CN108932340A - The construction method of financial knowledge mapping under a kind of non-performing asset operation field - Google Patents
The construction method of financial knowledge mapping under a kind of non-performing asset operation field Download PDFInfo
- Publication number
- CN108932340A CN108932340A CN201810767595.1A CN201810767595A CN108932340A CN 108932340 A CN108932340 A CN 108932340A CN 201810767595 A CN201810767595 A CN 201810767595A CN 108932340 A CN108932340 A CN 108932340A
- Authority
- CN
- China
- Prior art keywords
- ontology
- contract
- attribute
- library
- relationship
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of construction method of financial knowledge mapping under non-performing asset operation field, and the practice is as follows: firstly, carrying out triple building to the structured text in non-performing asset operation field;Secondly, extracting non-structured text information based on the intelligence of semantics recognition model and constructing triple, realize that the structuring of non-structured text is extracted;In turn, by the relationship of the attribute of ontology, ontology in contract and ontology pair, data fusion is carried out with original body library information, multi-source heterogeneous data are merged, visualization shows information relevant to user's search.Through the above steps, the present invention realizes knowledge reasoning, calculating and completion, to which comprehensive, true, effective information visualization is presented to business expert, to solve the problems, such as that non-performing asset operation field enterprise and practitioner lack air control decision support when commencing business.
Description
Technical field
The present invention provides a kind of construction method of financial knowledge mapping under non-performing asset operation field, belongs to financial field skill
Art.
Background technique
Non-performing asset operation field includes the purchase and disposition of non-performing asset packet, and it is fixed to be related to the valuation to non-performing asset packet
Valence and diversification method of disposal.With the arrival of data age, business personnel can more be facilitated in non-performing asset operation field
Ground obtains data information, however business personnel obtains high quality, high request, high accurately information still from the data information of magnanimity
It so requires a great deal of time and energy, workload is like looking for a needle in a haystack.Above situation not only reduces the work of business personnel
Make efficiency, also so that company is undertaken investment risk the imperfection being possible to because of information.Based on this status, need to establish not
Financial knowledge mapping in good assets management field is realized to the information management under the financial fields such as non-performing asset, and combines industry
Business rule efficiently auxiliary activities personnel carry out information penetrate, the risk prevention systems measure such as Risk-warning, and related service is carried out
Assistant analysis decision improves working efficiency.
The foundation of knowledge mapping is related to multiple fields, including natural language processing, graph theory, complex network, deep learning etc..
Knowledge mapping in financial field is not exclusively to above-mentioned field content, it is also necessary to by the stock of knowledge of expert, business is special
The thinking logic of family is converted to the expression logic of the ontology in knowledge mapping, increases the building difficulty of knowledge mapping.It uses for reference herein
The successful experience that medical domain knowledge mapping is established proposes the financial knowledge mapping building side in a kind of non-performing asset operation field
Method realizes the intelligent extraction to internal data and the intelligence fusion to multi-source heterogeneous data, and based on business expert's
Business rule and logic realize knowledge reasoning, calculating, completion, to be in by comprehensive, true, effective information visualization
Business expert is now given, to solve non-performing asset operation field enterprise and practitioner, air control decision is lacked when commencing business
The problem of support.
Summary of the invention
(1) purpose of the present invention
The purpose of the present invention is to provide a kind of construction methods of knowledge mapping financial under non-performing asset operation field, realize
Knowledge storage, reasoning to non-performing asset field.
(2) technical solution of the present invention
The construction method of financial knowledge mapping under a kind of non-performing asset operation field of the present invention, its step are as follows:
Step 1: being combed to the structural data in non-performing asset operation field, sorted out not using effective information
Ontology, Noumenon property in good assets management field, relationship, attribute of a relation form financial knowledge mapping dictionary, and then utilize and reflect
Resource description framework file of the file by Database Mapping at triple form is penetrated, RDF file is denoted as, for the original body of building
Library;
Step 2: intelligent by taking contract text as an example extract triple, the contract in non-performing asset operation field is carried out
Word segmentation processing, and character string identification is carried out to the vocabulary after word segmentation processing using special contract template, by the word after identification
Content converge as candidate entity;Candidate entity is screened using semantics recognition model, obtains the category of provider location, entity
Property, the relationship between entity and entity, construct the feature vector of the entity by physical contents and entity attribute, and using this to
Amount is matched with ontology library, determines the affiliated ontology of the entity;
Step 3: according to body contents and time term and financial knowledge mapping dictionary is combined, to the sheet in original body library
Body class and Noumenon property, ontological relationship, attribute of a relation are merged, and are stored using all information as historical data, with
Just knowledge reasoning is carried out, knowledge calculates, knowledge completion;
Step 4: using RDF query language for the specific information of user's input for the triple ontology library after merging
Speech, is denoted as sparql, is translated into relational query sentence inquiry triple ontology library, and return to relevant information;Then, it will look into
The triplet information ask carries out visualized operation, and wherein visualization tool utilizes the browser programming language of data-driven document
Frame is denoted as d3.js, generates dynamic relationship figure;
By above step, the present invention provides the financial knowledge mapping construction methods under non-performing asset field, by right
Triple building, the structuring of the non-structural data extraction, the fusion of multi-source heterogeneous data of structured data, realize knowledge and push away
Reason calculates, completion, so that comprehensive, true, effective information visualization is presented to business expert, to solve bad money
The problem of producing operation field enterprise and practitioner, air control decision support lacked when commencing business.
