form a node table and a relationship table, and forming graph data according to the node table and the relationship table;
when the relation path of the target enterprise is inquired, positioning is carried out in the graph data through the target enterprise information, and then all relation paths between two nodes are obtained.
In an embodiment of the present invention, the score value is obtained by the following formula:
Scoregeneral assembly＝max(F{F[F(X·W[1]+b[1])·W[2]+b[2]]·W[3]+b[3]})
Wherein, W[1]、W[2]、W[3]Coefficient matrix 1, coefficient matrix 2, coefficient matrix 3, b[1]、b[2]、b[3]Respectively as an offset matrix 1, an offset matrix 2, an offset matrix 3 and a transformation function F;
and taking the relationship path with the highest score value as the strongest relationship strength between the target enterprises, and further finishing the determination of the final relationship strength between the two target enterprises.
In an embodiment of the present invention, after obtaining the relationship strength of the target enterprise, the method further includes:
acquiring an object to be identified, incidence relation data among the objects to be identified and incidence relation data between the object to be identified and an existing object;
determining the incidence relation and the incidence relation strength among the objects to be identified and the incidence relation strength between the objects to be identified and the existing objects according to the incidence relation data;
and carrying out risk identification on the object to be identified through the graph data.
In an embodiment of the present invention, after obtaining the relationship strength of the target enterprise, the method further includes:
establishing a corporate linkage map of the target enterprise according to the target enterprise information;
acquiring a public credit comprehensive evaluation score of a legal person of a target enterprise through the legal person association map;
setting corresponding weights for the relationship strength and risk identification of the target enterprise and the public credit comprehensive evaluation score of the legal person of the target enterprise in advance, further acquiring a final score, and using the final score as a reference basis for enterprise activities.
The invention also provides a device for measuring and calculating the strength of the general association relationship, which comprises:
the information acquisition module is used for acquiring target enterprise information, and the target enterprise information at least comprises: the relationship between nodes and each node;
the path query module is used for acquiring all relationship paths among enterprises according to the target enterprise information;
the network model is used for measuring and calculating the relation strength of each relation path;
the model training module is used for training the network model according to the relationship among the nodes, the length of the relationship path and the coding matrix, the training comprises the steps of carrying out data coding on all the relationship paths according to the relationship among the nodes and the length of the relationship path to form the coding matrix, and mapping each coding rule uniquely to be a score value through the coding matrix;
and the measuring and calculating module is used for acquiring the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the measurement and calculation of the relation strengths among the target enterprises.
In an embodiment of the present invention, the present invention further includes an encoding matrix module and a fully connected neural network;
acquiring the relationship among the nodes, the length of the relationship path and the corresponding relationship strength, and performing data coding on all the relationship paths by the coding matrix module through an one-hot code to acquire a coding matrix;
the coding matrix module trains a fully-connected neural network through the coding matrix to obtain a coefficient matrix and an intercept vector of the fully-connected neural network;
using the trained fully-connected neural network to uniquely map each coding rule into a score value, and calculating the score values of the relationship paths of all target enterprises at one time
The invention also provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the method according to any one of the preceding claims when executing the computer program.
The invention also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method as in any one of the above.
The invention has the beneficial effects that: the general incidence relation strength measuring and calculating method, the general incidence relation strength measuring and calculating device, the electronic equipment and the medium can efficiently obtain all relation paths between companies, avoid the problems of low efficiency, missing judgment, instability and the like caused by an expert identification mode, can meet the requirement of real-time path query by firstly finding out all relation paths between two companies, then measuring and calculating the relation strength of each path and taking the strongest strength as the relation strength between the two companies, and can meet the actual situation of graph data and the expectation.
In addition, the method and the device improve the intensity calculation problem in the ultra-multipath scene by utilizing the characteristic of efficient operation of matrix operation in a mode based on full-connection neural network mapping, and the operation time is not influenced by the number of paths. Based on the acquired relationship strength, the overall effects of the enterprise on the work such as bidding and the like are further improved through risk assessment and enterprise credit comprehensive evaluation, and the benefits of the company are prevented from being damaged.
Drawings
Fig. 1 is a schematic flow chart of a method for measuring and calculating strength of a general association relationship in an embodiment of the present invention.
Fig. 2 is a schematic flow chart of the training of the fully-connected network model of the general incidence relation strength measuring and calculating method in the embodiment of the present invention.
Fig. 3 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a hardware structure of an electronic device according to another embodiment of the present invention.
Fig. 5 is a schematic diagram of a hardware structure of the apparatus for measuring and calculating the strength of a general association relationship according to the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
In the following description, numerous details are set forth to provide a more thorough explanation of embodiments of the present invention, however, it will be apparent to one skilled in the art that embodiments of the present invention may be practiced without these specific details, and in other embodiments, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring embodiments of the present invention.
