CN109063045A - A kind of financial service method and financial service terminal - Google Patents
A kind of financial service method and financial service terminal Download PDFInfo
- Publication number
- CN109063045A CN109063045A CN201810788831.8A CN201810788831A CN109063045A CN 109063045 A CN109063045 A CN 109063045A CN 201810788831 A CN201810788831 A CN 201810788831A CN 109063045 A CN109063045 A CN 109063045A
- Authority
- CN
- China
- Prior art keywords
- information
- financial
- financial service
- gain
- data
- 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
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of financial service method and financial service terminal, include the following steps: that step 100, Heterogeneous Information obtain automatically；Step 200, information processing, semi-structured processing is carried out to the Heterogeneous Information obtained automatically, the basic datas such as user's present quotation data, hot spot information, return of security investment data are supplied to, while providing industry analysis report, the prediction data etc. of specialty analysis teacher or institutional investor's publication for user；Step 300, development of judgment matrix compare Multiple factors index in item layer two-by-two, using analytic hierarchy process (AHP), determine its weight, development of judgment matrix；Step 400, Information Authentication, according to weight calculation result, financial Heterogeneous Information processing result is verified, guarantee the accuracy of data and the validity of financial knowledge services, internet Heterogeneous Information processing technique is applied to financial knowledge services, by financial knowledge services process modelling, and improve the accuracy rate of time series forecasting.
Description
Technical field
The present invention relates to financial services industry, specially a kind of financial service method and financial service terminal.
Background technique
Financial knowledge services are financial knowledge to be merged with financial service, knowledge is converted by finance data, information, with mutual
The platforms such as networking, Xinhua Finance Media provide service for carrier for user.With the continuous improvement of national life level and investment awareness,
Demand to financial knowledge services is also higher and higher, gradually develops towards diversification, personalized direction.In recent years, with mutual
The rapid development of networking and universal, internet has become the main data source and carrier of financial knowledge services.Towards net
Also there are unprecedented opportunities and challenge in the financial knowledge services of network, and there is also following shortcomings:
For example, patent name is a kind of patent of invention of financial service method application No. is 201510763272.1:
Its financial service terminal integrates the withdrawal function of the convenience service function of POS machine and ATM machine, can be realized
Convenience service and withdrawal, and the convenience widely distributed, raising financial service terminal uses can be carried out, it provides the user with conveniently.
But existing financial service method and financial service terminal have the following deficiencies:
(1) in financial knowledge services field, with emerging in large numbers for financial market increasingly keen competition and magnanimity Financial Information,
Requirement of the people to financial knowledge services is higher and higher, and following problem is come into being, and how to improve Internet-based
Information retrieval accuracy rate provides accurate available especially for the distinctive time sensitivity in financial field for Heterogeneous Information acquisition
Data source；The accuracy that feature extracts automatically how is improved, realizes automatic, efficient information pre-processing, for further text point
Analysis, data mining provide basic data etc.；
(2) large-scale financial services system is a kind of Complex Information System, and security risk is small with probability of happening, destroys journey
Big, assessment difficulty of degree etc. is held a little, the existing safety risk estimating method in current information security field or it is too simple and can not
Reflect the characteristics of Complex Information System completely or excessively theorize and lead to not implement to fall in actual operation system
Ground, therefore often lack applicable methods of risk assessment when floaing service system in face of Big Gold, it can not be to large-scale financial service
The security risk of system carries out more accurate evaluation and analyzes.
Summary of the invention
In order to overcome the shortcomings of prior art, the present invention provides a kind of financial service method and financial service terminal,
It can effectively solve the problem of background technique proposes.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of financial service method, includes the following steps:
Step 100, Heterogeneous Information obtain automatically, obtain automatically from data sources such as internet, Xinhua Finance Media and listed companies
The data and text information of various structures；
Step 200, information processing carry out semi-structured processing to the Heterogeneous Information obtained automatically, it is real-time to be supplied to user
The basic datas such as market data, hot spot information, return of security investment data, while specialty analysis teacher or institutional investment are provided for user
Industry analysis report, prediction data that person issues etc.；
Step 300, development of judgment matrix compare Multiple factors index in item layer, two-by-two using step analysis
Method determines its weight, development of judgment matrix；
Step 400, Information Authentication guarantee according to weight calculation as a result, verifying to financial Heterogeneous Information processing result
The validity of the accuracy of data and financial knowledge services.