Wherein, at " structural data " described in step 1, refer in Oracle databases, be denoted as Oracle data
Library, the list structured data of storage;Effective information refers to the financial knowledge mapping relevant information in building non-performing asset field, comprising:
Company's master data, corporate linkage data, company's family tree data, personal tenure data etc..
Wherein, " extracting triple " described in step 2, the process established is as follows: firstly, contract text screens
For Word text, for the text type of extended formatting, needing to convert tool change first with file is Word text, if conversion
It is unsuccessful, then abandon the text;Secondly, segmenting using stammerer participle tool to Word text, mode is segmented are as follows: full word cutting
+ new word discovery+customized bag of words;Specific contract template includes the contract templates types such as debt-to-equity swap contract, assignment of credit contract;
" the semantics recognition model " refers to and carries out candidate entity judgement according to business rule and context semanteme, obtain entity
Relationship between position, entity attribute, entity.
Wherein, " fusion " mentioned in step 3, in particular to after obtaining the ontology information in contract, in order to close
Ontology and financial knowledge mapping dictionary in are compared one by one, if not including in contract in the ontology class in original body library
Ontology, then the ontology class in original body library is updated, adds new contract ontology, wherein in contract ontology attribute
Noumenon property as ontology library after update;If the ontology class in ontology library includes the ontology in contract, to original body library
In ontology class be updated, according to time attribute to attribute identical in ontology, select the attribute value in the nearest time;If this
Ontology in contract is then added in original body library relationship, contract by body to the relationship is not present in original body library
Middle ontology is the attribute of relationship in original body library to the attribute of relationship;If ontology in original body library there are the relationship,
Then based on contract in ontology in time attribute in relationship and original body library time attribute compare, select in the nearest time
Attribute value, and be put into other attribute as historical status in History noumenon library.
(3) advantages of the present invention and effect
Financial knowledge mapping construction method in a kind of non-performing asset operation field of the present invention, compared with prior art,
Advantage and effect are: (1) inquiring compared to traditional database association, the present invention utilizes natural language processing technique, intelligence
Change, it is efficient realize knowledge reasoning function, improve search efficiency, increase the work efficiency of business personnel;(2) pass through
Structural data and unstructured data are merged, personnel is reduced in multiple data source information search efficiencies, reduces business
The insufficient risk of the acquisition of information of personnel;(3) data memory format of triple is knowledge reasoning, calculating, completion provide
Data basis realizes comprehensive displaying of data, provides strong data support for the information inference of business personnel.
Detailed description of the invention
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.Attached drawing is only used for showing preferred embodiment, and is not considered as to limit of the invention
System, and throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 is construction method flow chart of the present invention.
Fig. 2 is a kind of framework embodiment flow chart of specific financial knowledge mapping provided by the invention.
Fig. 3 is ontology and ontological relationship schematic diagram provided by the invention.
Fig. 4 is a kind of monomer query case provided by the invention.
Fig. 5 is a kind of incidence relation example provided by the invention.
Specific embodiment
Below in conjunction with the attached drawing in the present invention, technical solution of the present invention progress is described clear and completely, it is clear that institute
The case of description is only a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, originally
Field those of ordinary skill every other embodiment obtained without making creative work, belongs to the present invention
The range of protection.
The present invention provides a kind of construction methods of knowledge mapping financial under non-performing asset operation field, and combine example detailed
Describe in detail it is bright, as shown in Figure 1, including the following steps:
Step 1: being arranged when structural data combs in non-performing asset operation field first with effective information
Ontology in non-performing asset operation field, Noumenon property, relationship, attribute of a relation out form financial knowledge mapping dictionary, Jin Erli
With mapping file by Database Mapping at the RDF file of triple form, the original body library that is constructed;
Step 2: intelligent by taking contract text as an example extract triple, the contract in non-performing asset operation field is carried out
Word segmentation processing, and character string identification is carried out to the vocabulary after word segmentation processing using special contract template, by the word after identification
Content converge as candidate entity;Candidate entity is screened using semantics recognition model, obtains the category of provider location, entity
Property, the relationship between entity and entity, construct the feature vector of the entity by physical contents and entity attribute, and using this to
Amount is matched with ontology library, determines the affiliated ontology of the entity;
Step 3: according to body contents and time term and financial knowledge mapping dictionary is combined, to the sheet in original body library
Body class and Noumenon property, ontological relationship, attribute of a relation are merged, and are stored using all information as historical data, with
Just knowledge reasoning is carried out, knowledge calculates, knowledge completion;
Step 4: for merge after triple ontology library, for user input specific information, using sparql by its
It is converted into relational query sentence inquiry triple ontology library, and returns to relevant information.Then, by the triplet information inquired into
Row visualized operation, wherein visualization tool generates Dynamic Display figure and dynamic relationship figure using d3.js.
In non-performing asset operation field provided by the invention in financial knowledge mapping construction method, firstly for structuring text
Then this building original body library carries out vocabulary extraction and Entity recognition to contract data source, by the feature for constructing entity
Vector determines ontology belonging to entity, includes the information and entity attributes of entity in feature vector.It in turn, will be in contract
Triplet information carries out information with original body library information and merges, and updates ontology library and store historical data, after finally merging
Triple ontology library, inquired using relational query sentence, and return to specific information, automatically generate financial knowledge graph
Spectrum, so as to provide more full and accurate effective reference scheme for the business expert in non-performing asset operation field.
In order to make it easy to understand, each step in above method embodiment is described in detail below, and it is situated between in detail
Implementation method in the present invention that continues, as shown in Fig. 2, the embodiment of the present invention is hierarchically introduced.
