CN107402925A - Information-pushing method and device - Google Patents
Information-pushing method and device Download PDFInfo
- Publication number
- CN107402925A CN107402925A CN201610334269.2A CN201610334269A CN107402925A CN 107402925 A CN107402925 A CN 107402925A CN 201610334269 A CN201610334269 A CN 201610334269A CN 107402925 A CN107402925 A CN 107402925A
- Authority
- CN
- China
- Prior art keywords
- event
- event type
- type
- keyword
- present news
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a kind of information-pushing method and device, pass through the object event type being subordinate to according to Present News event, determine destination object corresponding with Present News event, according to history media event corresponding to object event type, influence force parameter of the object event type to destination object is determined, destination object and influence force parameter are pushed to user.What the present invention was got is the objects such as the product directly related with Present News event or service, and compared with prior art, ordinary individual can save analyzes the time that related object expends from the mass data for exist complex relationship.Further, due to ordinary individual and do not have professional experiences, the result accuracy obtained after analyzing mass data is poor, and the present invention directly provides the object such as Related product or service, improve the accuracy of analysis, can more meet the needs of ordinary individual, formulating capital project for ordinary individual provides significantly more efficient guiding or help.
Description
Technical field
The invention belongs to data processing field, more particularly to a kind of information-pushing method and device.
Background technology
In actual life, socialization event in part can affect for the tendency of some industries, for example, portion
Point media event can produce obvious front or negative effect to the product provided in some industries or service, and to these products
Or service shows metastable rule.
At present, using the free information on internet, Data Analysis Platform is established to analyze major issue to financial industry
Influence, such as external Kensho services, Kensho services are for the professional inside investment institution, by all history phases
Influence data of the pass event to financial industry show professional in a manner of enumerating, and professional needs rule of thumb to carry out
Analysis, the important evidence of financial scenario is formulated as mechanism.
Generally, ordinary individual is often interested just for portioned product or service, and ordinary individual lacks
The weary ability that the association between the specific products related to Present News event or service is found from the influence data enumerated,
For this characteristic of this ordinary individual, the analysis result of existing data analysing method can not meet the need of ordinary individual
Ask, it is impossible to extended efficient help for ordinary individual.
The content of the invention
The present invention provides a kind of information-pushing method and device, is for solving the existing analysis result to media event
The mass data of complex relationship be present, can not meet the needs of ordinary individual, it is impossible to extend efficient help for ordinary individual.
To achieve these goals, the invention provides a kind of information-pushing method, including：
Object event type according to belonging to Present News event, determine target pair corresponding to the Present News event
As；
According to history media event corresponding to the object event type, determine the object event type to the target
The influence force parameter of object；
The destination object and the influence force parameter are pushed to user.
To achieve these goals, the invention provides a kind of information push-delivery apparatus, including：
Object acquisition module, for the object event type according to belonging to Present News event, determine the Present News
Destination object corresponding to event；
Determining module, for the history media event according to corresponding to the object event type, determine the object event
Influence force parameter of the type to the destination object；
Pushing module, for the destination object and the influence force parameter to be pushed into user.
Information-pushing method and device provided by the invention, pass through the object event class being subordinate to according to Present News event
Type, the destination object related to Present News event is determined, according to history media event corresponding to object event type, it is determined that
Go out influence force parameter of the object event type to destination object, destination object and influence force parameter are pushed to user.This hair
It is bright that get is the objects such as the product directly related with Present News event or service, compared with prior art, ordinary individual
It can save and the time that related object expends is analyzed from the mass data for exist complex relationship.Further, due to common
People does not simultaneously have professional experiences, and the result accuracy obtained after analyzing mass data is poor, and the present invention directly provides phase
The object such as product or service is closed, the accuracy of analysis is improved, can more meet the needs of ordinary individual, formulates and throws for ordinary individual
Capital's case provides significantly more efficient guiding or help.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the information-pushing method of the embodiment of the present invention one；
Fig. 2 is the schematic flow sheet of the information-pushing method of the embodiment of the present invention two；
Fig. 3 is the schematic flow sheet of the information-pushing method of the embodiment of the present invention three；
Fig. 4 is the schematic flow sheet of the information-pushing method of the embodiment of the present invention four；
Fig. 5 is one of application schematic diagram of information-pushing method of the embodiment of the present invention four；
Fig. 6 is the two of the application schematic diagram of the information-pushing method of the embodiment of the present invention four；
Fig. 7 is the three of the application schematic diagram of the information-pushing method of the embodiment of the present invention four；
Fig. 8 is the four of the application schematic diagram of the information-pushing method of the embodiment of the present invention four；
Fig. 9 is the structural representation of the information push-delivery apparatus of the embodiment of the present invention five；
Figure 10 is the structural representation of the information push-delivery apparatus of the embodiment of the present invention six.
Embodiment
Information-pushing method provided in an embodiment of the present invention and device are described in detail below in conjunction with the accompanying drawings.