As shown in fig. 1, the method for measuring and calculating the strength of a general association relationship in this embodiment includes:
s1, obtaining target enterprise information, wherein the target enterprise information at least comprises: the relationship between nodes and each node;
s2, acquiring all relation paths among enterprises according to the target enterprise information;
s3, establishing a network model for measuring and calculating the relation strength of each relation path, and training the network model according to the relation among the nodes, the length of the relation path and a coding matrix, wherein the training comprises the steps of carrying out data coding on all the relation paths according to the relation among the nodes and the length of the relation path to form the coding matrix, and mapping each coding rule to a score value uniquely through the coding matrix;
and S4, obtaining the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the calculation of the relation strength among the target enterprises.
Before determining the strength of the relationship between companies in this embodiment, it is necessary to find a relationship path between companies, and target business information is first acquired through step S1 of this embodiment. The target enterprise information includes at least: the relationship between the nodes and each node is obtained through step S2 of this embodiment, and then all relationship paths between enterprises are obtained according to the target enterprise information. In this embodiment, the relationship path mainly includes a path formed by connecting the following relationships: 1. and (4) the investment relation is that the investment grade is divided according to the investment proportion. 2. The legal is divided into one level of the relationship of job and function. 3. High management, which is classified into different bases according to the level of the relationship (degree of importance). 4. And the branch office is regarded as the first level of investment. In order to quickly find out all relationship paths between companies in a specified relationship range (investment, legal, high-management and branch structure) and a specified relationship length (for example, a relationship of two degrees, such as A- > B- > C), the method can be realized by combining the characteristic that a graph database is good at finding the relationship paths by adopting a mode based on graph data. Specifically, the method comprises the following steps:
s101: and regarding companies, legal persons and high managers as nodes, regarding investment relations, legal person relations, high manager role relations and branch structure relations as edges among the nodes, and processing a corresponding node table and a corresponding relation table. The investment relation can cover the whole enterprise and commercial institutions, integrates stockholder information and external investment information, and processes the information into triples; the job relationship can cover the information of the high management and job of the whole quantity of the industrial and commercial institutions and is processed into a triplet; the legal relation can cover the whole enterprise and business legal enterprises and is processed into triples; the branch relationship can cover the branch information of the industrial and commercial enterprises and is processed into a triple;
s102: the node tables and relational tables are imported into the database, for example, using the neo4j database, to construct graph data.
S103: when searching all paths between companies, locating to a specific node according to information such as a company name in the target enterprise information acquired in step S1, and then querying all relationship paths between two nodes by using graph data characteristics, for example, all acyclic paths between two nodes can be queried by a path query method based on neo4j extension package, such as allsimplepaths (), and all shortest paths between two nodes can be queried by allshortpaths ().
In step S3 in this embodiment, a network model for measuring the relationship strength of each relationship path is established, and the network model is trained according to the relationship between the nodes, the length of the relationship path, and the coding matrix. After all the relationship paths are obtained through steps S1-S2, the relationship strength of each path needs to be measured, for example, the target enterprises are two companies, company a and company B, and can be linked with the investment relationship between companies through high management relationship. Company B < - > independent board < - > C < -100.00< -D < -100.00- > E < -100.00- > A < - >. The traditional mode can arrange rules of relationship strength levels under different relationship types and depths according to actual experience of business. For example, a combination of the investment relation 1 st gear and the occupational relation 2 nd gear is regarded as a strong relation when the path length is 2 degrees, a medium relation when the path length is 3 to 5 degrees, and a weak relation when the path length exceeds 5 degrees. According to the strength judgment criterion, the investment relation grade is one grade, the occupational relation grade is one grade, the path length is 5, and the relation strength is judged to be strong. On the basis of obtaining the relationship path, the conventional rule-based judgment method needs to sequentially judge the relationship composition type and the length of the relationship path of each path, and then obtain the strength level of the path. When a large number of routes between two companies are found (when the number of relation routes between two companies actually exceeds 10 ten thousand), the above method based on condition judgment is inefficient and difficult to deal with a real-time service scenario. In the embodiment, the strength calculation problem in a super-multipath scene is improved by adopting a full-connection neural network mapping-based mode and utilizing the characteristic of efficient operation of matrix operation. As shown in fig. 2, specifically:
s201: and the relationship composition and length and the corresponding relationship strength one-hot coding matrix are used for training the fully-connected neural network.
S202: and training the fully-connected neural network by using the coding matrix to obtain a coefficient matrix and an intercept vector of the network.
S203: and on the basis of inquiring to obtain the path, carrying out the same one-hot coding matrix on the path. Using the fully connected neural network model, the scores (strengths) of all paths are calculated at one time, and the highest score is obtained.
TABLE 1
As shown in table 1 above, in step S201, a first-order relationship is invested in a path having a path length of 1 degree, and the relationship strength is strong. The last one is: investment relation 2 shelves, and the job relation 2 shelves, and path length is the path of 3 degrees, and the relation intensity is weak. The score and intensity correspondence is: strong: 10 minutes, medium: 4 points, weak: 2 min, no: and 0 point. The level of the investment relation is divided according to the investment proportion, and the level of the job is divided according to the job ranking.