When structural data combs in non-performing asset operation field, bad money is sorted out first with effective information
Ontology, the Noumenon property, relationship, attribute of a relation in operation field are produced, forms financial knowledge mapping dictionary, and then utilize
Mapping file by Database Mapping at the RDF file of triple form, the original body library that is constructed.
Wherein, structured text information is stored in the form of oracle database table, the ontology library packet combed
It includes: ontology class, Noumenon property class, relation object, attribute of a relation class.In ontology class, the embodiment of the present invention include altogether enterprise, people,
Mechanism three classes;In Noumenon property class, the attribute of each ontology class is all different, enterprise's generic attribute include license the item to manage,
Unified credit code, address, enterprise name, management position, the type of business, the Date of Incorporation, grant date, legal representative/negative
Blame people/execution affairs partner, operating period from, operating period to, paid-up capital (ten thousand yuan), registered capital (ten thousand yuan), registration
Capital currency type, number of registration, registration authority, scopes of services, industrial sectors of national economy code, institutional framework code, registered capital
Currency Code, provinces and cities' information etc., the Attribute class of people include name, position, gender, ID card No. etc., and mechanical properties class includes
Organization names, class of establishment, administrative division, organization address etc.;Ontological relationship includes: investment, branch, legal person, case-involving etc., is being closed
In set attribute class, investment relation attribute includes enterprise's total quantity, cancelling date, the way of contributing investment, subscribes investment currency type, unified society
Credit code, enterprise institution's title, enterprise status, enterprise (mechanism) type, opening date, ratio between investments, legal representative's surname
Name etc., branch's attribute of a relation include branch address, branch's title, branch responsible person, enterprise of branch note
Volume number, general item to manage etc., legal person's attribute of a relation include name of judicial person, enterprise name, tenure time etc.;Case-involving attribute of a relation
Include: ID card No./Business Registration Number, Reference Number, case state, execute law court, execute target (member), executed person name/
Title, time of putting on record, executed person type, creation time, time of winding up the case, execution law court's area code etc..By above-mentioned data
Classification combs to form financial knowledge mapping dictionary.
For table structure specific in Oracle, the RDF file of mapping file generated triple is utilized.It is following with table 1
For table 2,
1. control paths of table-relationship array list
2. control path node array list of table
RDF file is obtained using the corresponding mapping file of Tables 1 and 2, wherein mapping file is various to include:
A table in relational database, is mapped as the class of RDF by nodes class and links class, and the NAME in table 1 corresponds to NODES
The NAME of class;ID corresponds to the NODES_ID of NODES class;NODE_TYPE corresponds to the NODE_TYPE of NODES class;ICL_ in table 2
FROM corresponds to the COMPANY_FROM of LINKS class;ICL_TO corresponds to the COMPANY_TO of LINKS class;TYPE corresponds to LINKS class
LINK_TYPE.The format of RDF file is showed with triple form, as shown in figure 3, being carried out in the form of<subject, relationship, attribute>
It shows, is denoted as { o, or, oa, wherein o is subject, orFor relationship, oaFor attribute.This triple form has established financial knowledge
It is merged in map using triple, the basis of operation, inquiry, while being provided the foundation for the formatted storage of data.
It is intelligent by taking contract text as an example to extract triple, the contract in non-performing asset operation field is carried out at participle
Reason, and character string identification is carried out to the vocabulary after word segmentation processing using special contract template, by the vocabulary content after identification
As candidate entity;Candidate entity is screened using semantics recognition model, obtains provider location, entity attributes, entity
Relationship between entity is constructed the feature vector of the entity by physical contents and entity attribute, and utilizes the vector and this
Body library is matched, and determines the affiliated ontology of the entity.
In embodiments of the present invention, it is cared in contract information comprising debenture transfer contract, purchasing contract, classified contract, finance
Ask the multiple types contract such as contract, certainly further include the data source that other record contract information, the embodiment of the present invention to this not
Make specific limit.
Essence of a contract is obtained according to business experience first, with X={ xi}1×nIndicate essence of a contract set, each xiIndicate one
A essence of a contract, content include ontology label.For a true contract C, text is divided using stammerer participle tool
Word takes participle mode are as follows: special dictionary+accurate participle+new word discovery.Autonomous dictionary source is divided into two parts: a part
To collect from network and business expert provides, another part is provides using probabilistic model.
It is collected on network and the special dictionary of business expert offer is main are as follows: real estate, " currency devaluation ", " Chinese
People bank " etc..Professional domain vocabulary step is obtained using probabilistic model are as follows: all legal contract files are extracted into text envelope
Breath, gets rid of non-Chinese character;The file information extracted is joined end to end with space;And then it is common using multiple characters
The left and right comentropy of the frequency of appearance and multiple characters, is screened based on certain threshold value, obtains multiple character compositions
Vocabulary;Using the vocabulary as special word of the invention, obligatory contract, herein etc. is specifically included that.Above two method is obtained
Vocabulary merge and become special dictionary as the special participle dictionary in of the invention.