Further, in the step 100, further includes:
Firstly, building power field text, is parsed by the OWL file to description ontology, obtains decimation rule, it will be general
Thought, relationship, hierarchical structure etc. parse deposit database；
Then, using dictionary editing machine record, editing conception, the corresponding keyword of relationship, decimation rule is carried out perfect；
The Web information of acquisition is pre-processed again；
Finally, removing the unwanted contributions in sentence by diathesis alternation.
Further, Web information carries out pretreated method in the step 100 are as follows:
Step 101, the amendment weighted value using Information Security Risk, to reflect the conduction effect of Information Security Risk；
Step 102 assumes that assets A is made of m functional module, and respectively A1, A2, A3 ... Am, wherein Ai is assets A's
I-th of functional module, similarly, it is assumed that assets B is made of n functional module, respectively B1, B2, B3 ... Bm, and wherein Bj is assets
J-th of functional module of B；
Step 103, using certain functional module Ai on A to the access relation of certain functional module Bj on assets B, attack assets
B；
Step 104, the risk that modules A i to module Bj is formed according to the dependence between Ai and Bj conduct RC, are denoted as
RCAi:Bj；
Step 105, RCAi:Bj risk information use<A, i, B, j, P, W>hexa-atomic group indicate, wherein P is indicated using should
The probability of happening for the security incident that risk conduction model calculates also becomes weight, P ∈ [0,1]；W indicates a functional module pair
The influence degree of another functional module, W ∈ [0 ,+∞].
Further, the step 200 further include:
Firstly, carrying out text classification to Heterogeneous Information:
Training set D={ (x equipped with n samplei, yi), x ∈ Rn, i=1,2,3...n, y ∈ [- 1,1], training D can
With right-on separated by certain hyperplane, and the distance between make two class samples and hyperplane and maximize；
Secondly, the characteristics of data object is divided into cluster, is had to cluster according to similarity is analysed in depth and is summarized:
Given news agregator X={ x1, x2... xn, for each text xiComprising m feature f, then xi={ fi1,
fi2... fim, i=1,2 ..., n obtain cluster result C=by feature calculation similarity, and according to Clustering Decision-Making standard
{c1, c2..., ck, and cluster result meets: ci∩cj,Wherein, i ≠ j, i, j=1,
2 ..., n；
Finally, establishing the dynamic relation of interdependence model in reflecting time series between data according to historical record, pass through
The model predicts future trend.
Further, the step 300 further include:
Step 301, the product M for calculating each row element of judgment matrix Ai；
Step 302 calculates MiN times root Wi；
Step 303, to resulting phasor W={ W1, W2, W3Standardize, make ∑ Wi=1 (i=1,2,3), then W is
For required weight.
Further, in the step 400 further include:
Step 401 introduces RI index, calculates the maximum eigenvalue of judgment matrix
Step 402 calculates random index CI,
Step 403 calculates random consistency ratio CR,
Further, need to find the optimum state sequence of each sub-stage in the step 400:
Firstly, ratio of profit increase of the gain network node based on original capital is calculated, the gain that each division beneficiating process obtains
Amount isOriginal capital isThe then node S in division gain networkijRatio of profit increase relative to original capital isIt is located at gain path interior joint SijWeight be Δ ωij, then
Secondly, determining separate unit as sub-stage；
Again, the optimal path of sub-stage is calculated, the state of each node is three-dimensional vector Si(ΔSi, ti, θi), wherein ti
It is in node SiThe time distributed, θiIt is corresponding gain recovery time, the optimal of sub-stage is calculated by class dijkstra's algorithm
Path；
Furthermore record each gain node amountCorresponding gain time of return θi, optimal road is obtained through the above steps
DiameterWith optimum state sequenceAccording to gain time of return θiIt calculates
The optimum gain amount of sub-stage out
Finally, searching again for next independent sub-stage through the above steps and calculating amount of gain, terminated until the gain time,
End loop.
