CN108628863A - Information acquisition method and device - Google Patents
Information acquisition method and device Download PDFInfo
- Publication number
- CN108628863A CN108628863A CN201710153107.3A CN201710153107A CN108628863A CN 108628863 A CN108628863 A CN 108628863A CN 201710153107 A CN201710153107 A CN 201710153107A CN 108628863 A CN108628863 A CN 108628863A
- Authority
- CN
- China
- Prior art keywords
- information
- financial
- ratio
- default
- events
- 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.)
- Granted
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
Abstract
This application discloses information acquisition methods and device.One specific implementation mode of this method includes：The information of the corresponding entity object of financial object is obtained, and extracts the keyword in information；Keyword is input to logic of propositions regression model, obtains output result, wherein logic of propositions regression model is trained based on the characteristic information for advancing with multiple financial objects and is generated；Based on output as a result, generating whether the financial object of instruction can occur the instruction information of default financial events.It realizes without relying on financial report data, it is only necessary to extract keyword from the information letter of such as news, pass through Logic Regression Models, you can complete financial object whether can occur the prediction of default financial events, to provide prediction result in time.
Description
Technical field
This application involves computer realms, and in particular to data analysis field more particularly to information acquisition method and device.
Background technology
Whether financial object (such as bond) can be occurred financial events (such as event of default) to carry out prediction to be that finance is right
The most key link in the investment decision of elephant.Currently, the mode of generally use is：According to the corresponding entity object of financial object
Financial report data, predict whether financial object can occur financial events.
However, since financial report data have hysteresis quality, lead to not obtain whether financial object can occur finance in time
The prediction result of event, and then influence the investment decision of financial object.
Invention content
This application provides information acquisition methods and device, are asked for solving technology existing for above-mentioned background technology part
Topic.
In a first aspect, this application provides information acquisition method, this method includes：Obtain the corresponding entity pair of financial object
The information of elephant, and extract the keyword in information；Keyword is input to logic of propositions regression model, is obtained
Export result, wherein logic of propositions regression model is trained based on the characteristic information for advancing with multiple financial objects and is given birth to
At characteristic information includes：Whether the financial object of instruction occurred to preset the markup information of financial events, the corresponding reality of financial object
Default financial events keyword in the information of body object；Based on output as a result, generating whether the financial object of instruction can be sent out
The instruction information of raw default financial events.
Second aspect, this application provides information acquisition device, which includes：Acquiring unit is configured to obtain gold
Melt the information of the corresponding entity object of object, and extracts the keyword in information；Predicting unit is configured to
Keyword is input to logic of propositions regression model, obtains output result, wherein logic of propositions regression model is based on advancing with
The characteristic information of multiple finance objects is trained and generates, and characteristic information includes：Whether the financial object of instruction occurred to preset
Default financial events keyword in the markup information of financial events, the information of the corresponding entity object of finance object；It is raw
At unit, it is configured to based on output as a result, generating whether the financial object of instruction can occur the instruction information of default financial events.
Information acquisition method and device provided by the present application, the information by obtaining the corresponding entity object of financial object are believed
Breath, and extract the keyword in information；Keyword is input to logic of propositions regression model, exported as a result,
Wherein, logic of propositions regression model is trained based on the characteristic information for advancing with multiple financial objects and is generated；Based on defeated
Go out as a result, generating whether the financial object of instruction can occur the instruction information of default financial events.It realizes without relying on financial report number
According to, it is only necessary to extract keyword from the information letter of such as news, pass through Logic Regression Models, you can complete to financial object whether
The prediction that default financial events can occur, to provide prediction result in time.
Description of the drawings
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is the exemplary system architecture figure that can be applied to the information acquisition method or device of the application；
Fig. 2 shows the flow charts according to information acquisition method one embodiment of the application；
Fig. 3 shows an exemplary process diagram of structure logic of propositions regression model；
Fig. 4 shows the structural schematic diagram of one embodiment of the information acquisition device according to the application；
Fig. 5 is adapted for the structural schematic diagram of the server of the information acquisition method for realizing the embodiment of the present application.
Specific implementation mode
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, is illustrated only in attached drawing and invent relevant part with related.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the exemplary system architecture of the information acquisition method or device that can be applied to the application.