After being segmented using aforesaid way to contract C, generating indicates contract C=with vector that word sequence is constituted
[w1, w2..., wm], wherein m indicates that contract C includes m vocabulary altogether.To each essence of a contract xiWith word order column vector C=[w1,
w2..., wm] string matching is carried out, by wordAs word wjWith essence of a contract xiSuccessful match is denoted as candidate entity.And then benefit
Contract ontology is extracted from candidate entity with semantic extraction modelAnd position and the contextual information of contract ontology are given, it takes out
Take the attribute of ontology relevant to the ontologyAnd the relationship between ontologyThe contract ontological construction triple that will be drawn into
Physical contents and entity attribute are constructed to the feature vector of the entity, the constituted mode of the vector are as follows:Wherein physical contents account for t with similarity in ontology library0Point, the category of some entity in each attribute and entity library
Property it is identical, then the matching value of the two entities pair add t1Point.Given threshold is Γ, if it exists highest matching value max t0+λ
t1> Γ then matches the ontology extracted in the ontology of highest scoring in ontology library and contract;If highest matching value max
t0+λt1< Γ is then not present the ontology, and is classified as most like ontology class according to its attribute in ontology library.
According to body contents and time term and financial knowledge mapping dictionary is combined, to the ontology class in original body library and originally
Body attribute, ontological relationship, attribute of a relation are merged, and are stored using all information as historical data, to be known
Know reasoning, knowledge calculates, knowledge completion.
In embodiments of the present invention, after obtaining the ontology information in contract, in order to by contract ontology and financial knowledge
Map dictionary is compared one by one, if not including the ontology in contract in the ontology class in original body library, to original body
Ontology class in library is updated, and adds new contract ontology, and wherein the attribute of ontology is as ontology library after updating in contract
Noumenon property;If the ontology class in ontology library includes the ontology in contract, the ontology class in original body library is carried out more
Newly, the attribute value in the nearest time is selected to attribute identical in ontology according to time attribute;If ontology is in original body library
In be not present the relationship, then the ontology in contract is added in original body library relationship, category of the ontology to relationship in contract
Property for relationship in original body library attribute;If ontology is in original body library, there are the relationships, based on contract in sheet
Body compares the time attribute in time attribute in relationship and original body library, selects the attribute value in the nearest time, and will be another
Outer attribute is put into History noumenon library as historical status.
Triple ontology library after merging is translated into the specific information of user's input using sparql
Relational query sentence inquires triple ontology library, and returns to relevant information.Then, the triplet information inquired is carried out visual
Change operation, wherein visualization tool generates Dynamic Display figure and dynamic relationship figure using d3.js.
In embodiments of the present invention, it by the triple ontology library of generation (RDF file), is looked into using sparql sentence
It askes, and carries out visualized operation using d3.js, generate the knowledge mapping under non-performing asset operation field.Example is had input in expert
After such as Business Name substance parameter, determine that expert wants after these substance parameters are carried out with participle and semantic parsing
The entity of input is being based on the generated knowledge mapping, can automatically generate and export in the information about the substance parameter
Hold, for business expert reference.
By taking inquiry is associated with 3 ontologies with ontology " Rui Tai Home Co., Ltd of Jixi County " as an example, need to utilize query statement
Are as follows:
SELECT? n WHERE
S rdf:type:Company.
The Jixi County s:companyName' Rui Tai Home Co., Ltd '
S:hasInvestIn? o.
O:companyName? n
}=
limit 3
The inquiry relational structure that then shows is as shown in figure 4, relationship to inquire between Liang Ge enterprise, as shown in Figure 5.
In the specific implementation, in ontology library here further include rule that preset entity needs to abide by.Specifically, this
In rule may include: someone be company A legal person, company B is the branch of company A, then the people is automatically replenished company B's
Independent legal person's information.
Therefore, method provided in an embodiment of the present invention further include:
In the rule and affiliated rule for needing to abide by by entity, the preset affiliated entity after participle operation identification
Associated another entity.
Specifically, for each entity identified, need to judge that the entity is to be stored in ontology library
The rule of the entity is limited, if so, then obtain based on another entity associated by the rule, it is final to obtain " entity, rule, reality
Body " triple.
Claims (4)
1. the construction method of financial knowledge mapping under a kind of non-performing asset operation field, it is characterised in that: its step are as follows:
Step 1: combing to the structural data in non-performing asset operation field, bad money is sorted out using effective information
Ontology, the Noumenon property, relationship, attribute of a relation in operation field are produced, forms financial knowledge mapping dictionary, and then utilize mapping text
Database Mapping at the resource description framework file of triple form, is denoted as RDF file by part, for the original body library of building;
Step 2: intelligent by taking contract text as an example extract triple, the contract in non-performing asset operation field is segmented
Processing, and character string identification is carried out to the vocabulary after word segmentation processing using special contract template, it will be in the vocabulary after identification
Hold as candidate entity;Candidate entity is screened using semantics recognition model, obtains provider location, entity attributes, reality
Relationship between body and entity, constructs the feature vector of the entity by physical contents and entity attribute, and using the vector with
Ontology library is matched, and determines the affiliated ontology of the entity;
Step 3: according to body contents and time term and financial knowledge mapping dictionary is combined, to the ontology class in original body library
Merged with Noumenon property, ontological relationship, attribute of a relation, and stored using all information as historical data, so as into
Row knowledge reasoning, knowledge calculates and knowledge completion;
Step 4:, for the specific information of user's input, using RDF query language, note for the triple ontology library after merging
For sparql, it is translated into relational query sentence inquiry triple ontology library, and returns to relevant information;Then, it will inquire
Triplet information carry out visualized operation, wherein visualization tool utilize data-driven document browser programming language frame
Frame is denoted as d3.js, generates dynamic relationship figure;
By above step, the present invention provides the spectrum construction methods of knowledge graph financial under non-performing asset field, by structure
Triple building, the structuring of the non-structural data extraction, the fusion of multi-source heterogeneous data of data, realize knowledge reasoning, meter
It calculates and completion passes through so that comprehensive, true, effective information visualization is presented to business expert to solve non-performing asset
Field enterprise of battalion and practitioner lack the problem of air control decision support when commencing business.
