CN117540027A - Multi-element evidence association analysis system and method based on domain ontology - Google Patents
Multi-element evidence association analysis system and method based on domain ontology Download PDFInfo
- Publication number
- CN117540027A CN117540027A CN202311341606.7A CN202311341606A CN117540027A CN 117540027 A CN117540027 A CN 117540027A CN 202311341606 A CN202311341606 A CN 202311341606A CN 117540027 A CN117540027 A CN 117540027A
- Authority
- CN
- China
- Prior art keywords
- case
- evidence
- crime
- ontology
- examination
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000012098 association analyses Methods 0.000 title claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 49
- 238000012552 review Methods 0.000 claims abstract description 29
- 230000000007 visual effect Effects 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 238000010276 construction Methods 0.000 claims description 13
- 238000007726 management method Methods 0.000 claims description 12
- 238000003058 natural language processing Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 8
- 238000012097 association analysis method Methods 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 230000002787 reinforcement Effects 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 238000010835 comparative analysis Methods 0.000 claims description 2
- 238000004900 laundering Methods 0.000 description 4
- 238000010521 absorption reaction Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000000470 constituent Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 239000010931 gold Substances 0.000 description 2
- 229910052737 gold Inorganic materials 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 238000012550 audit Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Accounting & Taxation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a multi-element evidence association analysis system and method based on a domain ontology, and belongs to the field of computer data processing. The invention relates to a financial crime case evidence examination rule management module, which is used for storing and managing financial crime case evidence examination bodies with different crime names; the financial crime case data resource processing module is used for realizing element identification and extraction of case cards and volume data of case handling, writing the case element library, and when a case handling person carries out multi-element evidence examination and analysis, calling the financial crime case evidence examination rule management module, returning to a financial crime case evidence examination body of a specific crime name, reading a case element data set and sending the case element data set to the visual multi-element evidence examination and analysis module; and the visual multi-evidence review analysis module is used for viewing, comparing and analyzing the relevance and conflict relation of the evidence of various sources and various types through the visual interface. The invention can realize continuous expansion and intelligent analysis of the examination rules.
Description
Technical Field
The invention belongs to the field of computer data processing, and particularly relates to a multi-element evidence association analysis system and method based on a domain ontology.
Background
The financial crime cases are of a plurality of types, the crime amount is large, and the social hazard is serious. Evidence review and fact identification for financial crime cases require consideration of crime constituent elements and corresponding evidence systems for different crime names. The evidence of the financial crime case obtained by investigation mainly comprises word evidence, book evidence, electronic evidence and the like, and plays different roles respectively. The crime constitution required to be examined is different for a specific financial crime case, and the focused evidence type and specific evidence element are also different. This presents significant difficulties in computer-aided case evidence review.
The prior method for solving the problems mainly comprises the following steps: (1) By adopting a natural language processing method, important evidences are classified and identified through supervised learning, semantic similarity calculation is carried out on the important evidences and the evidences of the same type, and therefore whether a plurality of evidences support each other or conflict with each other is judged. The method is widely applied to evidence examination and analysis of criminal cases and most civil cases of various criminals, and has good adaptability. (2) According to the requirement of the case handling business, an evidence examination rule of a specific crime name is formulated, and the evidence is examined by developing special software according to a certain sequence. The method is accurate and effective for the cases of accurate and detailed crime names defined by the evidence examination rules.
Disadvantages of the prior art: the crime names of the financial crimes are numerous, each crime name is a type of financial business, the corresponding crime constituent elements are greatly different, and the evidence examination system and the emphasis point are greatly different. The evidence association analysis accuracy realized by adopting a universal semantic matching algorithm based on a natural language processing technology is not high, and the systematic multidimensional analysis is lacking; by adopting a method for hard-coding analysis of the data based on the examination rules, a large number of examination rules are required to be independently written for each crime name, the service is changed, the software code is required to be modified, the response is slow, and the cost is high.