Embodiment one
As shown in figure 1, its schematic flow sheet for the information-pushing method of the embodiment of the present invention one.The information-pushing method
Comprise the following steps：
S101, the object event type according to belonging to Present News event, determine target pair corresponding to Present News event
As.
In actual life, and not all media event can all affect for the tendency of industry, and be formed
The rule of ensured sustained development by data mining and history from a large amount of history media events, it is necessary to return the technological means such as survey to news thing
Part is sorted out.
The basic event type identified at present includes socialization media event and company's bulletin class media event.Example
Such as, magnitude exceeds certain series and the density of population of scene can be used as a socialization news more than certain level
Event, bulletin is held in senior executive's increase and decrease and increasing and decreasing the ratio held can be used as a company to announce class media event more than certain level.
, it is necessary to according to the content of Present News event in the present embodiment, the target that Present News event is subordinate to is determined
Event type.Specifically, the keyword of Present News event can be got, according to keyword and the key of object event type
Word is matched, and determines the object event type of the current media event.
Further, according to influence of a large amount of history media events to industry, different event type institute shadows can be obtained
Specific product or service in loud industry, wherein these products or service can be used as object, can get with often
Object corresponding to individual event type., can be according to above-mentioned event class after the object event type of Present News event is determined
Corresponding relation between type and object, destination object corresponding with the object event type is obtained, i.e. Present News event is related
Destination object.
S102, the history media event according to corresponding to object event type, determine object event type to destination object
Influence force parameter.
After object event type is determined, history news thing included under the object event type can be got
Part, after all history media events are obtained, further obtain the influence to destination object after history media event occurs
Data, after the media event that can get object event type by analyzing these influence data occurs, to destination object
Influence power, wherein, influence power includes being embodied on the effect of destination object and on the time of influence.For example, work as mesh
When mark object is an investment product, the media event that can be analyzed under object event type is walked to the price of the investment product
The influence of gesture, and the duration that media event influences on the price trend of investment product.By object event type to target
The influence power of object is as affecting parameters.
S103, by destination object and influence force parameter be pushed to user.
, can be with after determining the history media event included by object event type to the affecting parameters of destination object
Destination object and history media event are pushed to user to the affecting parameters of destination object, in order to which user is according to target
The affecting parameters of object, formulate investment tactics.It is alternatively possible to after destination object is got, first destination object is pushed and used
Family, when user sounds out purpose object, the destination object can be chosen by operations such as clicks, now again by history news
Influence force parameter push user of the event to destination object.
The information-pushing method that the present embodiment provides, by the object event type being subordinate to according to Present News event,
The destination object related to Present News event is determined, according to history media event corresponding to object event type, is determined
Influence force parameter of the object event type to destination object, destination object and influence force parameter are pushed to user.The present invention
What is got is the objects such as the product directly related with Present News event or service, and compared with prior art, ordinary individual can
To save the time that related object is analyzed from the mass data for exist complex relationship and is expended.Further, due to ordinary individual
And not having professional experiences, the result accuracy obtained after analyzing mass data is poor, and the present invention directly provides correlation
The object such as product or service, the accuracy of analysis is improved, can more meet the needs of ordinary individual, investment is formulated for ordinary individual
Scheme provides significantly more efficient guiding or help.
Embodiment two
As shown in Fig. 2 its schematic flow sheet for the information-pushing method of the embodiment of the present invention two.In above-described embodiment one
Basis on, the object event type according to belonging to Present News event, determine corresponding to the Present News event
Destination object, comprise the following steps：
S201, by data mining to the classification of history media event handle.
In actual life, and not all media event can all affect for the tendency of industry, and be formed
The rule of ensured sustained development by data mining and history from a large amount of history media events, it is necessary to return the technological means such as survey to news thing
Part is sorted out.
The basic event type identified at present includes socialization media event and company's bulletin class media event.Example
Such as, magnitude exceeds certain series and the density of population of scene can be used as a socialization news more than certain level
Event, bulletin is held in senior executive's increase and decrease and increasing and decreasing the ratio held can be used as a company to announce class media event more than certain level.
S202, obtain basic event type, each type of foundation dictionary, each basic event corresponding to basic event type
Event type dictionary corresponding to event type and each event type included by type.
Specifically, in data mining process, keyword corresponding to each history media event is extracted, then by keyword
Intelligent Matching is carried out, during intelligent Matching, the similarity between keyword is obtained, obtains belonging to same foundation event class
All keywords of type, further, obtain the hyponymy between keyword, according to get between keyword up and down
All keywords are effectively sorted, obtain the type of foundation dictionary of the basic event type by position relation.
Similarly, during event classification is carried out using data mining to a large amount of history media events, can also obtain
Get the event type included by each basic event type, and event type dictionary corresponding to each event type.Specifically
Ground, above-mentioned matching degree highest crucial phrase cooperation is combined for optimal keyword, combined using optimal keyword to target base
Event type dictionary corresponding to all event types carries out intelligent Matching under plinth event type, obtains combining with optimal keyword
Matching degree highest event type, the object event type using the event type as Present News event.