2. the construction method of financial knowledge mapping, feature under a kind of non-performing asset operation field according to claim 1
It is:
At " structural data " described in step 1, refer in Oracle databases, is denoted as oracle database, the table of storage
Structured data;Effective information refers to the financial knowledge mapping relevant information in building non-performing asset field, comprising: company's basic number
According to, corporate linkage data, company's family tree data and personal tenure data.
3. the construction method of financial knowledge mapping, feature under a kind of non-performing asset operation field according to claim 1
It is:
" extracting triple " described in step 2, the process established is as follows: firstly, contract text screening is Word text
This, for the text type of extended formatting, needing to convert tool change first with file is Word text, if conversion is unsuccessful,
Then abandon the text;Secondly, segmenting using stammerer participle tool to Word text, mode is segmented are as follows: full word cutting+neologisms hair
Existing+customized bag of words;Specific contract template includes debt-to-equity swap contract, all contract template types of assignment of credit contract;Described
" semantics recognition model " refers to and carries out candidate entity judgement according to business rule and context semanteme, obtain provider location, reality
Relationship between body attribute and entity.
4. the construction method of financial knowledge mapping, feature under a kind of non-performing asset operation field according to claim 1
It is:
" fusion " mentioned in step 3, in particular to after obtaining the ontology information in contract, in order to by the sheet in contract
Body and financial knowledge mapping dictionary are compared one by one, if not including the ontology in contract in the ontology class in original body library,
Then the ontology class in original body library is updated, adds new contract ontology, wherein the attribute of ontology is used as more in contract
The Noumenon property of ontology library after new;If the ontology class in ontology library includes the ontology in contract, to the sheet in original body library
Body class is updated, and according to time attribute to attribute identical in ontology, selects the attribute value in the nearest time;If ontology to
The relationship is not present in original body library, then is added to the ontology in contract in original body library to relationship, ontology in contract
Attribute to relationship is the attribute of relationship in original body library;If ontology is in original body library, there are the relationship, bases
Ontology in contract compares the time attribute in time attribute in relationship and original body library, selects the attribute in the nearest time
Value, and be put into other attribute as historical status in History noumenon library.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810767595.1A CN108932340A (en) | 2018-07-13 | 2018-07-13 | The construction method of financial knowledge mapping under a kind of non-performing asset operation field |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810767595.1A CN108932340A (en) | 2018-07-13 | 2018-07-13 | The construction method of financial knowledge mapping under a kind of non-performing asset operation field |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108932340A true CN108932340A (en) | 2018-12-04 |
Family
ID=64447206
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810767595.1A Pending CN108932340A (en) | 2018-07-13 | 2018-07-13 | The construction method of financial knowledge mapping under a kind of non-performing asset operation field |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108932340A (en) |
Cited By (34)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109710775A (en) * | 2018-12-29 | 2019-05-03 | 北京航天云路有限公司 | A kind of knowledge mapping dynamic creation method based on more rules |
CN109885691A (en) * | 2019-01-08 | 2019-06-14 | 平安科技(深圳)有限公司 | Knowledge mapping complementing method, device, computer equipment and storage medium |
CN109918511A (en) * | 2019-01-29 | 2019-06-21 | 华融融通(北京)科技有限公司 | A kind of knowledge mapping based on BFS and LPA is counter to cheat feature extracting method |
CN109922075A (en) * | 2019-03-22 | 2019-06-21 | 中国南方电网有限责任公司 | Network security knowledge map construction method and apparatus, computer equipment |
CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
CN110059912A (en) * | 2019-02-20 | 2019-07-26 | 国网浙江省电力有限公司杭州供电公司 | Honest risk hidden danger panorama management-control method and device based on wisdom map |
CN110083823A (en) * | 2019-03-07 | 2019-08-02 | 平安科技(深圳)有限公司 | Dictionary sheet method for building up and device, computer installation and storage medium |
CN110197280A (en) * | 2019-05-20 | 2019-09-03 | 中国银行股份有限公司 | A kind of knowledge mapping construction method, apparatus and system |
CN110275894A (en) * | 2019-06-24 | 2019-09-24 | 恒生电子股份有限公司 | A kind of update method of knowledge mapping, device, electronic equipment and storage medium |
CN110489560A (en) * | 2019-06-19 | 2019-11-22 | 民生科技有限责任公司 | The little Wei enterprise portrait generation method and device of knowledge based graphical spectrum technology |