In step S203, after the path is found by the query, the same one-hot encoding matrix is performed on the path. Using the fully connected neural network model, the scores (strengths) of all paths are calculated at one time, and the highest score is obtained. For example, the matrix may be:
0 1 0 1 0 0 0 0 0 0 1
0 1 0 0 1 0 1 0 0 0 0
as above two codes, the following two paths are respectively represented:
1) a path with an investment level of 2, an occupational level of 1 and a path length of 6 degrees or more
2.) a path with an investment level of 2, an occupational level of 2, and a path length of 2 degrees.
In step S4 of this embodiment, the trained network model is used to obtain the score values of the relationship strengths of all paths between the target enterprises, and the relationship strength with the highest score value is used as the relationship strength of the target enterprise to complete the measurement and calculation of the relationship strength between the target enterprises, in this embodiment, the score values may be obtained by the following formula:
Scoregeneral assembly＝max(F{F[F(X·W[1]+b[1])·W[2]+b[2]]·W[3]+b[3]})
Wherein X is a coding matrix, W[1]、W[2]、W[3]Coefficient matrix 1, coefficient matrix 2, coefficient matrix 3, b[1]、b[2]、b[3]Respectively offset matrix 1, offset matrix 2, offset matrix 3, F is the transformation function, Relu. The dimension of the coefficient matrix 1 is 11 x 20, the dimension of the coefficient matrix 2 is 20 x 10, the dimension of the coefficient matrix 1 is 10 x 1, the dimension of the shift matrix 1 is 20 x 1, the dimension of the shift matrix 1 is 10 x 1, the dimension of the shift matrix 1 is n x 1, wherein,
[(n*11) (11*20) (20*10) (10*1)]＝n*1
the highest score is the strongest strength, and the strongest strength is taken as the relationship strength between the two companies, so that the final relationship strength between the two companies is identified.
In this embodiment, the fully-connected neural network is different from the prediction function in the conventional application, and the fully-connected neural network in this embodiment trains a neural network capable of uniquely mapping scores to each rule by using the data encoded by the intensity rule one-hot, and does not have an explicit prediction process when in use, but directly calculates the score vector by using the coefficient matrix and the coefficient matrix of the network, and fully utilizes the high efficiency of matrix operation. The fully-connected neural network (DNN) used in this embodiment is the least-sophisticated neural network, with the most network parameters and the most computation. Optionally, in this embodiment, the output value of each neuron may be calculated in the forward direction; then calculating the error term of each neuron in a reverse way; and finally, iteratively updating the weights w and b by using a random gradient descent algorithm to find the most suitable weight w and weight b, so that the whole loss function is as small as possible. The embodiment implements path query based on neo4j, implements path strength calculation based on a neural network, and can quickly and accurately find all paths between companies (nodes) that meet conditions (relationship types and path lengths). When a large number of paths are found (for example, hundreds of thousands of paths), the strength of all the paths can be efficiently calculated by the method in the embodiment. In the actual working process, the final result response is in millisecond level, and the real-time service requirement can be met. In actual activity application, the method takes the case of the girdling bidding, generally determines that the girdling bidding is suspected if the correlation result is strong and moderate, can effectively guarantee fair implementation of bidding work, and greatly maintains the benefits of companies.
In this embodiment, one-hot encoding is used for the discrete features, where N states are encoded by using an N-bit state register, each state is represented by its own independent register bit, and only one of the states is valid at any time, which makes the distance calculation between the features more reasonable. Since most algorithms are computed based on metrics in vector space, in order for the variable values of the non-partial order relationship to have no partial order and be equidistant to the dots. And (3) using one-hot coding to expand the value of the discrete feature to an Euclidean space, wherein a certain value of the discrete feature corresponds to a certain point of the Euclidean space. The one-hot coding of the discrete features can make the distance calculation between the features more reasonable. After one-hot encoding is carried out on the discrete features, the features of each dimension can be regarded as continuous features after encoding, and then normalization is carried out on the features of each dimension. Such as normalized to [ -1,1] or normalized to a mean of 0 and a variance of 1.
After step S4 of the present embodiment, based on the determination of the final relationship strength between the two companies, the method further includes:
and S5, identifying whether the object to be identified is an object with risk.
In this embodiment, after the risk value of the object marked with the risk value in the existing object is known, the risk value of the object to be recognized is predicted according to the determined association relationship and association relationship strength between the objects to be recognized and the existing object, and the association relationship between the objects is fully utilized in a manner of performing risk recognition on the objects to be recognized according to the determined association relationship and association relationship strength between the objects to be recognized and the existing object, so that the recognition rate of the objects with risks in the risk recognition is improved. Specifically, the method comprises the following steps:
s501, acquiring incidence relation data of objects to be identified, wherein the incidence relation data comprises incidence relation data between the objects to be identified and existing objects;
s502, determining the incidence relation and the incidence relation strength between the objects to be identified and the existing objects according to the incidence relation data;
s503, in the graph data, performing risk identification on the object to be identified according to the risk value of the existing object
In step S501, objects to be recognized are first acquired, and the objects to be recognized differ according to specific application fields. Taking the financial field as an example, the object to be identified can be a person or a business handling loan business at a financial institution; or enterprises needing to be enclosed in bidding activities.