As shown in Figure 1, system architecture may include server 101, network 102, server 103.Network 102 is taking
It is engaged in providing the medium of transmission link between device 101 and server 103.Server 103 can be to provide Internet resources such as finance and economics
The server of news.Server 101 may be used web crawlers and obtain what the Internet resources on server 103 were for example issued bond
The news of company.
Referring to FIG. 2, it illustrates the flow charts according to one embodiment of the information acquisition method of the application.This method
It can be executed by the server 101 in server such as Fig. 1, correspondingly, device can be set to the clothes in server such as Fig. 1
It is engaged in device 101.This approach includes the following steps：
Step 201, the information of the corresponding entity object of financial object is obtained, and extracts the pass in information
Keyword.
In the present embodiment, default financial events whether can occur for the financial object of prediction, it is right that finance can be obtained first
As the information of corresponding entity object.For example, financial object is bond, the non-event of default of financial events is preset, finance is right
As corresponding entity object is the company to issue bond, event of default whether can occur for prediction bond, information is distribution
The news of the company of bond can obtain the news of the company to issue bond first.
In the present embodiment, after getting information, the keyword in information, keyword can be extracted
Can entity object corresponding with financial object operation situation it is associated.Using financial object as bond, entity object is distribution
For the company of bond, in the news of the company to issue bond, including word associated with the operation situation of company.For example,
A development of projects for being related to the said firm described in the news of the progress of the said firm's investment project is slow, then can extract
Go out the keywords such as project name, progress, slow.
Step 202, keyword is input to logic of propositions regression model, obtains output result.
In the present embodiment, after extracting keyword by step 201, logic of propositions regression model base may be used
In the keyword extracted, predict whether financial object can occur default financial events.For example, financial object is bond, preset
Financial events are event of default, can predict whether bond can occur event of default according to the keyword extracted.
Using financial object as bond, financial events are preset as event of default, the spy of multiple bonds can be obtained in advance
Reference ceases, and the characteristic information of bond includes：Whether instruction bond occurred the corresponding entity of the markup information of event of default, bond
Default financial events keyword in the news for the company that object is for example issued bond.The feature of multiple bonds can be advanced with
Information is trained Logic Regression Models, obtains logic of propositions regression model.By after training, logic of propositions returns mould
Type can determine the weight i.e. regression coefficient of each default financial events keyword.Each default financial events keyword
Weight instruction presets financial events keyword for judging whether bond can occur the significance level of event of default.
In some optional realization methods of the present embodiment, the finance for occurring to preset financial events can be obtained in advance
The information of the corresponding entity object of object；Information is divided into multiple information sentences, and information sentence is carried out
Participle, obtains multiple words；Clustering is carried out to multiple words, obtains default financial events keyword.
Using financial object as bond, financial events are preset as that for event of default, can obtain in certain period of time in advance
Such as the news of the company of the bond of event of default occurred in 3 years, news is divided into multiple sentences, sentence is divided
After word, multiple words can be obtained.Clustering can be carried out to multiple words, obtained associated with event of default default
Financial events keyword.
In the present embodiment, after the keyword extracted by step 201 is input to logic of propositions regression model,
Logic of propositions regression model according to the weight i.e. regression coefficient of the default financial events keyword with the Keywords matching extracted,
Obtain output result.The output result of logic of propositions regression model, which can be the financial object of instruction, can occur default financial events
Probability.
In some optional realization methods of the present embodiment, following manner may be used and build logic of propositions recurrence in advance
Model：Can construction logic regression model first, obtain the characteristic information of multiple financial objects, and multiple characteristic informations are drawn
It is divided into for trained characteristic information and for the characteristic information of verification.In multiple finance objects, including occurring to preset finance
The financial object of event and the financial object for not occurring to preset financial events.The financial object of default financial events occurred
Characteristic information includes:The financial object of instruction occurred to preset the markup information of financial events, the corresponding entity pair of finance object
Default financial events keyword in the information of elephant.Do not occurred to preset the characteristic information packet of the financial object of financial events
It includes:The financial object of instruction did not occurred to preset the information of the markup information of financial events, the corresponding entity object of finance object
Default financial events keyword in information.