Disclosure of Invention
First, the technical problem to be solved
The technical problem to be solved by the invention is how to provide a multi-element evidence association analysis system and method based on a domain ontology so as to solve the problems that the evidence association analysis accuracy realized by adopting a general semantic matching algorithm based on a natural language processing technology is not high and the systematic multidimensional analysis is lacking; by adopting a method for hard-coding analysis of the data based on the examination rules, a large number of examination rules need to be independently written for each crime name, the business is changed, the software code needs to be modified, and the problems of slow response and high cost are solved.
(II) technical scheme
In order to solve the technical problems, the invention provides a multi-element evidence association analysis system based on a domain ontology, which comprises: a financial crime case evidence examination rule management module, a financial crime case data resource processing module and a visual multi-element evidence examination analysis module;
the financial crime case evidence examination rule management module is used for storing and managing financial crime case evidence examination bodies with different crime names;
the financial crime case data resource processing module is used for realizing element identification and extraction of case cards and volume data of case handling, writing the case element library, and when a case handling person performs multi-element evidence examination and analysis, calling the financial crime case evidence examination rule management module, returning a financial crime case evidence examination body with a specific crime name, reading a case element data set of a certain case and sending the case element data set to the visual multi-element evidence examination and analysis module;
and the visual multi-element evidence examination and analysis module is used for enabling a case transacting person to view, compare and analyze the relevance and conflict relation of various sources and various types of evidence through a visual interface.
Further, ontology (Ontology) refers to a graph data structure based on Semantic Web (Semantic Web) technology, and is used for describing a logic structure of a knowledge graph, the Ontology has a General Ontology (General Ontology) and a Domain Ontology (Domain Ontology), the General Ontology is used for construction and application of a common sense knowledge base and a cross-Domain knowledge base, and the Domain Ontology is used for construction and application of a specific Domain professional knowledge structure; the basic unit of the body is a triplet: head nodes, relations and tail nodes, wherein the head nodes are concepts and entities; tail nodes are concepts and entities, or attribute values; the relation is a semantic relation between head nodes and tail nodes, and is definitely defined by an ontology; the head node and the tail node form nodes of the data structure of the ontology graph, and the relationship forms a directed edge of the data structure of the ontology graph, so that the ontology is a directed graph.
Further, the financial crime case evidence censoring ontology is a domain ontology, and for each financial crime name, a specific ontology is defined, and the ontologies of a plurality of crime names form a financial crime case evidence censoring ontology set; the evidence examination ontology set for the financial crime cases has a certain shared ontology part, and each different crime name has a special part of the ontology.
Further, the financial crime case evidence review rules include: the financial crime case evidence examination body, the evidence examination rule script and the evidence examination key point realize the definition of case facts of the financial crime case of a specific crime name and examination contents and methods of corresponding evidence.
Further, the financial crime case data resource processing module processes case card data and file data of the case, the case card data are structured data, the file data are unstructured data, and the general concepts of the financial crime case and the case feature concepts of specific crimes defined in the financial crime case evidence examination ontology are identified and extracted through entity identification of natural language processing and case feature element identification models based on deep learning, and are stored in a case element library in a structured format.
Further, the visual multi-element evidence examination analysis module organizes the case element data of a certain case by taking a financial crime case evidence examination body as a structure, and realizes a user interface, and the interface organizes the case element data from different dimensions and displays the case element data according to the requirement of a case transacting person on evidence examination of the case; the case handling personnel sequentially conduct comparison analysis on the same contents in the case elements according to the time sequence of inquiring the records; or gathering related case elements, and carrying out relevance analysis of different evidences; the evidence examination rules in the evidence examination rules of the financial crime cases are displayed on an interface, and a case transactor runs software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the evidence examination assistance is provided for the case transactor.