In step S502, obtaining association relationship data between objects to be identified and association relationship data between the objects to be identified and existing objects, where the existing objects are objects recorded with graph data in the constructed graph database, and mainly include objects with marked risk values and objects without marked risk values, and the graph data of each object includes: the association relationship between the object and other objects and the strength of the association relationship. The number of the objects to be recognized in this embodiment may be one or multiple, and may be determined by a specific application scenario. Those skilled in the art will appreciate that the existing objects in the present embodiment may vary according to the specific application field.
In step S503, the association relationship and the association relationship strength between the objects to be identified and the existing objects are determined according to the association relationship data. In step S504, based on the determined association relationship, the strength of the association relationship, and the map data recorded in the constructed map database, it is identified whether the object to be identified is an object at risk. And determining the risk value of the object to be identified according to the risk of the existing object having the incidence relation with the object to be identified, the incidence relation and the incidence relation strength between the objects to be identified, the incidence relation and the incidence relation strength between the object to be identified and the existing object. After the risk values of the object to be identified and the risk values of the object without the marked risk values are obtained, the predicted risk values can be sorted according to the order of the risk values from large to small, after the risk value sorting result is obtained, a certain number of risk values can be selected according to a specific application scene, for example, the first 20% risk values in the sorting result can be selected, and then whether the risk values of the object to be identified are in the first 20% risk values or not is judged, if the risk values are in the first 20% risk values, the object to be identified is the object with the risk, and if the risk values are not in the first 20% risk values, the object to be identified is the object without the risk.
After step S4 of the present embodiment, based on the determination of the final relationship strength between the two companies, the method further includes:
and S6, identifying whether the object to be identified is an object with risk. Specifically, the method comprises the following steps:
s601, establishing a corporate linkage map of the target enterprise according to the target enterprise information; and establishing a target enterprise legal person relation chain set, wherein the enterprise legal person relation chain set comprises a plurality of enterprise legal persons with generalized relation with the target enterprise legal person.
S602, acquiring a public credit comprehensive evaluation score of a legal person of a target enterprise through a legal person association map; for example, the current comprehensive evaluation score value of the public credit of each enterprise legal person in the target enterprise legal person relationship chain set can be determined according to the previous iteration comprehensive evaluation score of the public credit of each enterprise legal person in the target enterprise legal person relationship chain set until the current comprehensive evaluation score value of the public credit of each enterprise legal person in the target enterprise legal person relationship chain set meets the preset convergence condition generalized association relationship strength, and the determination can be determined according to the share ratio between two enterprise legal persons, the similarity of high-level managers, the similarity of high-level members of a board party, the investment relationship, the loan relationship and the like, and the generalized association relationship strength can be set manually to be smaller than 1.
And S603, setting corresponding weights for the relationship strength and the risk identification of the target enterprise and the public credit comprehensive evaluation score of the legal person of the target enterprise in advance, further acquiring a final score, and using the final score as a reference basis for enterprise activities. For example, the current score value for the target business jurisdictions may be determined as the public credit integrated ratings score value for the target business jurisdictions. Calculating an initial comprehensive credit evaluation score vector; the target enterprise legal person in the target enterprise legal person relationship chain set the public credit comprehensive evaluation score value to be 0, other enterprise legal persons can obtain an initial score value according to the existing comprehensive credit evaluation score calculation method, and the comprehensive credit score values of the target enterprise legal person and other enterprise legal persons form an initial comprehensive credit evaluation score vector. For each enterprise legal person in the target enterprise legal person relationship chain set, determining the current public credit comprehensive evaluation score value of the target enterprise legal person according to the previous round of public credit comprehensive evaluation score value of the enterprise legal person having a generalized association relationship with the target enterprise legal person until the public credit comprehensive evaluation score of each enterprise legal person in the target enterprise legal person relationship chain set meets a preset convergence condition; determining the current public credit integrated evaluation score of the target enterprise legal person as the public credit integrated evaluation score of the target enterprise legal person.
In this embodiment, comprehensive consideration is performed in bidding activities of enterprises based on the approval of the final strength of the relationship between two companies and the comprehensive evaluation of corporate legal public credit. For example, taking the bidding scenario applied in the above embodiments as an example, in order to avoid affecting fairness of bidding activities and damaging interests of companies, relevant regulatory regulations are violated. On one hand, the evaluation can be carried out according to the acquired association strength through measuring and calculating the strength of the general association relationship, on the basis, the credit problem possibly occurring in the bidding activity can be further avoided through the combined judgment of two dimensions in combination with the public credit comprehensive evaluation of enterprise legal system personnel, and therefore the damage to the benefits of companies is effectively avoided. After the final relationship strength between the two companies and the comprehensive evaluation of the public credit of the enterprise legal person are obtained, the final score can be obtained in a weight setting mode, and the enterprises participating in bidding are effectively screened according to the final score. In this embodiment, the weight calculation method may be selected in combination with the characteristic condition of the data, for example, if the fluctuation between the data is an information amount, the CRITIC weight method or the information amount weight method may be considered; for the expert to score data, an AHP hierarchy method or a priority graph method may be used, which may be implemented by using a scheme mature in the prior art, and those skilled in the art should know that the description is omitted here. In the embodiment, multiple fields such as insurance, medical treatment, health and the like are fused to realize flexible configuration according to different business requirements.