In addition the present invention also provides a kind of financial service terminals for claim 1 the method, including finance clothes
Business platform, enterprise database and work station；
The signal end of the work station establishes financial service Heterogeneous Information library by service subsystem, and passes through wireless network
It is connected with financial service platform, the data terminal of the financial service platform and enterprise's numerical control library interconnect；
The financial service platform carries out dissection process for extracting Financial Information publication and law rules and regulations ordinances information,
Coincidence matching is carried out extracting information from enterprise database, obtains matching result；
The data terminal of the work station is connected with personal computer, and the data terminal of the personal computer passes through wireless network
It is connected with knowledge services customization end, knowledge services customization end is internally provided with service type definition module；
The signal end at knowledge services customization end is connected with server by knowledge services analytical algorithm.
Compared with prior art, the beneficial effects of the present invention are:
(1) internet Heterogeneous Information processing technique is applied to financial knowledge services by financial service terminal of the invention, will
Financial knowledge services process modelling；New stock investment return is realized using the division gain model of proposition to analyze；Utilize the time
Sequential forecasting models realize the real-time net value valuation of close-ended fund, and improve the accuracy rate of time series forecasting；It utilizes
Search result assessment models Internet-based, building financial field ontology etc. realize corporate bond Knowledge Service Platform from
Adaptability；It ensure that using information automatic verification method and the accuracy of data be provided；
(2) financial service terminal of the invention is commented using based on the large-scale financial services system security risk that risk is conducted
Two aspects of large-scale financial services system security risk assessment specification for estimating model and Test driver, form a set of suitable for large size
The safety risk estimating method of financial services system, the information security risk evaluation of large-scale financial services system is ground ingeniously will tool
There is certain reference, it will be with certain finger for the information security risk evaluation implementation of large-scale financial services system
Meaning is led, will also have good reference significance for other management domains of large-scale financial services system.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention；
Fig. 2 is structural block diagram of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in Fig. 2, the present invention provides a kind of financial service terminal, including financial service platform, enterprise database and
Work station；The signal end of the work station establishes financial service Heterogeneous Information library by service subsystem, and passes through wireless network
It is connected with financial service platform, the data terminal of the financial service platform and enterprise's numerical control library interconnect；The finance clothes
Business platform carries out dissection process for extracting Financial Information publication and law rules and regulations ordinances information, is mentioning from enterprise database
Breath of winning the confidence carries out coincidence matching, obtains matching result；The data terminal of the work station is connected with personal computer, the personal meter
The data terminal of calculation machine has knowledge services to customize end by wireless network connection, and knowledge services customization end is internally provided with clothes
Service type definition module；The signal end at knowledge services customization end is connected with server by knowledge services analytical algorithm.
In the present embodiment, internet finance Heterogeneous Information processing method is proposed, Heterogeneous Information is obtained, automatically at information
Reason method and the method verified automatically are analyzed and are summarized, and are established financial Heterogeneous Information and are handled frame, will be based on Heterogeneous Information
Financial knowledge services each stage of processing, modularization provide ginseng for the processing of financial field Heterogeneous Information and financial knowledge services
It examines, establishes financial Knowledge Service Platform, provide timely, accurate Financial Information service and personalized support for user.
It in the present embodiment, after the isomeries Financial Informations such as webpage, figure, table in internet are obtained, extracts, processing is half hitch
Structure data are further excavated using the methods of classification, cluster, time series analysis, discovery data, text variation relation, shape
It is supplied to user at structural data, verifying directly or indirectly is carried out to information by Information Authentication method, so as to further
Perfect information processing method.