It, can be by feature when utilizing the characteristic information for the financial object of training to be trained Logic Regression Models
Numerical value of the markup information as dependent variable in information indicates for example, the markup information in the characteristic information of financial object is 1
Default financial events occurred, the markup information in the characteristic information of financial object is 0, and expression did not occurred to preset financial thing
Part.Using the default financial events keyword in characteristic information as the numerical value of independent variable, Logic Regression Models are trained, are obtained
Logic Regression Models after to training.It is crucial for each default financial events in the characteristic information of trained financial object
Word corresponds to a regression coefficient, and regression coefficient can indicate default financial events keyword for judging whether financial object can be sent out
The significance level of raw default financial events.
It is then possible to utilize the default financial events keyword in each characteristic information for the financial object of verification
The Logic Regression Models being input to after training, obtain whether the financial object of multiple instructions can occur the recurrence knot of default financial events
Fruit.Characteristic information for including the financial object for not occurring to preset financial events in the characteristic information of the financial object of verification
With the characteristic information for the financial object for occurring to preset financial events.It can will be in the characteristic information of the financial object of verification
The characteristic information that the financial object of default financial events occurred is referred to as fisrt feature information, by the financial object for verification
Characteristic information in do not occurred to preset the characteristic information of the financial object of financial events and be referred to as second feature information.
After obtaining multiple regression results, corresponding regression result and mark in all fisrt feature information can be calculated
The ratio is referred to as the first ratio by the quantitative proportion of the quantity and all fisrt feature information of the consistent fisrt feature information of information
Example.In other words, the first ratio for using the Logic Regression Models after training to be useful for verification the corresponding gold of characteristic information
Melt and occurred to preset the financial objects of financial events in object and default financial events whether can occur predicted, obtained recurrence
As a result it is that the quantity of default financial events can occur and occurred to preset the ratio of the total quantity of the financial object of financial events.
After obtaining multiple regression results, corresponding regression result and mark in all second feature information can be calculated
The ratio is referred to as the second ratio by the quantitative proportion of the quantity and all second feature information of the consistent second feature information of information
Example.In other words, the second ratio for using the Logic Regression Models after training to be useful for verification the corresponding gold of characteristic information
Melt and do not occurred to preset the financial objects of financial events in object and default financial events whether can occur predicted, what is obtained returns
It is that the quantity of default financial events will not occur and do not occurred to preset the total quantity of the financial object of financial events to sum up fruit
Ratio.
After calculating the first ratio and the second ratio, it can be determined that it is default whether the first ratio and the second ratio meet
Condition, preset condition include：First ratio and the second ratio are all higher than proportion threshold value, when meeting preset condition, can will instruct
Logic Regression Models after white silk are as logic of propositions regression model.
When calculated first ratio and the second ratio are unsatisfactory for preset condition, Bayes can be carried out to regression result
Analysis adjusts the regression coefficient of each default financial events keyword until meeting preset condition；By the default financial thing of adjustment
Logic Regression Models after the regression coefficient of part keyword are as logic of propositions regression model.
Referring to FIG. 3, it illustrates an exemplary process diagrams of structure logic of propositions regression model.
In this application, negative sample clustering can be analyzed first, obtains financial events keyword.For example, finance is right
As for bond, negative sample is the news in nearly 3 years of the company for the bond that event of default occurred, to the language in news
Sentence is segmented, and obtains appearing in multiple words in news.Clustering can be carried out to multiple words, obtain multiple finance
Event keyword.
The Logic Regression Models based on financial events keyword are established, the probability value between 0-1 is exported.Based on financial events
In the Logic Regression Models of keyword, each financial events keyword corresponds to a regression coefficient, which can be with table
Show default financial events keyword for judging whether financial object can occur the significance level of default financial events.When judgement is worked as
When whether preceding finance object can occur default financial events, it can will believe from the information of the corresponding entity object of current financial object
Keyword associated with entity object operation state that is being extracted in breath is input to Logic Regression Models, Logic Regression Models
The probability of default financial events can occur for output instruction current financial object.
Bayesian analysis is carried out to regression result, and regression coefficient is adjusted.When the Logic Regression Models of foundation
When output result is undesirable, Bayesian analysis can be carried out to regression result, to presetting the regression coefficient of financial events keyword
It is adjusted.