Correspondingly, as shown in fig. 5, the present embodiment further provides a device for measuring and calculating the strength of a general association relationship, including:
the information acquisition module is used for acquiring target enterprise information, and the target enterprise information at least comprises: the relationship between nodes and each node;
the path query module is used for acquiring all relationship paths among enterprises according to the target enterprise information;
the network model is used for measuring and calculating the relation strength of each relation path;
the model training module is used for training the network model according to the relationship among the nodes, the length of the relationship path and the coding matrix, the training comprises the steps of carrying out data coding on all the relationship paths according to the relationship among the nodes and the length of the relationship path to form the coding matrix, and mapping each coding rule uniquely to be a score value through the coding matrix;
and the measuring and calculating module is used for acquiring the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the measurement and calculation of the relation strengths among the target enterprises.
The system also comprises a coding matrix module and a full-connection neural network;
acquiring the relationship among the nodes, the length of the relationship path and the corresponding relationship strength, and performing data coding on all the relationship paths by the coding matrix module through an one-hot code to acquire a coding matrix;
the coding matrix module trains a fully-connected neural network through the coding matrix to obtain a coefficient matrix and an intercept vector of the fully-connected neural network;
and mapping each coding rule uniquely into a score value through the trained fully-connected neural network, and calculating the score values of the relationship paths of all the target enterprises at one time.
In this embodiment, before determining the strength of the relationship between companies, it is necessary to find a relationship path between companies, and first obtain target business information. The target enterprise information includes at least: and acquiring all relationship paths among the enterprises according to the target enterprise information. In this embodiment, the relationship path mainly includes a path formed by connecting the following relationships: 1. and (4) the investment relation is that the investment grade is divided according to the investment proportion. 2. The legal is divided into one level of the relationship of job and function. 3. High management, which is classified into different bases according to the level of the relationship (degree of importance). 4. And the branch office is regarded as the first level of investment. In order to quickly find out all relationship paths between companies in a specified relationship range (investment, legal, high-management and branch structure) and a specified relationship length (for example, a relationship of two degrees, such as A- > B- > C), the method can be realized by combining the characteristic that a graph database is good at finding the relationship paths by adopting a mode based on graph data.
In this embodiment, a company, a legal person, and a high-level manager are regarded as nodes, and an investment relationship, a legal person relationship, a high-level manager relationship, and a branch structure relationship are regarded as edges between the nodes, so as to process a corresponding node table and a corresponding relationship table. The investment relation can cover the whole enterprise and commercial institutions, integrates stockholder information and external investment information, and processes the information into triples; the job relationship can cover the information of the high management and job of the whole quantity of the industrial and commercial institutions and is processed into a triplet; the legal relation can cover the whole enterprise and business legal enterprises and is processed into triples; the branch relationship can cover the branch information of the industrial and commercial enterprises and is processed into a triple. The node tables and relational tables are imported into the database, for example, using the neo4j database, to construct graph data. When searching all the paths between companies, the specific node is located according to the company name and other information in the target enterprise information acquired in step S1, and then all the relationship paths between the two nodes are queried using the graph data characteristics.
In this embodiment, a network model for measuring and calculating the relationship strength of each relationship path is established, and the network model is trained according to the relationship among the nodes, the length of the relationship path, and the coding matrix. After all the relationship paths are obtained, the relationship strength of each path needs to be measured, for example, the target enterprise is two companies, company a and company B, which can be linked by the high-management relationship and the investment relationship between the companies. Company B < - > independent board < - > C < -100.00< -D < -100.00- > E < -100.00- > A < - >. The traditional mode can arrange rules of relationship strength levels under different relationship types and depths according to actual experience of business. For example, a combination of the investment relation 1 st gear and the occupational relation 2 nd gear is regarded as a strong relation when the path length is 2 degrees, a medium relation when the path length is 3 to 5 degrees, and a weak relation when the path length exceeds 5 degrees. According to the strength judgment criterion, the investment relation grade is one grade, the occupational relation grade is one grade, the path length is 5, and the relation strength is judged to be strong. On the basis of obtaining the relationship path, the conventional rule-based judgment method needs to sequentially judge the relationship composition type and the length of the relationship path of each path, and then obtain the strength level of the path. When a large number of routes between two companies are found (when the number of relation routes between two companies actually exceeds 10 ten thousand), the above method based on condition judgment is inefficient and difficult to deal with a real-time service scenario. In the embodiment, the strength calculation problem in a super-multipath scene is improved by adopting a full-connection neural network mapping-based mode and utilizing the characteristic of efficient operation of matrix operation.