In the present embodiment, by internet and system connectivity, related data passes through for all kinds of Financial Informations, rules and regulations information
Interface enters system, the customer information of enterprises and business transaction in Intranet directly and system connectivity, with structure
A complete corporate information technology service system is built, in the publication of service and the interaction of end user, can be taken by financing
Business personnel directly pass through counter terminal, or by internet and system connectivity when making house calls；It can also be passed through by client mutual
Networking connection service platform, directly and system mutual, for the safety for guaranteeing system, in service publication and user interaction,
System set up in internet together with firewall.
In the present embodiment, as shown in Figure 1, financial service method includes the following steps:
Step 100, Heterogeneous Information obtain automatically, obtain automatically from data sources such as internet, Xinhua Finance Media and listed companies
The data and text information of various structures；
Step 200, information processing carry out semi-structured processing to the Heterogeneous Information obtained automatically, it is real-time to be supplied to user
The basic datas such as market data, hot spot information, return of security investment data, while specialty analysis teacher or institutional investment are provided for user
Industry analysis report, prediction data that person issues etc.；
Step 300, development of judgment matrix compare Multiple factors index in item layer, two-by-two using step analysis
Method determines its weight, development of judgment matrix；
Step 400, Information Authentication guarantee according to weight calculation as a result, verifying to financial Heterogeneous Information processing result
The validity of the accuracy of data and financial knowledge services.
In the present embodiment, the Heterogeneous Information in step 100 obtains automatically is divided into two parts: automatic acquisition of scientific information and information are taken out
It takes, after obtaining Heterogeneous Information and being pre-processed, by the information obtained automatically by the processing such as information extraction, feature extraction, obtains
To the useful information of structuring.
In the step 100, further includes:
Firstly, building power field text, is parsed by the OWL file to description ontology, obtains decimation rule, it will be general
Thought, relationship, hierarchical structure etc. parse deposit database；
Then, using dictionary editing machine record, editing conception, the corresponding keyword of relationship, decimation rule is carried out perfect；
The Web information of acquisition is pre-processed again；
Finally, removing the unwanted contributions in sentence by diathesis alternation.
Web information carries out pretreated method in the step 100 are as follows:
Step 101, the amendment weighted value using Information Security Risk, to reflect the conduction effect of Information Security Risk；
Step 102 assumes that assets A is made of m functional module, and respectively A1, A2, A3 ... Am, wherein Ai is assets A's
I-th of functional module, similarly, it is assumed that assets B is made of n functional module, respectively B1, B2, B3 ... Bm, and wherein Bj is assets
J-th of functional module of B；
Step 103, using certain functional module Ai on A to the access relation of certain functional module Bj on assets B, attack assets
B；
Step 104, the risk that modules A i to module Bj is formed according to the dependence between Ai and Bj conduct RC, are denoted as
RCAi:Bj；
Step 105, RCAi:Bj risk information use<A, i, B, j, P, W>hexa-atomic group indicate, wherein P is indicated using should
The probability of happening for the security incident that risk conduction model calculates also becomes weight, P ∈ [0,1]；W indicates a functional module pair
The influence degree of another functional module, W ∈ [0 ,+∞].
In the present embodiment, the final goal to the security risk of financial services system is the security risk value of output quantization,
It just needs to carry out quantitative research, groundwork to safety wind yin conduction effect thus include two aspect: one is to for finance
The instantiation of service system assets dependence is analyzed, on the basis of three kinds of dependences of parallel, sequence and bad conjunction, information skill
The equipment connection and functional module mode that art field is got used to refine to dependence existing for assets and retouch fan and patrol
Collect control；Second is that determining the wind in various dependences using expert graded on the basis of the assets dependence of instantiation
Dangerous conduction system numerical value, with the conduction effect of quantificational description security risk.
In the present embodiment, Web information is pre-processed in step 100, can be according to the field feature of information to be extracted, rule
Mould, Information heterogeneity degree etc. are selected, and are realized information extraction, are realized the structurizing process of non-structural information substantially, just
The relationship discovery to data and text can be achieved and excavate.