Step 203, based on output as a result, generating whether the financial object of instruction can occur the instruction letter of default financial events
Breath.
In the present embodiment, the keyword of financial object is being input to by logic of propositions regression model by step 202, obtained
It, can be according to output as a result, generating whether the financial object of instruction can occur the instruction of default financial events to after output result
Information.For example, the output result of logic of propositions regression model is the probability that default financial events can occur for the financial object of instruction, it can
According to the probability, to determine whether finance object can occur default financial events, it is pre- whether the financial object of generation instruction can occur
If the instruction information of financial events.
In some optional realization methods of the present embodiment, when the output result of logic of propositions regression model is instruction gold
The probability of default financial events can be occurred by melting object, can when the probability of logic of propositions regression model output is more than probability threshold value
To generate the instruction information for indicating that default financial events can occur for financial object.When the probability of logic of propositions regression model output is small
When probability threshold value, can generate the financial object of instruction will not occur the instruction information of default financial events.
Using financial object as bond, financial events are preset for event of default, to be exported when logic of propositions regression model
When probability is more than probability threshold value, can generate instruction bond can occur the instruction information of event of default.When logic of propositions returns mould
When the probability of type output is less than probability threshold value, can generate instruction bond will not occur the instruction information of event of default.
Referring to FIG. 4, it illustrates the structural schematic diagram according to one embodiment of the information acquisition device of the application, letter
Ceasing acquisition device includes：Acquiring unit 401, predicting unit 402, generation unit 403.Wherein, acquiring unit 401 is configured to obtain
The information of the corresponding entity object of financial object is taken, and extracts the keyword in information；Predicting unit 402 is matched
It sets for keyword to be input to logic of propositions regression model, obtains output result, wherein logic of propositions regression model is based on pre-
It is trained and generates first with the characteristic information of multiple financial objects, characteristic information includes：Whether the financial object of instruction occurs
The default financial events crossed in the markup information of default financial events, the information of the corresponding entity object of financial object are crucial
Word；Generation unit 403 is configured to based on output as a result, generating whether the financial object of instruction can occur the finger of default financial events
Show information.
In some optional realization methods of the present embodiment, generation unit 403 includes：Indicate that information generates subelement
(not shown) is configured to when output result be the probability that default financial events can occur for the financial object of instruction, when probability is more than
When probability threshold value, the instruction information of default financial events can be occurred by generating the financial object of instruction；When probability is less than probability threshold value,
The instruction information of default financial events will not be occurred by generating the financial object of instruction.
In some optional realization methods of the present embodiment, information acquisition device further includes：First model generation unit
(not shown) is configured to construction logic regression model；The characteristic information of multiple financial objects is obtained, and multiple features are believed
Breath divides the characteristic information for the characteristic information of training and for verification；Using for trained characteristic information, logic is returned
Model is returned to be trained, the Logic Regression Models after being trained, wherein each is for default in trained characteristic information
Financial events keyword corresponds to a regression coefficient；It determines in the characteristic information for verification and occurred comprising the financial object of instruction
It presets the first quantity of the fisrt feature information of the markup information of financial events and did not occurred to preset comprising the financial object of instruction
Second quantity of the second feature information of the markup information of financial events；It will be default in each characteristic information for verification
Financial events keyword is input to the Logic Regression Models after training, obtains whether the financial object of multiple instructions can occur default gold
Melt the regression result of event；Calculate the first ratio and the second ratio, wherein the first ratio is that corresponding regression result is believed with mark
The ratio of the quantity and the first quantity of consistent fisrt feature information is ceased, the second ratio is corresponding regression result and markup information
The ratio of the quantity and the second quantity of consistent second feature information；Judge whether the first ratio and the second ratio meet default item
Part, preset condition include：First ratio and the second ratio are all higher than proportion threshold value；It is preset when the first ratio and the second ratio meet
When condition, using the Logic Regression Models after training as logic of propositions regression model；Second model generation unit (not shown), matches
It sets for when the first ratio and the second ratio are unsatisfactory for preset condition, carrying out Bayesian analysis to regression result, adjustment is each
The regression coefficient of a default financial events keyword is until meet preset condition；Adjustment is preset to the recurrence of financial events keyword
Logic Regression Models after coefficient are as logic of propositions regression model；Keyword acquiring unit (not shown), is configured to obtain
The information of the corresponding entity object of financial object of default financial events occurred；Information is divided into multiple information
Sentence, and information sentence is segmented, obtain multiple words；Clustering is carried out to multiple words, obtains default finance
Event keyword.