In this embodiment, the relationship composition and length, and the corresponding relationship strength one-hot encoding matrix are used to train the fully-connected neural network. And training the fully-connected neural network by using the coding matrix to obtain a coefficient matrix and an intercept vector of the network. And on the basis of inquiring to obtain the path, carrying out the same one-hot coding matrix on the path. Using the fully connected neural network model, the scores (strengths) of all paths are calculated at one time, and the highest score is obtained.
In this embodiment, the fully-connected neural network is different from the prediction function in the conventional application, and the fully-connected neural network in this embodiment trains a neural network capable of uniquely mapping scores to each rule by using the data encoded by the intensity rule one-hot, and does not have an explicit prediction process when in use, but directly calculates the score vector by using the coefficient matrix and the coefficient matrix of the network, and fully utilizes the high efficiency of matrix operation. The fully-connected neural network (DNN) used in this embodiment is the least-sophisticated neural network, with the most network parameters and the most computation. Optionally, in this embodiment, the output value of each neuron may be calculated in the forward direction; then calculating the error term of each neuron in a reverse way; and finally, iteratively updating the weights w and b by using a random gradient descent algorithm to find the most suitable weight w and weight b, so that the whole loss function is as small as possible. The embodiment implements path query based on neo4j, implements path strength calculation based on a neural network, and can quickly and accurately find all paths between companies (nodes) that meet conditions (relationship types and path lengths). When a large number of paths are found (for example, hundreds of thousands of paths), the strength of all the paths can be efficiently calculated by the method in the embodiment. In the actual working process, the final result response is in millisecond level, and the real-time service requirement can be met. In actual activity application, the method takes the case of the girdling bidding, generally determines that the girdling bidding is suspected if the correlation result is strong and moderate, can effectively guarantee fair implementation of bidding work, and greatly maintains the benefits of companies. In this embodiment, one-hot encoding is used for the discrete features, where N states are encoded by using an N-bit state register, each state is represented by its own independent register bit, and only one of the states is valid at any time, which makes the distance calculation between the features more reasonable. Since most algorithms are computed based on metrics in vector space, in order for the variable values of the non-partial order relationship to have no partial order and be equidistant to the dots. And (3) using one-hot coding to expand the value of the discrete feature to an Euclidean space, wherein a certain value of the discrete feature corresponds to a certain point of the Euclidean space. The one-hot coding of the discrete features can make the distance calculation between the features more reasonable. After one-hot encoding is carried out on the discrete features, the features of each dimension can be regarded as continuous features after encoding, and then normalization is carried out on the features of each dimension.
The present embodiment also provides a computer-readable storage medium on which a computer program is stored, which when executed by a processor implements any of the methods in the present embodiments.
The present embodiment further provides an electronic terminal, including: a processor and a memory;
the memory is used for storing computer programs, and the processor is used for executing the computer programs stored by the memory so as to enable the terminal to execute the method in the embodiment.
The computer-readable storage medium in the present embodiment can be understood by those skilled in the art as follows: all or part of the steps for implementing the above method embodiments may be performed by hardware associated with a computer program. The aforementioned computer program may be stored in a computer readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
The electronic device provided by the embodiment comprises a processor, a memory, a transceiver and a communication interface, wherein the memory and the communication interface are connected with the processor and the transceiver and are used for completing mutual communication, the memory is used for storing a computer program, the communication interface is used for carrying out communication, and the processor and the transceiver are used for operating the computer program so as to enable the electronic terminal to execute the steps of the method.
As shown in fig. 3, the electronic device may include: an input device 1100, a first processor 1101, an output device 1102, a first memory 1103, and at least one communication bus 1104. The communication bus 1104 is used to implement communication connections between the elements. The first memory 1103 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and the first memory 1103 may store various programs for performing various processing functions and implementing the method steps of the present embodiment.
Fig. 4 is a hardware structure of an electronic device provided in another embodiment, and the electronic device in this embodiment may include a second processor 1201 and a second memory 1202. The second processor 1201 executes the computer program code stored in the second memory 1202 to implement the method described in fig. 1 in the above embodiment. The second memory 1202 is configured to store various types of data to support operations at the terminal device. Examples of such data include instructions for any application or method operating on the terminal device, such as messages, pictures, videos, and so forth. The second memory 1202 may include a Random Access Memory (RAM) and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory. The electronic device may further include: communication components 1203, power components 1204, multimedia components 1205, audio components 1206, input/output interfaces 1207, and/or sensor components 1208. The specific components included in the terminal device are set according to actual requirements, which is not limited in this embodiment.
In this embodiment, the Memory may include a Random Access Memory (RAM), and may also include a non-volatile Memory (non-volatile Memory), such as at least one disk Memory.