The step 200 further include:
Firstly, carrying out text classification to Heterogeneous Information:
Training set D={ (x equipped with n samplei, yi), x ∈ Rn, i=1,2,3...n, y ∈ [- 1,1], training D can
With right-on separated by certain hyperplane, and the distance between make two class samples and hyperplane and maximize；
Secondly, the characteristics of data object is divided into cluster, is had to cluster according to similarity is analysed in depth and is summarized:
Given news agregator X={ x1, x2... xn, for each text xiComprising m feature f, then xi={ fi1,
fi2... fim, i=1,2 ..., n obtain cluster result C=by feature calculation similarity, and according to Clustering Decision-Making standard
{c1, c2..., ck, and cluster result meets: ci∩cj,Wherein, i ≠ j, i, j=1,
2 ..., n；
Finally, establishing the dynamic relation of interdependence model in reflecting time series between data according to historical record, pass through
The model predicts future trend.
In the present embodiment, related news event can be found by unusual fluctuation finance data using clustering algorithm, analyzes unusual fluctuation
Reason, and topic tracking is carried out to highlight, time using on-line talking algorithm, so that further analysis text will logarithm
According to the influence of generation, before being clustered to news, the standardization of time phrase and topic time window determine also be important it is pre-
Treatment process also will have a direct impact on Clustering Effect.
The step 300 further include:
Step 301, the product M for calculating each row element of judgment matrix Ai；
Step 302 calculates MiN times root Wi；
Step 303, to resulting phasor W={ W1, W2, W3Standardize, make ∑ Wi=1 (i=1,2,3), then W is
For required weight.
In the step 400 further include:
Step 401 introduces RI index, calculates the maximum eigenvalue of judgment matrix
Step 402 calculates random index CI,
Step 403 calculates random consistency ratio CR,
Need to find the optimum state sequence of each sub-stage in the step 400:
Firstly, ratio of profit increase of the gain network node based on original capital is calculated, the gain that each division beneficiating process obtains
Amount isOriginal capital isThe then node S in division gain networkijRatio of profit increase relative to original capital isIt is located at gain path interior joint SijWeight be Δ ωij, then
Secondly, determining separate unit as sub-stage；
Again, the optimal path of sub-stage is calculated, the state of each node is three-dimensional vector Si(ΔSi, ti, θi), wherein ti
It is in node SiThe time distributed, θiIt is corresponding gain recovery time, the optimal of sub-stage is calculated by class dijkstra's algorithm
Path；
Furthermore record each gain node amountCorresponding gain time of return θ i, obtains optimal road through the above steps
DiameterWith optimum state sequenceAccording to gain time of return θiIt calculates
The optimum gain amount of sub-stage out
Finally, searching again for next independent sub-stage through the above steps and calculating amount of gain, terminated until the gain time,
End loop.
In the present embodiment, independence in time is distributed according to division gain network interior joint, a NDUGP is divided
It then, is adopted according to the similitude of each subprocess for multiple sub-stages firstly, finding the optimum state sequence of each sub-stage
Entire NDUGP is solved with recursive algorithm.
In the present embodiment, when calculating amount of gain according to optimum state sequence, such as in gain model, due to group
Hair and recovery time have intersection, acquired optimum state sequence S*In comprising having distributed before stop time point do not obtain gain
Node, therefore, it is necessary to classify to the method for calculating given period critical point amount of gain, if the corresponding time sequence of beneficiating process
It is classified as t1, t2... tkAnd tkFor critical point, then from t1To tkAmount of gain Δ S calculation method can be divided into following three classes:
(1) in tkMoment does not have amount of gain to return to or ignore tkThe amount of gain returned after moment:
(2) by tkThe amount of gain returned after moment is included in Δ S in a manner of average gain:
Wherein, L tkThe gain number returned after moment；
(3) by tkThe amount of gain returned after moment is included in Δ S:
Wherein, Δ StδFor in tkIn t after the subset acquisition gain distributed away before momentδBefore moment (including tδWhen
Carve) return amount of gain.