Present invention also provides a kind of server, which may include information acquisition device described in Fig. 4.The clothes
Business device can be configured with one or more processors；Memory, for storing one or more programs, in one or more programs
Can include to execute the instruction of the operation described in above-mentioned steps 201-203.When one or more programs are by one or more
When a processor executes so that one or more processors execute the operation described in above-mentioned steps 201-203.
Fig. 5 shows the structural schematic diagram of the server of the information acquisition method suitable for being used for realizing the embodiment of the present application.
As shown in figure 5, include central processing unit (CPU) 501, it can be according to being stored in read-only memory (ROM) 502
In program or executed various appropriate from the program that storage section 508 is loaded into random access storage device (RAM) 503
Action and processing.CPU 501, ROM502 and RAM 503 are connected with each other by bus 504.Input/output (I/O) interface 505
It is also connected to bus 504.
It is connected to I/O interfaces 505 with lower component：Importation 506；Output par, c 507；Storage section including hard disk etc.
508；And the communications portion 509 of the network interface card including LAN card, modem etc..Communications portion 509 is via all
As the network of internet executes communication process.Driver 510 is also according to needing to be connected to I/O interfaces 505.Detachable media 511,
Such as disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on driver 510, as needed in order to from thereon
The computer program of reading is mounted into storage section 508 as needed.
The process of above-mentioned each step description in the application may be implemented as computer program.The computer program can
To carry on a computer-readable medium, which includes the instruction for method shown in execution flow chart.The meter
Calculation machine program can be downloaded and installed by communications portion 509 from network, and/or be mounted from detachable media 511.
Present invention also provides a kind of computer-readable medium, which can be included in server
's；Can also be individualism, without in supplying server.The computer-readable medium carries one or more program,
When one or more program is executed by the server so that the server：Obtain the corresponding entity object of financial object
Information, and extract the keyword in information；Keyword is input to logic of propositions regression model, is exported
As a result, wherein logic of propositions regression model is trained based on the characteristic information for advancing with multiple financial objects and is generated, special
Reference ceases：Whether the financial object of instruction occurred to preset the markup information of financial events, the corresponding entity pair of financial object
Default financial events keyword in the information of elephant；Based on output as a result, generate the financial object of instruction whether can occur it is pre-
If the instruction information of financial events.
It should be noted that above computer readable medium can be computer-readable signal media or computer-readable
Storage medium either the two arbitrarily combines.Computer readable storage medium for example may be-but not limited to-
System, device or the device of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or the arbitrary above combination.It is computer-readable
The more specific example of storage medium can include but is not limited to：Electrical connection, portable computing with one or more conducting wires
Machine disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM
Or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or above-mentioned
Any appropriate combination.In this application, computer readable storage medium can be any include or storage program it is tangible
Medium, the program can be commanded the either device use or in connection of execution system, device.And in this application,
Computer-readable signal media may include in a base band or as the data-signal that a carrier wave part is propagated, wherein carrying
Computer-readable program code.The data-signal of this propagation can be diversified forms, including but not limited to electromagnetic signal,
Optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer readable storage medium with
Outer any computer-readable medium, the computer-readable medium can be sent, propagated or transmitted for by instruction execution system
System, device either device use or program in connection.The program code for including on computer-readable medium can be with
It transmits with any suitable medium, including but not limited to：Wirelessly, electric wire, optical cable, RF etc. or above-mentioned any appropriate
Combination.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art
Member should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the design of the application, it is carried out by above-mentioned technical characteristic or its equivalent feature
Other technical solutions of arbitrary combination and formation.Such as features described above has similar work(with (but not limited to) disclosed herein
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of information acquisition method, which is characterized in that the method includes：
The information of the corresponding entity object of financial object is obtained, and extracts the keyword in the information；
The keyword is input to logic of propositions regression model, obtains output result, wherein logic of propositions regression model is based on
The characteristic information for advancing with multiple financial objects is trained and generates, and characteristic information includes：Whether the financial object of instruction is sent out
The default financial events given birth in the markup information of default financial events, the information of the corresponding entity object of financial object are closed
Keyword；
Based on output as a result, generating whether the financial object of instruction can occur the instruction information of default financial events.