CN110532449A (en) * | 2019-08-30 | 2019-12-03 | 盈盛智创科技(广州)有限公司 | A kind of processing method of service profile, device, equipment and storage medium |
CN110569369A (en) * | 2019-09-16 | 2019-12-13 | 神州数码融信软件有限公司 | Generation method and device, application method and device of knowledge graph of bank financial system |
CN110598003A (en) * | 2019-08-15 | 2019-12-20 | 上海市大数据中心 | Knowledge graph construction system and construction method based on public data resource catalog |
CN110765272A (en) * | 2019-09-12 | 2020-02-07 | 平安医疗健康管理股份有限公司 | Knowledge graph-based signing method and device, computer equipment and computer storage medium |
CN110781213A (en) * | 2019-09-25 | 2020-02-11 | 中国电子进出口有限公司 | Multi-source mass data correlation searching method and system with personnel as center |
CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
CN110941612A (en) * | 2019-11-19 | 2020-03-31 | 上海交通大学 | Autonomous data lake construction system and method based on associated data |
CN110990585A (en) * | 2019-11-29 | 2020-04-10 | 上海勘察设计研究院(集团)有限公司 | Multi-source data and time sequence processing method and device for constructing industry knowledge graph |
CN111078868A (en) * | 2019-06-04 | 2020-04-28 | 中国人民解放军92493部队参谋部 | Knowledge graph analysis-based equipment test system planning decision method and system |
CN111160707A (en) * | 2019-11-29 | 2020-05-15 | 广东轩辕网络科技股份有限公司 | Intelligent work cooperation and resource sharing method and device |
CN111198852A (en) * | 2019-12-30 | 2020-05-26 | 浪潮通用软件有限公司 | Knowledge graph driven metadata relation reasoning method under micro-service architecture |
CN111291197A (en) * | 2020-03-02 | 2020-06-16 | 北京邮电大学 | Knowledge base construction system based on new word discovery algorithm |
CN111598702A (en) * | 2020-04-14 | 2020-08-28 | 徐佳慧 | Knowledge graph-based method for searching investment risk semantics |
CN111858963A (en) * | 2020-07-28 | 2020-10-30 | 中国银行股份有限公司 | Webpage customer service knowledge extraction method and device |
CN111858948A (en) * | 2019-04-30 | 2020-10-30 | 杭州海康威视数字技术股份有限公司 | Ontology construction method and device, electronic equipment and storage medium |
CN112069817A (en) * | 2020-07-17 | 2020-12-11 | 中国科学院计算机网络信息中心 | Student knowledge extraction and fusion method and device |
CN112131275A (en) * | 2020-09-23 | 2020-12-25 | 中国科学技术大学智慧城市研究院(芜湖) | Enterprise portrait construction method of holographic city big data model and knowledge graph |
CN112699245A (en) * | 2019-10-18 | 2021-04-23 | 北京国双科技有限公司 | Construction method and device and application method and device of budget management knowledge graph |
CN112905808A (en) * | 2021-03-29 | 2021-06-04 | 北京机电工程研究所 | Knowledge graph construction method and device and electronic equipment |
CN113064998A (en) * | 2020-12-18 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Map construction method and device for financial institution wind control and storage medium |
CN113486124A (en) * | 2021-05-24 | 2021-10-08 | 山东佳联电子商务有限公司 | Bank bad asset management and management system, method, equipment and storage medium based on PCA and knowledge graph technology |
CN114003734A (en) * | 2021-11-22 | 2022-02-01 | 四川大学华西医院 | Breast cancer risk factor knowledge system model, knowledge map system and construction method |
CN115510196A (en) * | 2021-06-07 | 2022-12-23 | 马上消费金融股份有限公司 | Knowledge graph construction method, question answering method, device and storage medium |
CN117217807A (en) * | 2023-11-08 | 2023-12-12 | 四川智筹科技有限公司 | Bad asset valuation algorithm based on multi-mode high-dimensional characteristics |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105574098A (en) * | 2015-12-11 | 2016-05-11 | 百度在线网络技术(北京)有限公司 | Knowledge graph generation method and device and entity comparing method and device |
CN106547733A (en) * | 2016-10-19 | 2017-03-29 | 中国国防科技信息中心 | A kind of name entity recognition method towards particular text |
CN107122444A (en) * | 2017-04-24 | 2017-09-01 | 北京科技大学 | A kind of legal knowledge collection of illustrative plates method for auto constructing |
CN107657063A (en) * | 2017-10-30 | 2018-02-02 | 合肥工业大学 | The construction method and device of medical knowledge collection of illustrative plates |
CN107783973A (en) * | 2016-08-24 | 2018-03-09 | 慧科讯业有限公司 | The methods, devices and systems being monitored based on domain knowledge spectrum data storehouse to the Internet media event |
CN107945024A (en) * | 2017-12-12 | 2018-04-20 | 厦门市美亚柏科信息股份有限公司 | Identify that internet finance borrowing enterprise manages abnormal method, terminal device and storage medium |
CN107958091A (en) * | 2017-12-28 | 2018-04-24 | 北京贝塔智投科技有限公司 | A kind of NLP artificial intelligence approaches and interactive system based on financial vertical knowledge mapping |
-
2018
- 2018-07-13 CN CN201810767595.1A patent/CN108932340A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105574098A (en) * | 2015-12-11 | 2016-05-11 | 百度在线网络技术(北京)有限公司 | Knowledge graph generation method and device and entity comparing method and device |
CN107783973A (en) * | 2016-08-24 | 2018-03-09 | 慧科讯业有限公司 | The methods, devices and systems being monitored based on domain knowledge spectrum data storehouse to the Internet media event |
CN106547733A (en) * | 2016-10-19 | 2017-03-29 | 中国国防科技信息中心 | A kind of name entity recognition method towards particular text |
CN107122444A (en) * | 2017-04-24 | 2017-09-01 | 北京科技大学 | A kind of legal knowledge collection of illustrative plates method for auto constructing |