In the present embodiment, optimum gain path is obtained through the above steps, and is calculated by critical point gain algorithm whole
The amount of gain of a beneficiating process is asked finally, being evaluated by GEC actual gain amount with solving entire N member division gain
Topic.
In the present embodiment, information is verified automatically can be divided into directly verifying and indirect verification, to the base of financial service platform
This information, present quotation data etc., can by the data with the announcements such as stock exchange official website, mainstream financial web site directly into
Row contrast verification；And for the information such as earning rate can by construction feature function carry out indirect verification, for example, by present price,
Interest rate etc. calculates the feature construction characteristic function of earning rate, passes through the verifying indirect verification earning rate to feature in function.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
Claims (8)
1. a kind of financial service method, characterized by the following steps:
Step 100, Heterogeneous Information obtain automatically, obtain automatically from data sources such as internet, Xinhua Finance Media and listed companies various
The data and text information of structure；
Step 200, information processing carry out semi-structured processing to the Heterogeneous Information obtained automatically, are supplied to user's present quotation
The basic datas such as data, hot spot information, return of security investment data, while specialty analysis teacher or institutional investor's hair are provided for user
The industry analysis report of cloth, prediction data etc.；
Step 300, development of judgment matrix compare Multiple factors index in item layer, using analytic hierarchy process (AHP), really two-by-two
Its fixed weight, development of judgment matrix；
Step 400, Information Authentication guarantee data according to weight calculation as a result, verifying to financial Heterogeneous Information processing result
Accuracy and financial knowledge services validity.
2. a kind of financial service method according to claim 1, it is characterised in that: in the step 100, further includes:
Firstly, building power field text, by description ontology OWL file parse, obtain decimation rule, by concept,
Relationship, hierarchical structure etc. parse deposit database；
Then, using dictionary editing machine record, editing conception, the corresponding keyword of relationship, decimation rule is carried out perfect；
The Web information of acquisition is pre-processed again；
Finally, removing the unwanted contributions in sentence by diathesis alternation.
3. a kind of financial service method according to claim 2, it is characterised in that: Web information carries out in the step 100
Pretreated method are as follows:
Step 101, the amendment weighted value using Information Security Risk, to reflect the conduction effect of Information Security Risk；
Step 102 assumes that assets A is made of m functional module, and respectively A1, A2, A3 ... Am, wherein Ai is the i-th of assets A
A functional module, similarly, it is assumed that assets B is made of n functional module, respectively B1, B2, B3 ... Bm, and wherein Bj is assets B's
J-th of functional module；
Step 103, using certain functional module Ai on A to the access relation of certain functional module Bj on assets B, attack assets B；
Step 104, the risk that modules A i to module Bj is formed according to the dependence between Ai and Bj conduct RC, are denoted as RCAi:
Bj；
Step 105, RCAi:Bj risk information use<A, i, B, j, P, W>hexa-atomic group indicate, wherein P indicates to utilize the risk
The probability of happening for the security incident that conduction model calculates also becomes weight, P ∈ [0,1]；W indicates a functional module to another
The influence degree of a functional module, W ∈ [0 ,+∞].
4. a kind of financial service method according to claim 1, it is characterised in that: the step 200 further include:
Firstly, carrying out text classification to Heterogeneous Information:
Training set D={ (x equipped with n samplei, yi), x ∈ Rn, i=1,2,3...n, y ∈ [- 1,1], training D can be by
Certain hyperplane is right-on separated, and the distance between makes two class samples and hyperplane and maximize；
Secondly, the characteristics of data object is divided into cluster, is had to cluster according to similarity is analysed in depth and is summarized:
Given news agregator X={ x1, x2... xn, for each text xiComprising m feature f, then xi={ fi1, fi2,
...fim, i=1,2 ..., n obtain cluster result C={ c by feature calculation similarity, and according to Clustering Decision-Making standard1,
c2..., ck, and cluster result meets: ci∩cj,Wherein, i ≠ j, i, j=1,2 ...,
n；
Finally, establishing the dynamic relation of interdependence model in reflecting time series between data according to historical record, pass through the mould
Type predicts future trend.