2. according to the method described in claim 1, it is characterized in that, the output result is to indicate that financial object can be preset
The probability of financial events；And
Indicate that the instruction information whether financial object can occur default financial events includes as a result, generating based on output：
When the probability is more than probability threshold value, the instruction information of default financial events can be occurred by generating the financial object of instruction；
When the probability is less than probability threshold value, the instruction information of default financial events will not be occurred by generating the financial object of instruction.
3. according to the method described in claim 2, it is characterized in that, in the information letter for obtaining the corresponding entity object of financial object
Before breath, the method further includes：
Construction logic regression model；
The characteristic information of multiple financial objects is obtained, and multiple characteristic informations are divided into the characteristic information for training and are used for
The characteristic information of verification；
Using for trained characteristic information, Logic Regression Models are trained, the Logic Regression Models after being trained,
In, each corresponds to a regression coefficient for the default financial events keyword in trained characteristic information；
It determines in the characteristic information for verification and occurred to preset the of the markup information of financial events comprising the financial object of instruction
First quantity of one characteristic information and the second spy for not occurring to preset the markup information of financial events comprising the financial object of instruction
Second quantity of reference breath；
Default financial events keyword in each characteristic information for verification is input to the logistic regression mould after training
Type, obtains whether the financial object of multiple instructions can occur the regression result of default financial events；
Calculate the first ratio and the second ratio, wherein the first ratio is first consistent with markup information of corresponding regression result
The ratio of the quantity of characteristic information and the first quantity, the second ratio are corresponding regression result second spy consistent with markup information
The ratio of the quantity and the second quantity of reference breath；
Judge whether the first ratio and the second ratio meet preset condition, preset condition includes：First ratio and the second ratio are equal
More than proportion threshold value；
When the first ratio and the second ratio meet preset condition, returned the Logic Regression Models after training as logic of propositions
Model.
4. according to the method described in claim 3, it is characterized in that, the method further includes：
When the first ratio and the second ratio are unsatisfactory for the preset condition, Bayesian analysis is carried out to regression result, adjustment is every
The regression coefficient of one default financial events keyword is until meet preset condition；
Adjustment is preset to the Logic Regression Models after the regression coefficient of financial events keyword as logic of propositions regression model.
5. according to the method described in claim 4, it is characterized in that, in the information letter for obtaining the corresponding entity object of financial object
Before breath, the method further includes：
Obtain the information for the corresponding entity object of financial object for occurring to preset financial events；
The information is divided into multiple information sentences, and the information sentence is segmented, obtains multiple words；
Clustering is carried out to multiple words, obtains default financial events keyword.
6. a kind of information acquisition device, which is characterized in that described device includes：
Acquiring unit is configured to obtain the information of the corresponding entity object of financial object, and extracts the information
Keyword in information；
Predicting unit is configured to the keyword being input to logic of propositions regression model, obtains output result, wherein pre-
If Logic Regression Models are trained based on the characteristic information for advancing with multiple financial objects and are generated, characteristic information includes：
Indicate whether financial object occurred to preset the information of the markup information of financial events, the corresponding entity object of financial object
In default financial events keyword；
Generation unit is configured to based on output as a result, generating whether the financial object of instruction can occur the finger of default financial events
Show information.
7. device according to claim 6, which is characterized in that generation unit includes：
It indicates that information generates subelement, is configured to that default financial events can occur when the output result is the financial object of instruction
Probability, when the probability is more than probability threshold value, the instruction information of default financial events can be occurred by generating the financial object of instruction；
When the probability is less than probability threshold value, the instruction information of default financial events will not be occurred by generating the financial object of instruction.