The invention also provides a multi-element evidence association analysis method based on the domain ontology, which comprises the following steps:
step 1: financial crime case evidence censoring ontology set and evidence censoring rule construction
The financial crime case evidence examination rule management module constructs a financial crime case evidence examination ontology set and evidence examination rules based on relevant law and judicial interpretation, various normative files of financial crime case examination and experience of a case transactor;
step 2: construction of feature element identification model of financial crime case
Labeling specific named entities in the case document based on a natural language processing entity recognition technology, constructing a batch of labeling data, and performing reinforcement learning by using a deep learning-based natural language processing model to obtain a financial crime case named entity recognition model; labeling case feature elements of specific crimes defined in a financial crime case evidence examination body, constructing a batch of labeling data, training the data by using a deep learning-based long text feature representation learning method, and constructing a case feature element identification model of the specific crimes;
step 3: case element identification and extraction of financial crime cases
The financial crime case data resource processing module invokes a case feature element recognition model of a specific crime name through recognition of the crime name information in the case card data, performs case feature element recognition and extraction on the text data of the case document of the financial crime in progress, and stores the case feature element recognition and extraction in a case element library; calling a financial crime case naming entity recognition model, recognizing and extracting general case elements of the text data of the in-process financial crime case files, and storing the general case elements into a case element library;
step 4: instantiating a financial crime case evidence review ontology for a particular crime name with case elements
When multi-element evidence association analysis is carried out, after a case name or a case number to be inspected is selected by a case handling person, a financial crime case evidence inspection body corresponding to the case name or the case number is automatically selected through crime name information in case card information, and general elements and characteristic elements of the case in a case element library are automatically filled in the body, so that instantiation of the body is realized;
step 5: visual multiple evidence association review analysis
The visual multi-element evidence examination analysis module calls the ontology data instantiated in the step 4 into a computer memory, and organizes and displays the case element data from different dimensions on an interface according to the requirement of a case handling person on evidence examination of the case; the evidence examination rules in the evidence examination rules of the financial crime cases are displayed on an interface, and a case transactor runs software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the evidence examination assistance is provided for the case transactor.
Further, the step 1 includes:
step 1.1: constructing a set of evidence review ontologies for financial crime cases
Aiming at the common points and different characteristics of case facts and evidences of various financial crime cases of specific crime names, constructing a financial crime case evidence examination body of each crime name; the general concepts and relations of the ontology are common to all financial crime cases, and are defined by constructing a financial crime case general ontology, and the four conditions are established with crime constitution: subject, object, subjective aspect and objective aspect association; the financial crime case evidence examination ontology of each crime name inherits the general ontology of the financial crime case, inherits the related concepts of the general ontology, defines new subclasses, and realizes the definition of case elements such as specific crime facts and evidences of the crime name;
step 1.2: constructing evidence review rules for financial crime cases
Financial crime case evidence review rules are also two broad categories, namely general rules and specific rules related to crime names; the evidence examination general rule is suitable for case evidence examination of different crimes, and the crime related examination rule aims at a case of a specific crime; evidence examination rules are formulated in a script mode, and through various judging methods of data comparison and similarity scoring, the evidence examination rules are used for prompting a case transactor and recorded in a corresponding examination result data set for later retrieval.
Further, in the step 4, the office manually selects the financial crime case evidence examination ontology of the specific crime name, and fills the case elements to implement ontology instantiation.
Further, in the step 5, the office staff sequentially performs comparative analysis on the same content in the case elements according to the time sequence of inquiring the records; or aggregating related case elements, and carrying out relevance analysis of different evidences.
(III) beneficial effects
The invention provides a multi-element evidence association analysis system and method based on a domain ontology, which has the technical effects that:
1) A set of multi-element evidence examination software tool can examine evidence systems of different crimes by switching domain bodies (including examination rules and elements of different crimes) of different crimes. Thereby realizing continuous expansion and intelligent analysis of the examination rules.
2) The domain ontology is a structure of data and rules that can be recognized and processed by a computer. The method can well utilize the technical advantages of graph data calculation and graph neural network, and find the direct incidence relation of the evidence faster and more accurately.
Drawings
FIG. 1 is a schematic diagram of a multi-element evidence association analysis system based on a domain ontology;
FIG. 2 is a flow chart of a method for multi-element evidence association analysis based on domain ontology.
Detailed Description
To make the objects, contents and advantages of the present invention more apparent, the following detailed description of the present invention will be given with reference to the accompanying drawings and examples.