CN107657063A (en) * | 2017-10-30 | 2018-02-02 | 合肥工业大学 | The construction method and device of medical knowledge collection of illustrative plates |
CN107945024A (en) * | 2017-12-12 | 2018-04-20 | 厦门市美亚柏科信息股份有限公司 | Identify that internet finance borrowing enterprise manages abnormal method, terminal device and storage medium |
CN107958091A (en) * | 2017-12-28 | 2018-04-24 | 北京贝塔智投科技有限公司 | A kind of NLP artificial intelligence approaches and interactive system based on financial vertical knowledge mapping |
Cited By (49)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109710775A (en) * | 2018-12-29 | 2019-05-03 | 北京航天云路有限公司 | A kind of knowledge mapping dynamic creation method based on more rules |
CN109885691A (en) * | 2019-01-08 | 2019-06-14 | 平安科技(深圳)有限公司 | Knowledge mapping complementing method, device, computer equipment and storage medium |
CN109918511A (en) * | 2019-01-29 | 2019-06-21 | 华融融通(北京)科技有限公司 | A kind of knowledge mapping based on BFS and LPA is counter to cheat feature extracting method |
CN109918511B (en) * | 2019-01-29 | 2021-06-08 | 华融融通(北京)科技有限公司 | BFS and LPA based knowledge graph anti-fraud feature extraction method |
CN110008288B (en) * | 2019-02-19 | 2021-06-29 | 武汉烽火技术服务有限公司 | Construction method and application of knowledge map library for network fault analysis |
CN110008288A (en) * | 2019-02-19 | 2019-07-12 | 武汉烽火技术服务有限公司 | The construction method in the knowledge mapping library for Analysis of Network Malfunction and its application |
CN110059912A (en) * | 2019-02-20 | 2019-07-26 | 国网浙江省电力有限公司杭州供电公司 | Honest risk hidden danger panorama management-control method and device based on wisdom map |
CN110083823A (en) * | 2019-03-07 | 2019-08-02 | 平安科技(深圳)有限公司 | Dictionary sheet method for building up and device, computer installation and storage medium |
CN110083823B (en) * | 2019-03-07 | 2024-03-29 | 平安科技(深圳)有限公司 | Dictionary table establishing method and device, computer device and storage medium |
CN109922075A (en) * | 2019-03-22 | 2019-06-21 | 中国南方电网有限责任公司 | Network security knowledge map construction method and apparatus, computer equipment |
CN111858948A (en) * | 2019-04-30 | 2020-10-30 | 杭州海康威视数字技术股份有限公司 | Ontology construction method and device, electronic equipment and storage medium |
CN110197280B (en) * | 2019-05-20 | 2021-08-06 | 中国银行股份有限公司 | Knowledge graph construction method, device and system |
CN110197280A (en) * | 2019-05-20 | 2019-09-03 | 中国银行股份有限公司 | A kind of knowledge mapping construction method, apparatus and system |
CN111078868A (en) * | 2019-06-04 | 2020-04-28 | 中国人民解放军92493部队参谋部 | Knowledge graph analysis-based equipment test system planning decision method and system |
CN110489560A (en) * | 2019-06-19 | 2019-11-22 | 民生科技有限责任公司 | The little Wei enterprise portrait generation method and device of knowledge based graphical spectrum technology |
CN110275894A (en) * | 2019-06-24 | 2019-09-24 | 恒生电子股份有限公司 | A kind of update method of knowledge mapping, device, electronic equipment and storage medium |
CN110598003A (en) * | 2019-08-15 | 2019-12-20 | 上海市大数据中心 | Knowledge graph construction system and construction method based on public data resource catalog |
CN110532449B (en) * | 2019-08-30 | 2022-05-31 | 盈盛智创科技(广州)有限公司 | Method, device, equipment and storage medium for processing service document |
CN110532449A (en) * | 2019-08-30 | 2019-12-03 | 盈盛智创科技(广州)有限公司 | A kind of processing method of service profile, device, equipment and storage medium |
CN110765272B (en) * | 2019-09-12 | 2022-08-26 | 深圳平安医疗健康科技服务有限公司 | Knowledge graph-based signing method and device, computer equipment and computer storage medium |
CN110765272A (en) * | 2019-09-12 | 2020-02-07 | 平安医疗健康管理股份有限公司 | Knowledge graph-based signing method and device, computer equipment and computer storage medium |
CN110569369A (en) * | 2019-09-16 | 2019-12-13 | 神州数码融信软件有限公司 | Generation method and device, application method and device of knowledge graph of bank financial system |
CN110781213A (en) * | 2019-09-25 | 2020-02-11 | 中国电子进出口有限公司 | Multi-source mass data correlation searching method and system with personnel as center |
CN112699245A (en) * | 2019-10-18 | 2021-04-23 | 北京国双科技有限公司 | Construction method and device and application method and device of budget management knowledge graph |
CN110825721B (en) * | 2019-11-06 | 2023-05-02 | 武汉大学 | Method for constructing and integrating hypertension knowledge base and system in big data environment |
CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
CN110941612A (en) * | 2019-11-19 | 2020-03-31 | 上海交通大学 | Autonomous data lake construction system and method based on associated data |
CN110941612B (en) * | 2019-11-19 | 2020-08-11 | 上海交通大学 | Autonomous data lake construction system and method based on associated data |
CN111160707A (en) * | 2019-11-29 | 2020-05-15 | 广东轩辕网络科技股份有限公司 | Intelligent work cooperation and resource sharing method and device |
CN110990585B (en) * | 2019-11-29 | 2024-01-30 | 上海勘察设计研究院(集团)股份有限公司 | Multi-source data and time sequence processing method and device for building industry knowledge graph |
CN110990585A (en) * | 2019-11-29 | 2020-04-10 | 上海勘察设计研究院(集团)有限公司 | Multi-source data and time sequence processing method and device for constructing industry knowledge graph |
CN111198852A (en) * | 2019-12-30 | 2020-05-26 | 浪潮通用软件有限公司 | Knowledge graph driven metadata relation reasoning method under micro-service architecture |