8. device according to claim 7, which is characterized in that described device further includes：
First model generation unit, is configured to construction logic regression model；The characteristic information of multiple financial objects is obtained, and
Multiple characteristic informations are divided for the characteristic information of training and for the characteristic information of verification；Using for trained feature letter
Breath, is trained Logic Regression Models, the Logic Regression Models after being trained, wherein each is for trained feature
Default financial events keyword in information corresponds to a regression coefficient；It determines in the characteristic information for verification comprising instruction gold
Melt object to occur to preset the first quantity of the fisrt feature information of the markup information of financial events and comprising the financial object of instruction
Do not occurred to preset the second quantity of the second feature information of the markup information of financial events；By each feature for verification
Default financial events keyword in information is input to the Logic Regression Models after training, whether obtains the financial object of multiple instructions
The regression result of default financial events can occur；Calculate the first ratio and the second ratio, wherein the first ratio is corresponding recurrence
As a result the ratio of the quantity and the first quantity of fisrt feature information consistent with markup information, the second ratio are tied for corresponding recurrence
The ratio of the quantity and the second quantity of fruit and the consistent second feature information of markup information；Judging the first ratio and the second ratio is
No to meet preset condition, preset condition includes：First ratio and the second ratio are all higher than proportion threshold value；When the first ratio and second
When ratio meets preset condition, using the Logic Regression Models after training as logic of propositions regression model；
Second model generation unit is configured to when the first ratio and the second ratio are unsatisfactory for the preset condition, to returning
As a result Bayesian analysis is carried out, adjusts the regression coefficient of each default financial events keyword until meeting preset condition；It will
Logic Regression Models after the regression coefficient of the default financial events keyword of adjustment are as logic of propositions regression model；
Keyword acquiring unit is configured to obtain the money for the corresponding entity object of financial object for occurring to preset financial events
Interrogate information；The information is divided into multiple information sentences, and the information sentence is segmented, obtains multiple words
Language；Clustering is carried out to multiple words, obtains default financial events keyword.
9. a kind of server, which is characterized in that including：
One or more processors；
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors so that one or more of processors
Realize the method as described in any in claim 1-5.
10. a kind of readable computer storage medium, which is characterized in that be stored thereon with computer program, which is characterized in that the journey
The method as described in any in claim 1-5 is realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153107.3A CN108628863B (en) | 2017-03-15 | 2017-03-15 | Information acquisition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710153107.3A CN108628863B (en) | 2017-03-15 | 2017-03-15 | Information acquisition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108628863A true CN108628863A (en) | 2018-10-09 |
| CN108628863B CN108628863B (en) | 2021-07-20 |
Family
ID=63687386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710153107.3A Active CN108628863B (en) | 2017-03-15 | 2017-03-15 | Information acquisition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108628863B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110033284A (en) * | 2019-03-13 | 2019-07-19 | 平安城市建设科技(深圳)有限公司 | Source of houses verification method, apparatus, equipment and storage medium |
| CN111786802A (en) * | 2019-04-03 | 2020-10-16 | 北京嘀嘀无限科技发展有限公司 | Event detection method and device |
| CN112989165A (en) * | 2021-03-26 | 2021-06-18 | 杭州有数金融信息服务有限公司 | Method for calculating public opinion entity relevance |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135967A (en) * | 2010-01-27 | 2011-07-27 | 华为技术有限公司 | Webpage keywords extracting method, device and system |
| CN103176983A (en) * | 2011-12-20 | 2013-06-26 | 中国科学院计算机网络信息中心 | Event warning method based on Internet information |
| US20140344130A1 (en) * | 2001-01-30 | 2014-11-20 | Goldman, Sachs & Co. | Systems And Methods For Automated Political Risk Management |
| CN105528652A (en) * | 2015-12-03 | 2016-04-27 | 北京金山安全软件有限公司 | Method and terminal for establishing prediction model |
| CN106372961A (en) * | 2016-08-23 | 2017-02-01 | 北京小米移动软件有限公司 | Commodity recommendation method and device |
-
2017
- 2017-03-15 CN CN201710153107.3A patent/CN108628863B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140344130A1 (en) * | 2001-01-30 | 2014-11-20 | Goldman, Sachs & Co. | Systems And Methods For Automated Political Risk Management |
| CN102135967A (en) * | 2010-01-27 | 2011-07-27 | 华为技术有限公司 | Webpage keywords extracting method, device and system |
| CN103176983A (en) * | 2011-12-20 | 2013-06-26 | 中国科学院计算机网络信息中心 | Event warning method based on Internet information |
| CN105528652A (en) * | 2015-12-03 | 2016-04-27 | 北京金山安全软件有限公司 | Method and terminal for establishing prediction model |
| CN106372961A (en) * | 2016-08-23 | 2017-02-01 | 北京小米移动软件有限公司 | Commodity recommendation method and device |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110033284A (en) * | 2019-03-13 | 2019-07-19 | 平安城市建设科技(深圳)有限公司 | Source of houses verification method, apparatus, equipment and storage medium |
| CN111786802A (en) * | 2019-04-03 | 2020-10-16 | 北京嘀嘀无限科技发展有限公司 | Event detection method and device |
| CN111786802B (en) * | 2019-04-03 | 2023-07-04 | 北京嘀嘀无限科技发展有限公司 | Event detection method and device |
| CN112989165A (en) * | 2021-03-26 | 2021-06-18 | 杭州有数金融信息服务有限公司 | Method for calculating public opinion entity relevance |
| CN112989165B (en) * | 2021-03-26 | 2022-07-01 | 浙江有数数智科技有限公司 | Method for calculating public opinion entity relevance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108628863B (en) | 2021-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109493977B (en) | Text data processing method and device, electronic equipment and computer readable medium | |
| US11645517B2 (en) | Information processing method and terminal, and computer storage medium | |
| US9081411B2 (en) | Rapid development of virtual personal assistant applications | |
| CN105701088B (en) | The method and apparatus manually talked with are switched to from machine dialogue | |
| CN111859960A (en) | Semantic matching method and device based on knowledge distillation, computer equipment and medium | |
| CN111310436B (en) | Text processing method and device based on artificial intelligence and electronic equipment | |
| CN111104512B (en) | Game comment processing method and related equipment | |
| CN107506434A (en) | Method and apparatus based on artificial intelligence classification phonetic entry text | |
| CN112951203B (en) | Speech synthesis method, device, electronic equipment and storage medium | |
| CN112328761B (en) | Method and device for setting intention label, computer equipment and storage medium | |
| CN112860841A (en) | Text emotion analysis method, device and equipment and storage medium | |
| CN108111399B (en) | Message processing method, device, terminal and storage medium | |
| CN111428010A (en) | Man-machine intelligent question and answer method and device | |
| CN112863529B (en) | Speaker voice conversion method based on countermeasure learning and related equipment | |
| CN112463968B (en) | Text classification method and device and electronic equipment | |
| CN112395391B (en) | Concept graph construction method, device, computer equipment and storage medium | |
| CN108121699A (en) | For the method and apparatus of output information | |
| CN108628863A (en) | Information acquisition method and device | |
| CN109582788A (en) | Comment spam training, recognition methods, device, equipment and readable storage medium storing program for executing | |
| CN110457476A (en) | Method and apparatus for generating disaggregated model | |
| CN110543633A (en) | Sentence intention identification method and device | |
| CN110851650B (en) | Comment output method and device and computer storage medium | |
| CN110427454A (en) | Text mood analysis method and device, electronic equipment and non-transient storage media | |
| CN113220828B (en) | Method, device, computer equipment and storage medium for processing intention recognition model | |
| CN113821587A (en) | Text relevance determination method, model training method, device and storage medium |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20191122 Address after: 201210 room j1328, floor 3, building 8, No. 55, Huiyuan Road, Jiading District, Shanghai Applicant after: SHANGHAI YOUYANG NEW MEDIA INFORMATION TECHNOLOGY Co.,Ltd. Address before: 100085 Beijing, Haidian District, No. ten on the ground floor, No. 10 Baidu building, layer three Applicant before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20181009 Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. Assignor: SHANGHAI YOUYANG NEW MEDIA INFORMATION TECHNOLOGY Co.,Ltd. Contract record no.: X2020990000201 Denomination of invention: Network service description information acquisition method and network service description information acquisition device License type: Exclusive License Record date: 20200420 |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 401120 b7-7-2, Yuxing Plaza, No.5, Huangyang Road, Yubei District, Chongqing Patentee after: Chongqing duxiaoman Youyang Technology Co.,Ltd. Address before: 201210 room j1328, 3 / F, building 8, 55 Huiyuan Road, Jiading District, Shanghai Patentee before: SHANGHAI YOUYANG NEW MEDIA INFORMATION TECHNOLOGY Co.,Ltd. |