The Processor may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component.
In the above embodiments, unless otherwise specified, the description of common objects by using "first", "second", etc. ordinal numbers only indicate that they refer to different instances of the same object, rather than indicating that the objects being described must be in a given sequence, whether temporally, spatially, in ranking, or in any other manner. In the above-described embodiments, reference in the specification to "the embodiment," "an embodiment," "another embodiment," or "other embodiments" means that a particular feature, structure, or characteristic described in connection with the embodiments is included in at least some embodiments, but not necessarily all embodiments. The various appearances of the phrase "the present embodiment," "one embodiment," or "another embodiment" are not necessarily all referring to the same embodiment. If the specification states a component, feature, structure, or characteristic "may", "might", or "could" be included, that particular component, feature, structure, or characteristic is not necessarily included. If the specification or claim refers to "a" or "an" element, that does not mean there is only one of the element. If the specification or claim refers to "a further" element, that does not preclude there being more than one of the further element.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The invention is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.
The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
In the embodiments described above, although the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art in light of the foregoing description. For example, other memory structures (e.g., dynamic ram (dram)) may use the discussed embodiments. The embodiments of the invention are intended to embrace all such alternatives, modifications and variances that fall within the broad scope of the appended claims.
The foregoing embodiments are merely illustrative of the principles of the present invention and its efficacy, and are not to be construed as limiting the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (10)
1. A method for measuring and calculating the strength of a general association relationship is characterized by comprising the following steps:
acquiring target enterprise information, wherein the target enterprise information at least comprises: nodes and relationships between nodes;
acquiring all relation paths among enterprises according to the target enterprise information;
establishing a network model for measuring and calculating the relation strength of each relation path, and training the network model according to the relation among the nodes, the length of the relation path and a coding matrix, wherein the training comprises the steps of carrying out data coding on all the relation paths according to the relation among the nodes and the length of the relation path to form a coding matrix, and uniquely mapping each coding rule into a score value through the coding matrix;
and obtaining the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the calculation of the relation strength among the target enterprises.
2. The method for measuring and calculating the strength of a general association relationship according to claim 1, wherein the step of performing data encoding on all the relationship paths according to the relationship between the nodes and the lengths of the relationship paths to form an encoding matrix, and uniquely mapping each encoding rule to a score value through the encoding matrix comprises:
acquiring the relationship among the nodes, the length of the relationship path and the corresponding relationship strength, and performing data coding on all the relationship paths through the one-hot code to acquire a coding matrix;
training a fully-connected neural network through the coding matrix;
and (3) uniquely mapping each coding rule into a score value by using the trained fully-connected neural network, and calculating the score values of the relationship paths of all the target enterprises at one time.
3. The method for measuring and calculating the strength of the general association relationship according to claim 2, wherein the nodes comprise companies, legal persons and high governments, and the relationships among the nodes comprise investment relationships, job-taking relationships, legal person relationships and branch office relationships;
taking the relationship between the nodes as the edges between the nodes to form a node table and a relationship table, and acquiring graph data corresponding to the node table and the relationship table according to the node table and the relationship table;
when the relation path of the target enterprise is inquired, positioning is carried out in the graph data through the target enterprise information, and then all relation paths between two nodes are obtained.
4. The method for measuring and calculating the strength of a general association relationship according to claim 3, wherein the score value is obtained by the following formula:
Scoregeneral assembly＝max(F{F[F(X·W[1]+b[1])·W[2]+b[2]]·W[3]+b[3]})
Wherein, W[1]、W[2]、W[3]Coefficient matrix 1, coefficient matrix 2, coefficient matrix 3, b[1]、b[2]、b[3]Respectively as an offset matrix 1, an offset matrix 2, an offset matrix 3 and a transformation function F;
and taking the relationship path with the highest score value as the strongest relationship strength between the target enterprises to finish the determination of the final relationship strength between the two target enterprises.
5. The method for measuring and calculating the general relationship strength as claimed in claim 3, further comprising, after obtaining the relationship strength of the target enterprise:
acquiring incidence relation data of objects to be identified, wherein the incidence relation data comprises incidence relation data between the objects to be identified and existing objects;
determining the incidence relation and the incidence relation strength between the objects to be identified and the existing objects according to the incidence relation data;
and in the graph data, performing risk identification on the object to be identified according to the risk value of the existing object.
6. The method for measuring and calculating the general relationship strength as claimed in claim 5, further comprising, after obtaining the relationship strength of the target enterprise:
establishing a corporate linkage map of the target enterprise according to the target enterprise information;
acquiring a public credit comprehensive evaluation score of a legal person of a target enterprise through the legal person association map;
setting weights for the relationship strength and risk identification of the target enterprise and the public credit comprehensive evaluation score of the legal person of the target enterprise in advance, and further acquiring a final score serving as an enterprise activity evaluation reference basis.