CN111291197B (en) * | 2020-03-02 | 2021-05-11 | 北京邮电大学 | Knowledge base construction system based on new word discovery algorithm |
CN111291197A (en) * | 2020-03-02 | 2020-06-16 | 北京邮电大学 | Knowledge base construction system based on new word discovery algorithm |
CN111598702A (en) * | 2020-04-14 | 2020-08-28 | 徐佳慧 | Knowledge graph-based method for searching investment risk semantics |
CN112069817A (en) * | 2020-07-17 | 2020-12-11 | 中国科学院计算机网络信息中心 | Student knowledge extraction and fusion method and device |
CN111858963A (en) * | 2020-07-28 | 2020-10-30 | 中国银行股份有限公司 | Webpage customer service knowledge extraction method and device |
CN111858963B (en) * | 2020-07-28 | 2024-02-23 | 中国银行股份有限公司 | Webpage customer service knowledge extraction method and device |
CN112131275B (en) * | 2020-09-23 | 2023-07-25 | 长三角信息智能创新研究院 | Enterprise portrait construction method of holographic city big data model and knowledge graph |
CN112131275A (en) * | 2020-09-23 | 2020-12-25 | 中国科学技术大学智慧城市研究院(芜湖) | Enterprise portrait construction method of holographic city big data model and knowledge graph |
CN113064998A (en) * | 2020-12-18 | 2021-07-02 | 开鑫金服(南京)信息服务有限公司 | Map construction method and device for financial institution wind control and storage medium |
CN112905808A (en) * | 2021-03-29 | 2021-06-04 | 北京机电工程研究所 | Knowledge graph construction method and device and electronic equipment |
CN113486124B (en) * | 2021-05-24 | 2022-02-11 | 山东佳联电子商务有限公司 | Bank bad asset management and management system, method, equipment and storage medium based on PCA and knowledge graph technology |
CN113486124A (en) * | 2021-05-24 | 2021-10-08 | 山东佳联电子商务有限公司 | Bank bad asset management and management system, method, equipment and storage medium based on PCA and knowledge graph technology |
CN115510196A (en) * | 2021-06-07 | 2022-12-23 | 马上消费金融股份有限公司 | Knowledge graph construction method, question answering method, device and storage medium |
CN114003734A (en) * | 2021-11-22 | 2022-02-01 | 四川大学华西医院 | Breast cancer risk factor knowledge system model, knowledge map system and construction method |
CN114003734B (en) * | 2021-11-22 | 2023-06-30 | 四川大学华西医院 | Knowledge system and knowledge map system of breast cancer risk factors and construction method |
CN117217807A (en) * | 2023-11-08 | 2023-12-12 | 四川智筹科技有限公司 | Bad asset valuation algorithm based on multi-mode high-dimensional characteristics |
CN117217807B (en) * | 2023-11-08 | 2024-01-26 | 四川智筹科技有限公司 | Bad asset estimation method based on multi-mode high-dimensional characteristics |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108932340A (en) | The construction method of financial knowledge mapping under a kind of non-performing asset operation field | |
CN110825882B (en) | Knowledge graph-based information system management method | |
CN110633409B (en) | Automobile news event extraction method integrating rules and deep learning | |
CN111026842B (en) | Natural language processing method, natural language processing device and intelligent question-answering system | |
US20200334418A1 (en) | Applied Artificial Intelligence Technology for Using Natural Language Processing and Concept Expression Templates to Train a Natural Language Generation System | |
Korom | A bibliometric visualization of the economics and sociology of wealth inequality: a world apart? | |
CN106776711B (en) | Chinese medical knowledge map construction method based on deep learning | |
RU2637992C1 (en) | Method of extracting facts from texts on natural language | |
KR100533810B1 (en) | Semi-Automatic Construction Method for Knowledge of Encyclopedia Question Answering System | |
Tang et al. | Using Bayesian decision for ontology mapping | |
RU2662688C1 (en) | Extraction of information from sanitary blocks of documents using micromodels on basis of ontology | |
US20170213157A1 (en) | Method and system to provide related data | |
CN111708773A (en) | Multi-source scientific and creative resource data fusion method | |
CN110990590A (en) | Dynamic financial knowledge map construction method based on reinforcement learning and transfer learning | |
CN110297872A (en) | A kind of building, querying method and the system of sciemtifec and technical sphere knowledge mapping | |
CN103425740B (en) | A kind of material information search method based on Semantic Clustering of internet of things oriented | |
CN113673943B (en) | Personnel exemption aided decision making method and system based on historical big data | |
CN111858940A (en) | Multi-head attention-based legal case similarity calculation method and system | |
CN112183059B (en) | Chinese structured event extraction method | |
Wick et al. | A unified approach for schema matching, coreference and canonicalization | |
CN112508743B (en) | Technology transfer office general information interaction method, terminal and medium | |
CN116127090B (en) | Aviation system knowledge graph construction method based on fusion and semi-supervision information extraction | |
CN116127084A (en) | Knowledge graph-based micro-grid scheduling strategy intelligent retrieval system and method | |
CN105335510A (en) | Text data efficient searching method | |
CN105160046A (en) | Text-based data retrieval method |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181204 |
|
WD01 | Invention patent application deemed withdrawn after publication |