5. a kind of financial service method according to claim 1, it is characterised in that: the step 300 further include:
Step 301, the product M for calculating each row element of judgment matrix Ai；
Step 302 calculates MiN times root Wi；
Step 303, to resulting phasor W={ W1, W2, W3Standardize, make ∑ Wi=1 (i=1,2,3), then W is required
Weight.
6. a kind of financial service method according to claim 1, it is characterised in that: in the step 400 further include:
Step 401 introduces RI index, calculates the maximum eigenvalue of judgment matrix
Step 402 calculates random index CI,
Step 403 calculates random consistency ratio CR,
7. a kind of financial service method according to claim 1, it is characterised in that: need to find in the step 400 every
The optimum state sequence of a sub-stage:
Firstly, calculating ratio of profit increase of the gain network node based on original capital, the amount of gain that each division beneficiating process obtains is
Original capital isThe then node S in division gain networkijRatio of profit increase relative to original capital isIt is located at gain path interior joint SijWeight be Δ ωij, then
Secondly, determining separate unit as sub-stage；
Again, the optimal path of sub-stage is calculated, the state of each node is three-dimensional vector Si(ΔSi, ti, θi), wherein tiBe
Node SiThe time distributed, θiIt is corresponding gain recovery time, the optimal road of sub-stage is calculated by class dijkstra's algorithm
Diameter；
Furthermore record each gain node amountCorresponding gain time of return θi, optimal path is obtained through the above stepsWith optimum state sequenceAccording to gain time of return θiIt calculates
The optimum gain amount of sub-stage
Finally, searching again for next independent sub-stage through the above steps and calculating amount of gain, terminate, terminates until the gain time
Circulation.
8. a kind of financial service terminal for claim 1 the method, it is characterised in that: including financial service platform, enterprise
Industry database and work station；
The signal end of the work station establishes financial service Heterogeneous Information library by service subsystem, and passes through wireless network and gold
Melt service platform to be connected, the data terminal of the financial service platform and enterprise's numerical control library interconnect；
The financial service platform for extract Financial Information publication and law rules and regulations ordinances information carry out dissection process, from
Information is extracted in enterprise database and carries out coincidence matching, obtains matching result；
The data terminal of the work station is connected with personal computer, and the data terminal of the personal computer passes through wireless network connection
There are knowledge services to customize end, knowledge services customization end is internally provided with service type definition module；
The signal end at knowledge services customization end is connected with server by knowledge services analytical algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788831.8A CN109063045A (en) | 2018-07-18 | 2018-07-18 | A kind of financial service method and financial service terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810788831.8A CN109063045A (en) | 2018-07-18 | 2018-07-18 | A kind of financial service method and financial service terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063045A true CN109063045A (en) | 2018-12-21 |
Family
ID=64817124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810788831.8A Pending CN109063045A (en) | 2018-07-18 | 2018-07-18 | A kind of financial service method and financial service terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063045A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110674943A (en) * | 2019-09-16 | 2020-01-10 | 上海云从企业发展有限公司 | Financial knowledge network management method, system, medium and equipment |
| CN110838060A (en) * | 2019-10-30 | 2020-02-25 | 成都数融科技有限公司 | Financial risk measurement method and device and electronic equipment |
| CN113628032A (en) * | 2021-08-12 | 2021-11-09 | 上海上湖信息技术有限公司 | Method and device for determining user relationship |
| CN113781239A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Policy determination method and device, electronic equipment and storage medium |
| CN117729264A (en) * | 2024-02-08 | 2024-03-19 | 神州数码融信云技术服务有限公司 | Digital financial service mass information transmission method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964730A (en) * | 2010-01-28 | 2011-02-02 | 北京邮电大学 | Network vulnerability evaluation method |
| CN106934071A (en) * | 2017-04-27 | 2017-07-07 | 北京大学 | Recommendation method and device based on Heterogeneous Information network and Bayes's personalized ordering |
-
2018
- 2018-07-18 CN CN201810788831.8A patent/CN109063045A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101964730A (en) * | 2010-01-28 | 2011-02-02 | 北京邮电大学 | Network vulnerability evaluation method |
| CN106934071A (en) * | 2017-04-27 | 2017-07-07 | 北京大学 | Recommendation method and device based on Heterogeneous Information network and Bayes's personalized ordering |
Non-Patent Citations (3)
| Title |
|---|
| 牟伟明: "《供应链金融信用风险评价》", 《财会月刊》 * |
| 舒卫平: "《基于传导效应的大型金融服务系统安全风险评估方法研究》", 《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）2016年第03期》 * |
| 黄冬: "《面向网络金融知识服务的模型与方法研究》", 《中国博士学位论文全文数据库 信息科技辑（月刊）2016年第02期》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110674943A (en) * | 2019-09-16 | 2020-01-10 | 上海云从企业发展有限公司 | Financial knowledge network management method, system, medium and equipment |
| CN110838060A (en) * | 2019-10-30 | 2020-02-25 | 成都数融科技有限公司 | Financial risk measurement method and device and electronic equipment |
| CN113628032A (en) * | 2021-08-12 | 2021-11-09 | 上海上湖信息技术有限公司 | Method and device for determining user relationship |
| CN113628032B (en) * | 2021-08-12 | 2024-02-09 | 上海上湖信息技术有限公司 | Method and device for determining user relationship |
| CN113781239A (en) * | 2021-09-10 | 2021-12-10 | 未鲲(上海)科技服务有限公司 | Policy determination method and device, electronic equipment and storage medium |
| CN117729264A (en) * | 2024-02-08 | 2024-03-19 | 神州数码融信云技术服务有限公司 | Digital financial service mass information transmission method |
| CN117729264B (en) * | 2024-02-08 | 2024-04-26 | 神州数码融信云技术服务有限公司 | Digital financial service mass information transmission method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063045A (en) | A kind of financial service method and financial service terminal | |
| CN108629633A (en) | A kind of method and system for establishing user's portrait based on big data | |
| CN107977798A (en) | A kind of risk evaluating method of e-commerce product quality | |
| CN106067094A (en) | A kind of dynamic assessment method and system | |
| CN107368718A (en) | A kind of user browsing behavior authentication method and system | |
| CN112989761B (en) | Text classification method and device | |
| CN110751361B (en) | Bank demand entry level management method and system | |
| CN101819585A (en) | Device and method for constructing forum event dissemination pattern | |
| CN109325860A (en) | Network public-opinion detection method and system for overseas investment Risk-warning | |
| CN109472462A (en) | A kind of project risk ranking method and device based on the fusion of multi-model storehouse | |
| CN107808246A (en) | The intelligent evaluation method and system of collage-credit data | |
| CN114021970A (en) | Enterprise data asset model construction method based on data middlebox | |
| CN112396437A (en) | Trade contract verification method and device based on knowledge graph | |
| CN108564465A (en) | A kind of enterprise credit management method | |
| CN109359299A (en) | A kind of internet of things equipment ability ontology based on commodity data is from construction method | |
| CN112347254A (en) | News text classification method and device, computer equipment and storage medium | |
| CN115187066A (en) | Risk identification method and device, electronic equipment and storage medium | |
| CN110222180A (en) | A kind of classification of text data and information mining method | |
| CN107798137B (en) | A kind of multi-source heterogeneous data fusion architecture system based on additive models | |
| CN113610626A (en) | Bank credit risk identification knowledge graph construction method and device, computer equipment and computer readable storage medium | |
| CN113378024A (en) | Deep learning-based public inspection field-oriented related event identification method | |
| CN117290508A (en) | Post-loan text data processing method and system based on natural language processing | |
| CN106991171A (en) | Topic based on Intelligent campus information service platform finds method | |
| CN107609921A (en) | A kind of data processing method and server | |
| CN116485185A (en) | Enterprise risk analysis system and method based on comparison data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |
|
| RJ01 | Rejection of invention patent application after publication |