7. A general correlation strength measuring and calculating device is characterized by comprising:
the information acquisition module is used for acquiring target enterprise information, and the target enterprise information at least comprises: the relationship between nodes and each node;
the path query module is used for acquiring all relationship paths among enterprises according to the target enterprise information;
the network model is used for measuring and calculating the relation strength of each relation path;
the model training module is used for training the network model according to the relationship among the nodes, the length of the relationship path and the coding matrix, the training comprises the steps of carrying out data coding on all the relationship paths according to the relationship among the nodes and the length of the relationship path to form the coding matrix, and mapping each coding rule uniquely to be a score value through the coding matrix;
and the measuring and calculating module is used for acquiring the score values of the relation strengths of all paths among the target enterprises through the trained network model, and taking the relation strength with the highest score value as the relation strength of the target enterprises to complete the measurement and calculation of the relation strengths among the target enterprises.
8. The device for measuring and calculating the strength of general association relationship according to claim 7, further comprising a coding matrix module and a fully-connected neural network;
acquiring the relationship among the nodes, the length of the relationship path and the corresponding relationship strength, and performing data coding on all the relationship paths by the coding matrix module through an one-hot code to acquire a coding matrix;
the coding matrix module trains a fully-connected neural network through the coding matrix to obtain a coefficient matrix and an intercept vector of the fully-connected neural network;
and mapping each coding rule uniquely into a score value through the trained fully-connected neural network, and calculating the score values of the relationship paths of all the target enterprises at one time.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111271844.6A CN113988638A (en) | 2021-10-29 | 2021-10-29 | Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111271844.6A CN113988638A (en) | 2021-10-29 | 2021-10-29 | Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113988638A true CN113988638A (en) | 2022-01-28 |
Family
ID=79744366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111271844.6A Pending CN113988638A (en) | 2021-10-29 | 2021-10-29 | Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113988638A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115983921A (en) * | 2022-12-29 | 2023-04-18 | 广州市玄武无线科技股份有限公司 | Offline store commodity association combination method, device, equipment and storage medium |
| CN116226460A (en) * | 2022-12-09 | 2023-06-06 | 中科世通亨奇(北京)科技有限公司 | Method and equipment for extracting most valuable path based on character pattern |
-
2021
- 2021-10-29 CN CN202111271844.6A patent/CN113988638A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116226460A (en) * | 2022-12-09 | 2023-06-06 | 中科世通亨奇(北京)科技有限公司 | Method and equipment for extracting most valuable path based on character pattern |
| CN115983921A (en) * | 2022-12-29 | 2023-04-18 | 广州市玄武无线科技股份有限公司 | Offline store commodity association combination method, device, equipment and storage medium |
| CN115983921B (en) * | 2022-12-29 | 2023-11-14 | 广州市玄武无线科技股份有限公司 | Off-line store commodity association combination method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mukhametzyanov et al. | A sensitivity analysis in MCDM problems: A statistical approach | |
| Samoilenko et al. | Using Data Envelopment Analysis (DEA) for monitoring efficiency-based performance of productivity-driven organizations: Design and implementation of a decision support system | |
| CN108763277B (en) | Data analysis method, computer readable storage medium and terminal device | |
| CN109858740A (en) | Appraisal procedure, device, computer equipment and the storage medium of business risk | |
| Shekhovtsov et al. | Towards reliable decision-making in the green urban transport domain | |
| CN113988638A (en) | Method and device for measuring and calculating strength of general association relationship, electronic equipment and medium | |
| CN114565429B (en) | Intelligent provider matching method based on semantic graph model | |
| CN113656690B (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| US20140156544A1 (en) | Non-Tangible Assets Valuation Tool | |
| Volk et al. | Ask the Right Questions: Requirements Engineering for the Execution of Big Data Projects. | |
| CN109767333A (en) | Select based method, device, electronic equipment and computer readable storage medium | |
| Chen et al. | An intuitionstic fuzzy factorial analysis model for multi-attribute decision-making under random environment | |
| US20110213692A1 (en) | Computer-Implemented Systems And Methods For Automatic Triangulation Of Forward Conversion Curves | |
| CN113779116B (en) | Object ordering method, related equipment and medium | |
| Tchangani et al. | BOCR framework for decision analysis | |
| CN110490749A (en) | A kind of price fixing method and device | |
| Minnoor et al. | Nifty price prediction from Nifty SGX using machine learning, neural networks and sentiment analysis | |
| CN111506826A (en) | User recommendation method, device, equipment and storage medium based on intimacy | |
| CN112632197A (en) | Service relation processing method and device based on knowledge graph | |
| Ebrahimi | Efficiency measurement to identify the best efficient unit in the presence of dual-role factors | |
| CN110443646A (en) | Product competition relational network analysis method and system | |
| CN111126465B (en) | Node classification method and device, terminal equipment and computer readable storage medium | |
| CN113919957A (en) | Method, device, electronic equipment and medium for calculating cyclic investment proportion and representing path | |
| CN115345716B (en) | Method, system, medium and electronic device for estimating order fulfillment duration | |
| CN115422303A (en) | Intelligent relationship determination method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |