CN113553444A - Audit knowledge graph representation model based on excess edges and associated reasoning method - Google Patents
Audit knowledge graph representation model based on excess edges and associated reasoning method Download PDFInfo
- Publication number
- CN113553444A CN113553444A CN202110849600.5A CN202110849600A CN113553444A CN 113553444 A CN113553444 A CN 113553444A CN 202110849600 A CN202110849600 A CN 202110849600A CN 113553444 A CN113553444 A CN 113553444A
- Authority
- CN
- China
- Prior art keywords
- audit
- association
- super
- knowledge graph
- edge
- 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
- 238000012550 audit Methods 0.000 title claims abstract description 158
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000013598 vector Substances 0.000 claims description 16
- 238000009472 formulation Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 6
- 238000012552 review Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000006399 behavior Effects 0.000 abstract description 2
- 238000011160 research Methods 0.000 description 14
- 238000007726 management method Methods 0.000 description 5
- 238000010276 construction Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
Images
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Animal Behavior & Ethology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an audit knowledge graph representation model based on a super edge and an associated reasoning method. Firstly, constructing an audit knowledge graph representation model based on a transfinite, and describing key entities related to audit and relations thereof from the aspects of project type, audit analysis, audit specification and behavior data; in order to accurately represent the logical relationship in the audit point in the atlas, a super edge And a super node (including the four types of If, And, Or And Basic) are introduced on the basis of the triple. Then, a graph representation learning method is introduced on the model, and three inference methods of entity semantic association (including clause-clause association, field-field association and audit point-field association) are provided. The invention guides the audit analysis process by means of the established model and method, enhances the understanding of auditors to the standard when doubts occur, carries out association recommendation on the next audit direction, and changes the manual audit into intelligent audit, thereby greatly improving the audit efficiency.
Description
Technical Field
The invention belongs to the field of artificial intelligence audit, and particularly relates to an audit knowledge graph representation model based on a super edge and an associated reasoning method.
Background
Audit is an independent economic supervision activity for examining and evaluating the authenticity, legality and profitability of the finance, financial income and other economic activities of an audited unit by full-time institutions and personnel, and the core of the audit is compliance inspection. In recent years, with the rise of big data and artificial intelligence, the audit is gradually changed from artificial audit to intelligent audit, and the method specifically comprises the following three parts: the efficiency of the auditing work can be greatly improved through intelligent auditing.
In the auditing process, auditing norms (such as regulations and rules, legal terms and the like) are an important auditing basis, but the norms mainly exist in the brains of auditors or files are temporarily consulted, so that knowledge points are scattered, reasonable organization is lacked, key points are easy to miss, and meanwhile, the association contrast analysis capability of association behaviors is limited, so that the problem is solved by constructing an auditing knowledge map, a machine can understand and utilize the auditing norms, discover the association between auditing norms and effectively extract and update new knowledge, and auditing intellectualization is realized.
However, the following five problems are common in the existing audit knowledge maps: timeliness (the specification will be updated continuously, and the expired specification will be invalid); reliability (specifications from different sources should have different confidence levels); complex dependencies between multiple entities (simple triplet form cannot be represented, possibly consisting of primary relationships and constraints); standardizing flow knowledge (consisting of a plurality of links, and an order relationship exists among the links); associated reasoning cannot be realized in combination with specific audit services.
Disclosure of Invention
The invention aims to provide an audit knowledge graph representation model and an associated reasoning method based on a super-edge aiming at the defects of the existing audit knowledge graph.
In order to realize the purpose, the invention is realized by the following technical scheme: a kind of audit knowledge map representation model based on the excess edge, including concept, entity and its relation type; the concepts include items involved in auditing, audit analysis, specifications, and data; the items comprise a total item type, a primary item type and a secondary item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association; connecting items, specifications and data in series by taking the secondary audit points as axes;
the invention provides a correlation reasoning method of an audit knowledge graph based on a super edge, which specifically comprises the following steps:
(1) constructing an audit knowledge map representation model: constructing concepts, entities and relationship types thereof, and constructing triples by using the concepts and the relationship types thereof; the triples form an audit knowledge map representation model;
(2) constructing an audit knowledge graph representation model based on the excess edge: introducing a super edge and a super node on the audit knowledge graph representation model constructed in the step (1); the super edge is an edge connecting more than two nodes; the super nodes are introduced for representing super edges, And the super node types comprise If, And, Or And Basic;
(3) performing associated reasoning by using the ultralimit-based audit knowledge graph representation model obtained in the step (2): embedding an audit knowledge graph by using a graph representation learning model to obtain a vector of each entity and relationship type, and obtaining a semantic association reasoning result by using cosine similarity between every two vectors, wherein the semantic association reasoning result comprises a term-term, a field-field and a review point-field;
further, the step (1) includes the sub-steps of:
(1.1) constructing concepts, entities and relationship types thereof: the concepts include items involved in auditing, audit analysis, specifications, and data; the items comprise a total item type, a primary item type and a secondary item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association; connecting items, specifications and data in series by taking the secondary audit points as axes;
(1.2) constructing a triple by using the concept entity and the relationship type thereof: the first two parameters in the triple are selected from concepts related to auditing, and the last parameter is selected from a relationship type to obtain a triple; the triples form an audit knowledge map representation model;
the audit knowledge graph representation model includes, but is not limited to, the following triples: document _ creation unit _ association, secondary audit point _ clause _ association, clause _ document _ inclusion, menu _ system _ inclusion, primary project type _ secondary audit point _ association, field _ presence, field _ table association, historical clause _ historical version, field _ menu _ inclusion, secondary historical audit point _ historical clause _ association, historical document _ historical version, secondary audit point _ primary audit point _ belonging, secondary historical audit point _ secondary audit point _ historical version, historical clause _ historical document _ inclusion, primary project type _ secondary historical audit point _ association, primary audit point _ audit point type _ belonging, secondary project type _ secondary audit point _ association, secondary project type _ primary project type _ belonging, supernode _ superedge association, historical document _ historical item _ historical version, historical item _ historical version _ historical item _ historical version, historical item _ historical version _ historical item _ historical version _ historical item _ historical _ association, historical item _ historical item _ historical _ association, historical _, Supernode _ field _ superedge association, and secondary audit point _ supernode _ superedge association.
Further, characterized in that said step (3) comprises the following sub-steps:
(3.1) arranging the audit knowledge graph based on the excess edges into a triple form as an input of the graph representation learning model;
(3.2) numbering each entity and relationship type in the triplet output in the step (3.1), wherein the numbering starts from 0;
(3.3) utilizing the graph to represent learning model embedding to obtain a vector of each entity and the relationship type;
(3.4) calculating cosine similarity between every two vectors, sequencing the cosine similarity from high to low, taking the first N cosine similarity values as a semantic association reasoning result, wherein N is a self-defined numerical value, and the semantic association reasoning result comprises a term-term, a field-field and a check point-field;
compared with the prior art, the invention has the beneficial effects that: the invention provides an overlimit-based audit knowledge graph representation model, which respectively takes the specified validity period and reliability as attributes to be placed in an audit knowledge graph entity, thereby solving the problems of timeliness and reliability of the existing audit knowledge graph; aiming at the problems of complex dependency relationship and normative process knowledge among a plurality of entities, the invention constructs an audit knowledge map based on excess edges, and integrates audit rules with concept entities contained in the audit knowledge map and the relationship among the concept entities; the invention can be combined with specific audit service, applies the graph representation learning method to the proposed audit knowledge map based on the excess edge, and solves the problem that the prior audit system can not realize associated reasoning and association recommendation in combination with the specific audit service.
Drawings
FIG. 1 is a schematic diagram of a super-edge based audit knowledge graph framework of the present invention;
FIG. 2 is a partial atlas with secondary audit points as axes.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings and specific examples in order to enable those skilled in the art to better understand the invention, but the invention is not limited thereto.
The invention discloses an audit knowledge graph representation model based on a super edge, as shown in FIG. 1, the structure of the audit knowledge graph representation model comprises concepts, entities and relationship types thereof; the concepts include items involved in auditing, audit analysis, specifications, and data; the items comprise a total item type, a primary item type and a secondary item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association;
the items comprise a total item type, a primary item type and a secondary item type, and the total item type comprises all the items; the primary project type comprises a self-setting project, a transverse project and a longitudinal project; the secondary project types comprise a youth fund project, a major project, an open topic and a scientific research platform construction and equipment development project.
The audit point types comprise four categories of data integrity audit, task book audit, scientific research data audit and financial data audit; the first-level audit points are corresponding contents under four types of audit points, the data integrity audit comprises project item establishing materials, project process management materials and project acceptance materials, the task book audit comprises project expense budget, project personnel and project duration, the scientific research data audit comprises expert argumentation opinion, project achievements, project expense approval, project important item change and project state management, and the financial data audit comprises project expense execution conditions, purchase expenses, equipment expenses, research experiment expenses, labor expenses, expert consultation expenses, conference expenses, travel expenses, management expenses, incentive expenses and basic construction expenses; the secondary audit points are further refined of the primary audit points;
the terms include XX units of internal regulation and specific terms of audit-related laws and regulations; the document comprises XX units of internal regulation and related laws and regulations for auditing; the preparation unit is a preparation unit corresponding to the document;
the field is a specific field corresponding to the secondary audit point; the menu is a top-level directory of fields; the system is a system corresponding to the menu, such as a scientific research system, a financial system and the like;
the invention discloses an audit knowledge graph representation model based on a super edge and an associated reasoning method, which take the field of auditing of specific scientific research projects as an example, and specifically comprises the following steps:
(1) establishing an audit knowledge graph representation model, and describing key entities involved in audit and relationship types thereof, wherein the model mainly comprises concepts, entities and relationship types thereof; the concepts include items involved in auditing, audit analysis, specifications, and data; the item comprises an item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association; the first two parameters in the triple are selected from concepts related to auditing, and the last parameter is selected from a relationship type to obtain a triple; the triples form an audit knowledge map representation model;
connecting items, specifications and data in series by taking the specific secondary check point as an axis; fig. 2 is a local map with secondary audit points as axes, wherein the series connection mode of the secondary audit points specifically includes:
A. item (primary item type): the audit point belongs to the audit range of the self-setting project;
B. and (4) checking points: when the contents of the second-level audit points are that the project cannot be completed according to time, whether a project principal applies for processing a postponed procedure in advance for a certain threshold time or not, the first-level audit points are project important item changes, and the type of the audit points is scientific research data audit;
C. and (3) specification: terms associated with the secondary review point: the term of the item is changed. When the project cannot be completed according to the schedule, the project responsible person should apply for a delay procedure in advance for a certain threshold time, and the project responsible person is approved by the scientific research development department after approval of the project responsible person in the research center. In principle, the extension period of the first application cannot exceed one fourth of the original period of the project, and the extension period of the second application cannot exceed one sixth of the original period. The project that cannot be completed after the extension of two periods is terminated in principle. (ii) a
D. Data: the data fields corresponding to the secondary audit points are 3: project end time, project original end time and submission deadline change application time.
(2) Constructing an audit knowledge graph representation model based on the excess edge: introducing a super edge and a super node on the audit knowledge graph representation model constructed in the step (1); the super edge is an edge connecting more than two nodes; the super node is introduced for representing super edges And comprises If, And, Or And Basic; the super edge is an edge connecting more than two nodes;
for example, some secondary audit points are: when the project cannot be completed according to the schedule, whether a project principal applies for processing a delay procedure in advance for a certain threshold time or not is judged, wherein the delay procedure comprises two types of super nodes including If and Basic, and at the moment, the logical relationship of the secondary audit points (audit rules) is shown on an audit knowledge graph based on super edges: if the project end time is greater than the intended end date (If-1-Basic), then the filing date change application time should be one month (If-2-Basic) before the intended end time of the project. Executing the If in sequence from small to large according to the value of the order attribute, specifically executing the If-1-Basic, and executing the If-2-Basic If the item ending time is larger than the preset ending date; specific rules are written in the super edge and super node Basic.
For example, some secondary audit points are: whether the expenses of the condition construction project And the scientific research starting project are outdialized or not, If the expenses are outdialized or whether the principal And the task are approved by the official business office, the two types of the supernodes including If And, at the moment, the logical relationship of the second-level audit point (audit rule) is expressed on an audit knowledge graph based on the superedge: and If the (If-1) outbound expense is more than 0(If-1-And-1) And the project type is a condition construction project or a scientific research starting project (If-1-And-2), checking whether the result of the approval of the trade And the office of the outbound expense owner is available (If-2). The method comprises the steps that the If execution sequence is executed from small to large according to values of order attributes, specifically, If-1 is executed first, then a local graph with a super node type of an And is executed, If-1-an-1 is executed first, then If-1-an-2 is executed, then If-2 is executed; the specific rules are written in the properties of the super edge.
For example, some secondary audit points are: the direct charge adjustment amount exceeds 10% of the subject approval budget amount, Or is more than 50 ten thousand yuan but not more than 200 ten thousand yuan, whether the direct charge adjustment amount is applied by a project principal, and after the direct charge adjustment amount is audited by a research center principal, the direct charge adjustment amount reports approval of a scientific research development department, wherein the approval includes two types of supernodes, namely If and Or, and at the moment, the logical relationship of the secondary audit point (audit rule) is expressed on an audit knowledge graph based on a superedge: if the (If-1) direct charge adjustment amount exceeds the subject approval budget amount by 10% (If-1-Or-1) Or the (If-1-Or-2) direct charge adjustment amount is greater than 50 ten thousand yuan but not greater than 200 ten thousand yuan, it is checked whether there is a corresponding approval result (If-2). Executing the If in sequence from small to large according to the value of the order attribute, specifically executing If-1 and then executing a local graph with a super node type of Or, Or executing If-1-Or-1 and then executing If-1-Or-2, and then executing If-2 If either one of the two is true; specific rules are written in the attributes of the super edges and the super nodes Or.
(3) Performing associated reasoning by using the ultralimit-based audit knowledge graph representation model obtained in the step (2): embedding an audit knowledge graph by using a graph representation learning model to obtain a vector of each entity and relationship type, and obtaining a semantic association reasoning result by using cosine similarity between every two vectors, wherein the semantic association reasoning result comprises a term-term, a field-field and a review point-field; the method comprises the following steps:
(3.1) arranging the audit knowledge graph based on the excess edges into a triple form as an input of the graph representation learning model;
(3.2) numbering each entity and relationship type in the triplet output in the step (3.1), wherein the numbering starts from 0;
(3.3) utilizing the graph to represent learning model embedding to obtain a vector of each entity and the relationship type;
(3.4) calculating cosine similarity between every two vectors, sequencing the cosine similarity from high to low, taking the first N cosine similarity values as a semantic association reasoning result, wherein N is a self-defined numerical value, and the semantic association reasoning result comprises a term-term, a field-field and a check point-field;
the method is introduced to the specific scientific research project auditing field to obtain the semantically related reasoning result, including clause-clause, field-field and auditing point-field.
The audit point-field is specifically as follows:
(a) the auditing point is used for judging whether scientific research data auditing-published written information conforms to the following regulation or not: work x term. The cosine similarity value of the corresponding vector of the first 5 audit points-fields is used as the reasoning result of semantic association of the audit points, the reasoning result of semantic association is the number of writings extracted from the task book text, the number of copyright items of the project actual application software, and the like, and the detailed results are shown in the following table 1:
TABLE 1
(B) The check point is used for judging whether the task book exists or not: the reasoning results of the cosine similarity value semantic association of the corresponding vectors of the first 5 audit points-fields are a project task book, a subject task book, a project establishment notification book and the like. The detailed results are shown in table 2 below:
TABLE 2
The clauses-clauses are specifically: the clause content is that project management specialists regularly collect project progress conditions at each stage of a project, evaluate project problem risks, form project stage progress reports and report the project stage progress reports to scientific research development departments. The related terms are used for supervising, checking and evaluating the execution condition of the project task at the completion time and the project middle period appointed by each stage task and establishing a monitoring and evaluating mechanism. The terms and associated terms respectively describe project process management from different angles and have certain relevance. The detailed results are shown in table 3 below:
TABLE 3
The field-field is specifically:
(a) where the fields are the entry start time: cosine similarity values of vectors corresponding to the first 5 fields are the title, the name of a person in charge, the certificate type, the certificate number and the information of the person in charge of the project, and all relate to basic information of the project. The detailed results are shown in table 4 below:
TABLE 4
(b) Where the fields are item IDs: the cosine similarity values of the vectors corresponding to the first 5 field-fields are the project name, the title, the name of a person in charge, the certificate type and the belonging center, and all relate to the basic information of the project. The detailed results are shown in table 5 below:
TABLE 5
The above embodiments are illustrative of the principles of the present invention and its efficacy, but are not intended to limit the scope of the invention. Any person skilled in the art can modify and change the embodiments without departing from the technical principle and spirit of the present invention. The protection scope of the present invention shall be subject to the claims.
Claims (4)
1. A superadjacent-based audit knowledge graph representation model is characterized by comprising concepts, entities and relation types thereof; the concepts include items involved in auditing, audit analysis, specifications, and data; the items comprise a total item type, a primary item type and a secondary item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association; and connecting the project, the specification and the data in series by taking the secondary audit point as an axis.
2. The correlation reasoning method based on the over-edge audit knowledge graph of the application claim 1 is characterized by comprising the following steps:
(1) constructing an audit knowledge map representation model: constructing concepts, entities and relationship types thereof, and constructing triples by using the entities and the relationship types thereof; the triples form an audit knowledge map representation model;
(2) constructing an audit knowledge graph representation model based on the excess edge: introducing a super edge and a super node on the audit knowledge graph representation model constructed in the step (1); the super edge is an edge connecting more than two nodes; the super nodes are introduced for representing super edges, And the super node types comprise If, And, Or And Basic;
(3) performing associated reasoning by using the ultralimit-based audit knowledge graph representation model obtained in the step (2): and embedding an audit knowledge map by using a graph representation learning model to obtain a vector of each entity and relationship type, and obtaining a semantic association reasoning result by using cosine similarity between every two vectors, wherein the semantic association reasoning result comprises a term-term, a field-field and a review point-field.
3. The correlation inference method of ultra-edge based audit knowledge maps according to claim 2, wherein the step (1) comprises the following sub-steps:
(1.1) constructing concepts, entities and relationship types thereof: the concepts include items involved in auditing, audit analysis, specifications, and data; the items comprise a total item type, a primary item type and a secondary item type; the audit analysis comprises audit point types, primary audit points and secondary audit points; the specification includes terms, documents and a formulation unit; the data comprises fields, menus and systems; connecting items, specifications and data in series by taking the secondary audit points as axes; the entity is a concrete representation under each class of concept; the relationship type comprises association, presence, inclusion, historical version, belonging, table association and super edge association;
(1.2) constructing a triple by using the entities and the relationship types thereof: the first two parameters in the triple are selected from concepts related to auditing, and the last parameter is selected from a relationship type to obtain a triple; the triples form an audit knowledge graph representation model.
4. The correlation inference method of ultra-edge based audit knowledge maps of claim 2, wherein the step (3) comprises the following sub-steps:
(3.1) arranging the audit knowledge graph based on the excess edges into a triple form as an input of the graph representation learning model;
(3.2) numbering each entity and relationship type in the triplet output in the step (3.1), wherein the numbering starts from 0;
and (3.3) utilizing the graph to represent learning model embedding to obtain a vector of each entity and relationship type.
And (3.4) calculating cosine similarity between every two vectors, sequencing the cosine similarity from high to low, taking the first N cosine similarity values as a semantic association reasoning result, wherein N is a self-defined numerical value, and the semantic association reasoning result comprises a term-term, a field-field and an audit point-field.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110849600.5A CN113553444A (en) | 2021-07-27 | 2021-07-27 | Audit knowledge graph representation model based on excess edges and associated reasoning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110849600.5A CN113553444A (en) | 2021-07-27 | 2021-07-27 | Audit knowledge graph representation model based on excess edges and associated reasoning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113553444A true CN113553444A (en) | 2021-10-26 |
Family
ID=78132904
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110849600.5A Pending CN113553444A (en) | 2021-07-27 | 2021-07-27 | Audit knowledge graph representation model based on excess edges and associated reasoning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113553444A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114117066A (en) * | 2021-11-12 | 2022-03-01 | 中国银行股份有限公司 | Recommendation method and device for audit retrieval data folder |
| CN114780619A (en) * | 2022-06-07 | 2022-07-22 | 国网浙江省电力有限公司金华供电公司 | Abnormity early warning method for automatic engineering audit data |
| CN116204875A (en) * | 2023-05-06 | 2023-06-02 | 天津南大通用数据技术股份有限公司 | Method for realizing reasoning capacity of database audit record |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110137919A1 (en) * | 2009-12-09 | 2011-06-09 | Electronics And Telecommunications Research Institute | Apparatus and method for knowledge graph stabilization |
| US20110276362A1 (en) * | 2010-05-05 | 2011-11-10 | Oracle International Corporation | Auditing client - service provider relationships with reference to internal controls assessments |
| CN106874261A (en) * | 2017-03-17 | 2017-06-20 | 中国科学院软件研究所 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
| CN107015963A (en) * | 2017-03-22 | 2017-08-04 | 重庆邮电大学 | Natural language semantic parsing system and method based on deep neural network |
| CN107609052A (en) * | 2017-08-23 | 2018-01-19 | 中国科学院软件研究所 | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle |
| CN110111337A (en) * | 2019-04-16 | 2019-08-09 | 中山大学 | A kind of general human body analytical framework and its analytic method based on figure transfer learning |
| CN111104522A (en) * | 2019-12-20 | 2020-05-05 | 武汉理工大学 | Regional industry association effect trend prediction method based on knowledge graph |
| CN111191047A (en) * | 2019-12-31 | 2020-05-22 | 武汉理工大学 | Knowledge graph construction method for human-computer cooperation disassembly task |
| CN112131882A (en) * | 2020-09-30 | 2020-12-25 | 绿盟科技集团股份有限公司 | Multi-source heterogeneous network security knowledge graph construction method and device |
| CN112732944A (en) * | 2021-01-30 | 2021-04-30 | 吉林大学 | New method for text retrieval |
| CN112765486A (en) * | 2021-01-22 | 2021-05-07 | 重庆邮电大学 | Knowledge graph fused attention mechanism movie recommendation method |
| CN113157922A (en) * | 2021-04-28 | 2021-07-23 | 江苏易安联网络技术有限公司 | Network entity behavior evaluation and visualization method based on graph |
-
2021
- 2021-07-27 CN CN202110849600.5A patent/CN113553444A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110137919A1 (en) * | 2009-12-09 | 2011-06-09 | Electronics And Telecommunications Research Institute | Apparatus and method for knowledge graph stabilization |
| US20110276362A1 (en) * | 2010-05-05 | 2011-11-10 | Oracle International Corporation | Auditing client - service provider relationships with reference to internal controls assessments |
| CN106874261A (en) * | 2017-03-17 | 2017-06-20 | 中国科学院软件研究所 | A kind of domain knowledge collection of illustrative plates and querying method based on semantic triangle |
| CN107015963A (en) * | 2017-03-22 | 2017-08-04 | 重庆邮电大学 | Natural language semantic parsing system and method based on deep neural network |
| CN107609052A (en) * | 2017-08-23 | 2018-01-19 | 中国科学院软件研究所 | A kind of generation method and device of the domain knowledge collection of illustrative plates based on semantic triangle |
| CN110111337A (en) * | 2019-04-16 | 2019-08-09 | 中山大学 | A kind of general human body analytical framework and its analytic method based on figure transfer learning |
| CN111104522A (en) * | 2019-12-20 | 2020-05-05 | 武汉理工大学 | Regional industry association effect trend prediction method based on knowledge graph |
| CN111191047A (en) * | 2019-12-31 | 2020-05-22 | 武汉理工大学 | Knowledge graph construction method for human-computer cooperation disassembly task |
| CN112131882A (en) * | 2020-09-30 | 2020-12-25 | 绿盟科技集团股份有限公司 | Multi-source heterogeneous network security knowledge graph construction method and device |
| CN112765486A (en) * | 2021-01-22 | 2021-05-07 | 重庆邮电大学 | Knowledge graph fused attention mechanism movie recommendation method |
| CN112732944A (en) * | 2021-01-30 | 2021-04-30 | 吉林大学 | New method for text retrieval |
| CN113157922A (en) * | 2021-04-28 | 2021-07-23 | 江苏易安联网络技术有限公司 | Network entity behavior evaluation and visualization method based on graph |
Non-Patent Citations (2)
| Title |
|---|
| 刘胜航;陈辉;朱嘉奇;武汇岳;熊金泉;王宏安;: "基于语义三角形的自然人机交互模型", 中国科学:信息科学, no. 04, 20 April 2018 (2018-04-20) * |
| 王亚珅，黄河燕: "短文本表示建模及应用", 31 May 2021, 北京理工大学出版社, pages: 32 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114117066A (en) * | 2021-11-12 | 2022-03-01 | 中国银行股份有限公司 | Recommendation method and device for audit retrieval data folder |
| CN114780619A (en) * | 2022-06-07 | 2022-07-22 | 国网浙江省电力有限公司金华供电公司 | Abnormity early warning method for automatic engineering audit data |
| CN116204875A (en) * | 2023-05-06 | 2023-06-02 | 天津南大通用数据技术股份有限公司 | Method for realizing reasoning capacity of database audit record |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113553444A (en) | Audit knowledge graph representation model based on excess edges and associated reasoning method | |
| Zhang et al. | Factors influencing BIM adoption for construction enterprises in China | |
| CN111061679B (en) | Method and system for rapid configuration of technological innovation policy based on rete and drools rules | |
| Montoni et al. | Applying grounded theory to understand software process improvement implementation | |
| CN110851576A (en) | Question and answer processing method, device, equipment and readable medium | |
| Browne et al. | Distributed data and ontologies: An integrated semantic web architecture enabling more efficient data management | |
| CN114238655A (en) | Enterprise association relation identification method, device, equipment and medium | |
| Cummaudo et al. | What should I document? A preliminary systematic mapping study into API documentation knowledge | |
| Getman et al. | Ontologies as a Set to Describe Legal Information. | |
| CN116595191A (en) | Construction method and device of interactive low-code knowledge graph | |
| Natt och Dag et al. | An experiment on linguistic tool support for consolidation of requirements from multiple sources in market-driven product development | |
| CN114385820A (en) | Enterprise-level audit knowledge map construction method | |
| CN111611387A (en) | Method and device for consulting civil case, computer equipment and storage medium | |
| Gil | Interactive knowledge capture in the new millennium: how the Semantic Web changed everything | |
| Carter et al. | Desperately Seeking Information in Information Systems Research. | |
| Ping | An empirical study of collection assessment based on the quality–utility–value theory | |
| Guo et al. | [Retracted] Analysis on Influence of Business Intelligence Information Quality over User Information Adoption Based on Multiple Mediating Effects | |
| CN113190582B (en) | Data real-time interactive mining flow modeling analysis system | |
| CN115309987A (en) | Enterprise information matching system | |
| Tolle et al. | User experience design of malang city public service information mobile application using human-centered design method | |
| Yang et al. | Function mechanism of psychological driving factors and professional transformation: migrant workers to industrial workers | |
| Mao et al. | The effect of organizational environment on engineering project cost management: A field study on CREC 2nd Bureau Co. Ltd | |
| Terrientes | Hierarchical outranking methods for multi-criteria decision aiding | |
| Montoni et al. | Using grounded theory to acquire knowledge about critical success factors for conducting software process improvement implementation initiatives | |
| Wadjdi et al. | The Implementation of Framing, Agenda-Setting, and Data Mining in Evaluation of Public Policies Case Study of the State Defense Program in Indonesia |
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 |