In order to solve the problems, the invention provides a multi-element evidence association analysis method based on a financial crime case examination field ontology. The inventive structure of the present invention is shown in fig. 1:
the invention relates to a multi-element evidence association analysis system based on a domain ontology, which comprises the following components: a financial crime case evidence examination rule management module, a financial crime case data resource processing module and a visual multi-element evidence examination analysis module;
and the financial crime case evidence examination rule management module is used for storing and managing the financial crime case evidence examination ontology of different crime names.
And the financial crime case data resource processing module is used for realizing element identification and extraction of case cards and file data in case handling and writing the case elements into a case element library. When the case transacting personnel performs multi-element evidence examination and analysis, the financial crime case evidence examination rule management module is called, the financial crime case evidence examination body with a specific crime name is returned, the case element data set of a certain case is read, and the case element data set is sent to the visual multi-element evidence examination and analysis module.
And the visual multi-element evidence examination and analysis module is used for enabling a case transacting person to view, compare and analyze relations such as relevance, conflict and the like of various sources and various types of evidence through a visual interface.
The main concepts of the present invention are described below.
0. Body
Ontology refers to a graph data structure based on Semantic Web (Semantic Web) technology, and is used to describe the logical structure of a knowledge graph. The Ontology has a General Ontology (General Ontology) and a Domain Ontology (Domain Ontology), the General Ontology is used for construction and application of a common sense knowledge base and a cross-Domain knowledge base, and the Domain Ontology is used for construction and application of a specific Domain professional knowledge structure.
The basic unit of an ontology is a triplet (head node, relationship, tail node), also called (master, slave, guest). Wherein the head node is a concept and an entity (instantiation of the concept); tail nodes can be concepts and entities, and can also be attribute values (character strings or numerical values, which cannot be used as head nodes); relationships are semantic relationships between head nodes pointing to tail nodes, defined explicitly by the ontology. The head node and the tail node form nodes of the data structure of the ontology graph, and the relationship forms a directed edge (opposite direction is opposite semantic) of the data structure of the ontology graph, so that the ontology is a directed graph.
The triples of an ontology mainly define the relationship of concepts (also called classes) and concepts, for example: (Unit, subscossOf, principal), (suspect, typeOf, class), (fund electronic data evidence, subscossOf, evidence), (query entry, subscossOf, evidence); relationships are also defined, such as: (payroll, subtypeof, attribute), (payroll, subtypeof, crime facts), domain of payroll is "suspect", range is a numerical value.
After instantiation of an ontology, relationships of entities to concepts are defined, such as: (Lifour, typeOf, entity), (Zhang three, typeOf, suspect); relationships of entities to entities are also defined, for example: (Lifour, actual control bank account, some first account), (Zhang San, payroll case, 153,400 Yuan). Where typeOf and subClassOf, etc., relationships, class, entity, attribute, etc., are reserved words of the ontology.
1. Evidence review ontology for financial crime cases
The financial crime case evidence censoring ontology is an domain ontology, and for each financial crime's crime name, a specific ontology is defined, and the ontologies of a plurality of crime names all form a financial crime case evidence censoring ontology set. The evidence examination ontology set of the financial crime cases has a certain common ontology part, such as concepts of suspects, evidence and the like, and actually controls the relationship of bank accounts and the like. Each distinct crime name has its own specific part of the ontology, e.g. for illegally absorbing public deposit crimes, there are concepts and related attributes of funding pools, etc. For money laundering, there are concepts of self money laundering and other money laundering, and there are attributes of upstream crimes.
2. Evidence review rules for financial crime cases
The financial crime case evidence examination rules comprise contents such as a financial crime case evidence examination body, an evidence examination rule script, an evidence examination key point and the like, and the case facts of the financial crime case with a specific crime name and the examination contents and methods of corresponding evidence are defined.
3. Financial crime case data resource processing module
The financial crime case data resource processing module processes case card data (structured data) and file data (unstructured data, mainly text data) of a case, and recognizes and extracts general concepts (suspects, time to arrive, and the like) of the financial crime case and case feature concepts (fund pools, gold absorption scales, and the like) of a specific crime name defined in a financial crime case evidence examination ontology through entity recognition of natural language processing and a case feature element recognition model based on deep learning, and stores the general concepts and the case feature concepts in a structured format into a case element library.
4. Visual multi-element evidence examination and analysis module
The visual multi-element evidence examination analysis module can organize the case element data (including case facts, evidence and the like) of a certain case by taking a financial crime case evidence examination ontology as a structure. And realizing a user interface, and organizing and displaying the case element data from different dimensions according to the requirement of the case transactor on evidence examination of the case. The case handling personnel can sequentially compare and analyze the same contents in the case elements according to the time sequence of inquiring the records; the related case elements can be gathered, and the relevance analysis of different evidences can be performed. The evidence examination rules in the evidence examination rules of the financial crime cases are displayed on the interface, and a case transactor can run software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the assistance of evidence examination is provided for the case transactor.
The invention relates to a multi-element evidence association analysis method based on a domain ontology, which comprises the following steps:
step 1: financial crime case evidence censoring ontology set and evidence censoring rule construction
The financial crime case evidence examination rule management module constructs a financial crime case evidence examination ontology set, evidence examination rules and the like based on various normative files of financial crime case examination and experiences of case staff based on law and judicial interpretation such as criminal litigation law, people inspection institute criminal case handling rules and the like. The specific construction steps are as follows:
step 1.1: constructing a set of evidence review ontologies for financial crime cases
Aiming at specific crime names, such as common points and different characteristics of case facts and evidences of various financial crime cases such as illegal absorption of public deposit crimes, fund fraud crimes, money laundering crimes, loan fraud crimes, insurance fraud crimes and the like, a financial crime case evidence examination body of each crime name is constructed. The general concepts and relations of the ontology are common to all financial crime cases, and are defined by constructing a financial crime case general ontology, and an association relation of four elements (subject, object, subjective aspect and objective aspect) with crime is established. The financial crime case evidence examination ontology of each crime name inherits the general ontology of the financial crime case, inherits the related concepts of the general ontology, defines new subclasses, and realizes the definition of case elements such as specific crime facts and evidences of the crime name.
Step 1.2: constructing evidence review rules for financial crime cases
Financial crime case evidence review rules are also two broad categories of general rules and specific rules related to crime names. Wherein the evidence review generic rule is suitable for case evidence review of different crimes, such as: the fact of the same case is that the records are inconsistent through multiple interrogation; the fact of the same case is that a plurality of people inquire that the strokes are inconsistent; etc. The crime name related review rules are mainly directed to cases of a specific crime name, for example: the amount of the recruited funds for illegally absorbing public deposit crimes is a plurality of digital wisdom such as inquiry records, audit reports, and evidence examination and analysis results of electronic funds data, and comprehensive relevance analysis is required for the evidence. Evidence examination rules are formulated in a script mode, and can be used for prompting a case transactor through various judging methods of data comparison and similarity scoring, and are recorded in a corresponding examination result data set for later retrieval.
Step 2: construction of feature element identification model of financial crime case
The entity recognition technology based on natural language processing is used for marking specific named entities (such as suspects, criminal events, criminal amounts and the like) in case documents, constructing a batch of marking data, and performing reinforcement learning by using a natural language processing model based on deep learning to obtain a financial criminal case named entity recognition model. The method comprises the steps of marking case characteristic elements (case facts and evidences, such as fund pool scale, gold absorption scale, public propaganda condition of absorbed deposit, training experience, investor condition and the like) of specific crimes defined in a financial crime case evidence examination ontology, constructing a batch of marking data (one batch of data for each crime name), training the data by using a long text characteristic representation learning method based on deep learning, and constructing a case characteristic element identification model of the specific crimes.
Step 3: case element identification and extraction of financial crime cases
The financial crime case data resource processing module calls a case feature element identification model of a specific crime name through identifying the crime name information in the case card data (structure data), performs case feature element identification and extraction on the document text data of the financial crime case under office, and stores the document feature element identification and extraction in a case element library. And calling a financial crime case naming entity recognition model, recognizing and extracting general case elements (mainly suspicious names, crime amounts, finance obtained by crimes and the like) for the text data of the case files of the financial crimes in the office, and storing the general case elements in a case element library.
Step 4: instantiating a financial crime case evidence review ontology for a particular crime name with case elements
When multi-element evidence association analysis is carried out, after a case name or a case number to be inspected is selected by a case handling person, a financial crime case evidence inspection body corresponding to the case name or the case number is automatically selected through crime name information in case card information, and general elements and characteristic elements of the case in a case element library are automatically filled in the body, so that instantiation of the body is realized. The 'financial crime case evidence examination ontology' with specific crime name can also be manually selected by a case transactor, and case elements are filled, so that ontology instantiation is realized.
Step 5: visual multiple evidence association review analysis
The visual multi-element evidence examination and analysis module calls the ontology data instantiated in the step 4 into a computer memory, and organizes and displays the case element data from different dimensions on an interface according to the requirement of a case transactor for evidence examination of the case. The case handling personnel can sequentially compare and analyze the same contents in the case elements according to the time sequence of inquiring the records; the related case elements can be gathered, and the relevance analysis of different evidences can be performed. The evidence examination rules in the evidence examination rules of the financial crime cases are displayed on the interface, and a case transactor can run software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the assistance of evidence examination is provided for the case transactor.
The technical effects are as follows:
1) A set of multi-element evidence examination software tool can examine evidence systems of different crimes by switching domain bodies (including examination rules and elements of different crimes) of different crimes. Thereby reaching a constant expansion of the censoring rules and intelligent analysis.
2) The domain ontology is a structure of data and rules that can be recognized and processed by a computer. The method can well utilize the technical advantages of graph data calculation and graph neural network, and find the direct incidence relation of the evidence faster and more accurately.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (10)
1. A domain ontology-based multivariate evidence association analysis system, comprising: a financial crime case evidence examination rule management module, a financial crime case data resource processing module and a visual multi-element evidence examination analysis module;
the financial crime case evidence examination rule management module is used for storing and managing financial crime case evidence examination bodies with different crime names;
the financial crime case data resource processing module is used for realizing element identification and extraction of case cards and volume data of case handling, writing the case element library, and when a case handling person performs multi-element evidence examination and analysis, calling the financial crime case evidence examination rule management module, returning a financial crime case evidence examination body with a specific crime name, reading a case element data set of a certain case and sending the case element data set to the visual multi-element evidence examination and analysis module;
and the visual multi-element evidence examination and analysis module is used for enabling a case transacting person to view, compare and analyze the relevance and conflict relation of various sources and various types of evidence through a visual interface.
2. The Domain Ontology-based multivariate evidence association analysis system of claim 1, wherein Ontology (ontolog) refers to a graph data structure based on Semantic Web (Semantic Web) technology for describing a logical structure of a knowledge graph, the Ontology has a General Ontology (General ontolog) and a Domain Ontology (Domain ontolog), the General Ontology is used for construction and application of a knowledge base and a cross-Domain knowledge base, and the Domain Ontology is used for construction and application of a specific Domain specialized knowledge structure; the basic unit of the body is a triplet: head nodes, relations and tail nodes, wherein the head nodes are concepts and entities; tail nodes are concepts and entities, or attribute values; the relation is a semantic relation between head nodes and tail nodes, and is definitely defined by an ontology; the head node and the tail node form nodes of the data structure of the ontology graph, and the relationship forms a directed edge of the data structure of the ontology graph, so that the ontology is a directed graph.
3. The domain ontology-based multi-element evidence association analysis system of claim 2, wherein the financial crime case evidence censoring ontology is a domain ontology and for each of the crime names of the financial crimes, a specific ontology is defined, the ontologies of the plurality of crime names forming a set of financial crime case evidence censoring ontologies; the evidence examination ontology set for the financial crime cases has a certain shared ontology part, and each different crime name has a special part of the ontology.
4. The domain ontology-based multi-evidence association analysis system of claim 2, wherein the financial crime case evidence review rules include: the financial crime case evidence examination body, the evidence examination rule script and the evidence examination key point realize the definition of case facts of the financial crime case of a specific crime name and examination contents and methods of corresponding evidence.
5. The domain ontology-based multi-element evidence association analysis system according to claim 2, wherein the financial crime case data resource processing module processes case card data and file data of a case, the case card data is structured data, the file data is unstructured data, and the general concepts of the financial crime case and the case feature concepts of specific crimes defined in the financial crime case evidence examination ontology are identified and extracted through entity identification of natural language processing and case feature element identification model based on deep learning, and stored in a case element library in a structured format.
6. The domain ontology-based multi-element evidence association analysis system according to any one of claims 1-5, wherein the visual multi-element evidence review analysis module organizes case element data of a case with a financial crime case evidence review ontology as a structure and implements a user interface that organizes and displays the case element data from different dimensions according to requirements of a case transactor for evidence review of the case; the case handling personnel sequentially conduct comparison analysis on the same contents in the case elements according to the time sequence of inquiring the records; or gathering related case elements, and carrying out relevance analysis of different evidences; the evidence examination rules in the evidence examination rules of the financial crime cases are displayed on an interface, and a case transactor runs software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the evidence examination assistance is provided for the case transactor.
7. A domain ontology-based multivariate evidence association analysis method based on the system of any one of claims 1-6, the method comprising the steps of:
step 1: financial crime case evidence censoring ontology set and evidence censoring rule construction
The financial crime case evidence examination rule management module constructs a financial crime case evidence examination ontology set and evidence examination rules based on relevant law and judicial interpretation, various normative files of financial crime case examination and experience of a case transactor;
step 2: construction of feature element identification model of financial crime case
Labeling specific named entities in the case document based on a natural language processing entity recognition technology, constructing a batch of labeling data, and performing reinforcement learning by using a deep learning-based natural language processing model to obtain a financial crime case named entity recognition model; labeling case feature elements of specific crimes defined in a financial crime case evidence examination body, constructing a batch of labeling data, training the data by using a deep learning-based long text feature representation learning method, and constructing a case feature element identification model of the specific crimes;
step 3: case element identification and extraction of financial crime cases
The financial crime case data resource processing module invokes a case feature element recognition model of a specific crime name through recognition of the crime name information in the case card data, performs case feature element recognition and extraction on the text data of the case document of the financial crime in progress, and stores the case feature element recognition and extraction in a case element library; calling a financial crime case naming entity recognition model, recognizing and extracting general case elements of the text data of the in-process financial crime case files, and storing the general case elements into a case element library;
step 4: instantiating a financial crime case evidence review ontology for a particular crime name with case elements
When multi-element evidence association analysis is carried out, after a case name or a case number to be inspected is selected by a case handling person, a financial crime case evidence inspection body corresponding to the case name or the case number is automatically selected through crime name information in case card information, and general elements and characteristic elements of the case in a case element library are automatically filled in the body, so that instantiation of the body is realized;
step 5: visual multiple evidence association review analysis
The visual multi-element evidence examination analysis module calls the ontology data instantiated in the step 4 into a computer memory, and organizes and displays the case element data from different dimensions on an interface according to the requirement of a case handling person on evidence examination of the case; the evidence examination rules in the evidence examination rules of the financial crime cases are displayed on an interface, and a case transactor runs software on one or more of the evidence examination rules, so that the result of computer calculation and analysis is seen, and the evidence examination assistance is provided for the case transactor.
8. The domain ontology-based multivariate evidence association analysis system of claim 7, wherein said step 1 comprises:
step 1.1: constructing a set of evidence review ontologies for financial crime cases
Aiming at the common points and different characteristics of case facts and evidences of various financial crime cases of specific crime names, constructing a financial crime case evidence examination body of each crime name; the general concepts and relations of the ontology are common to all financial crime cases, and are defined by constructing a financial crime case general ontology, and the four conditions are established with crime constitution: subject, object, subjective aspect and objective aspect association; the financial crime case evidence examination ontology of each crime name inherits the general ontology of the financial crime case, inherits the related concepts of the general ontology, defines new subclasses, and realizes the definition of case elements such as specific crime facts and evidences of the crime name;
step 1.2: constructing evidence review rules for financial crime cases
Financial crime case evidence review rules are also two broad categories, namely general rules and specific rules related to crime names; the evidence examination general rule is suitable for case evidence examination of different crimes, and the crime related examination rule aims at a case of a specific crime; evidence examination rules are formulated in a script mode, and through various judging methods of data comparison and similarity scoring, the evidence examination rules are used for prompting a case transactor and recorded in a corresponding examination result data set for later retrieval.
9. The domain ontology-based multi-element evidence association analysis system according to claim 7, wherein in the step 4, a case clerk manually selects a financial crime case evidence review ontology of a specific crime name, and fills in case elements to implement ontology instantiation.
10. The domain ontology-based multivariate evidence association analysis system of claim 7, wherein in step 5, the office staff performs comparative analysis on the same contents in the case elements in sequence according to the time sequence of inquiring the records; or aggregating related case elements, and carrying out relevance analysis of different evidences.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311341606.7A CN117540027A (en) | 2023-10-17 | 2023-10-17 | Multi-element evidence association analysis system and method based on domain ontology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311341606.7A CN117540027A (en) | 2023-10-17 | 2023-10-17 | Multi-element evidence association analysis system and method based on domain ontology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117540027A true CN117540027A (en) | 2024-02-09 |
Family
ID=89788798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311341606.7A Pending CN117540027A (en) | 2023-10-17 | 2023-10-17 | Multi-element evidence association analysis system and method based on domain ontology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117540027A (en) |
-
2023
- 2023-10-17 CN CN202311341606.7A patent/CN117540027A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11860865B2 (en) | Analytic systems, methods, and computer-readable media for structured, semi-structured, and unstructured documents | |
| US11663254B2 (en) | System and engine for seeded clustering of news events | |
| Hutchins et al. | Hiding in plain sight: criminal network analysis | |
| CN112927082A (en) | Credit risk prediction method, apparatus, device, medium, and program product | |
| Chou et al. | Integrating XBRL data with textual information in Chinese: A semantic web approach | |
| CA2956627A1 (en) | System and engine for seeded clustering of news events | |
| CN110826315B (en) | Method for identifying timeliness of short text by using neural network system | |
| CN111027832A (en) | Tax risk determination method, apparatus and storage medium | |
| CN113378090B (en) | Internet website similarity analysis method and device and readable storage medium | |
| CN116151967A (en) | Fraudulent party identification system based on transaction knowledge graph | |
| CN112464670A (en) | Recognition method, recognition model training method, device, equipment and storage medium | |
| CN112149413A (en) | Method and device for identifying state of internet website based on neural network and computer readable storage medium | |
| Das et al. | A review on text analytics process with a CV parser model | |
| Hamad et al. | Sentiment analysis of restaurant reviews in social media using naïve bayes | |
| CN117540027A (en) | Multi-element evidence association analysis system and method based on domain ontology | |
| CN114741501A (en) | Public opinion early warning method and device, readable storage medium and electronic equipment | |
| CN113807920A (en) | Artificial intelligence based product recommendation method, device, equipment and storage medium | |
| CN113220843A (en) | Method, device, storage medium and equipment for determining information association relation | |
| CN113222471A (en) | Asset wind control method and device based on new media data | |
| Wang et al. | Preprocessing and feature extraction methods for microfinance overdue data | |
| Kulothungan | Loan Forecast by Using Machine Learning | |
| Correa et al. | Neural text classification for digital transformation in the financial regulatory domain | |
| Zhang et al. | A semantic search framework for similar audit issue recommendation in financial industry | |
| Chen et al. | Towards accurate search for e-commerce in steel industry: a knowledge-graph-based approach | |
| Anand et al. | Detecting sensitive information from unstructured text in a data-constrained environment |
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 |