S203, the time series for establishing each event type.
S204, the time series to each event type carry out history and return survey, obtain the related object of each event type,
Build the mapping relations between event type and object.
Event classification is carried out to media event using the mode of data mining to a large amount of history media events, got each
After event type included by basic event type, the time series of each event type is established, then to each event class
The time series of type carries out history and returns survey, obtains the related object of each event type, builds between event type and object
Mapping relations.
S205, destination object obtained according to mapping relations and object event type.
, can be according between above-mentioned event type and object after the object event type of Present News event is determined
Mapping relations, obtain destination object corresponding with the object event type, the i.e. related destination object of Present News event.
In the present embodiment, basic event type and corresponding type of foundation dictionary, basis are constructed by data mining
Event type and event type dictionary included by event type, returned and surveyed between acquisition event type and object by history
Mapping relations, so as to go out the event type that Present News event is subordinate to, root by the keyword intelligent Matching of Present News event
According to above-mentioned mapping relations, the destination object related to Present News event is determined.What the present embodiment was got be with it is current new
The object such as the directly related product of news event or service, compared with prior art, ordinary individual can save closes from the presence of complicated
The time that related object expends is analyzed in the mass data of system.Further, due to ordinary individual and do not have professional experiences, it is right
The result accuracy obtained after mass data analysis is poor, and the present invention directly provides the object such as Related product or service, carries
The high accuracy of analysis, can more meet the needs of ordinary individual, be provided for ordinary individual's formulation capital project significantly more efficient
Guiding helps.
Embodiment three
As shown in figure 3, its schematic flow sheet for the information-pushing method of the embodiment of the present invention three.In above-described embodiment
On basis, the object event type according to belonging to Present News event, target pair corresponding to the Present News event is determined
It is further comprising the steps of as before：
S301, the keyword for obtaining Present News event.
In the present embodiment, feature information extraction is carried out to the body matter of Present News event, with from Present News event
In extract corresponding to multiple keywords.It is, for example, possible to use closest (K-NearestNeighbor, the abbreviation KNN) texts of K
Sorting algorithm, feature selecting/SVMs text algorithm carry out feature information extraction to the body matter of Present News event.
For example, Present News event is " National Development and Reform Committee gave notice December 16 titles, the up-regulation petrol and diesel oil highest zero from December 17
Price lattice ", after carrying out keyword extraction to above-mentioned media event, obtain following keyword：National Development and Reform Committee, the Committee of Development and Reform, December
16 days, from December 17, raise, petrol and diesel oil, retail price etc..
S302, event type classification carried out to Present News event according to keyword, be subordinate to obtaining when media event
Object event type.
In the present embodiment, different basic event types is previously provided with, and construct for each basic event type
One type of foundation dictionary.The setting of basic event type and corresponding type of foundation dictionary, it is by largely going through
History media event carries out what data mining obtained.Specifically, in data mining process, it is corresponding to extract each history media event
Keyword, keyword is then subjected to intelligent Matching, during intelligent Matching, obtain keyword between similarity,
All keywords for belonging to same foundation event type are obtained, further, obtain the hyponymy between keyword, according to
The hyponymy between keyword is got, all keywords are effectively sorted, obtains the base of the basic event type
Plinth type dictionary.
Specifically, all keywords got in S301 are entered into any combination, the institute for obtaining Present News event is relevant
Key contamination, each keyword is then combined into type of foundation dictionary corresponding with basic event type and carries out intelligent Matching, will
Basic event type is as target basis event type corresponding to the combination of matching degree highest keyword.In the present embodiment, to closing
When keyword is combined with event type dictionary progress intelligent Matching, each keyword combination is mainly calculated in event using hash algorithm
Score in type dictionary, the crucial phrase cooperation of highest scoring is combined for matching degree highest keyword.
Each basic event type also includes multiple event types, for example, socialization media event can include：Industry is advised
Draw, industry adjusting and optimizing, great scientific and technological progress, burst/focus incident etc..It is in advance each basic event class in the present embodiment
Type sets multiple event types, is every in the present embodiment to identify the event type being subordinate to of different media events
Event type dictionary corresponding to individual event type setting.After each basic event type is got, to institute in basic event type
There is the keyword in history media event to carry out intelligent Matching, during intelligent Matching, obtain the semanteme between keyword
Similarity, all keywords for being under the jurisdiction of same event type can be obtained, obtain the hyponymy between keyword, according to
The hyponymy between keyword is got, all keywords are effectively sorted, obtains the event class of the event type
Type dictionary.
After the basic event type of Present News event is got, it is also necessary to determine mesh corresponding to Present News event person in servitude
Mark event type.Specifically, above-mentioned matching degree highest crucial phrase cooperation is combined for optimal keyword, utilizes optimal key
Word combination to corresponding to event type all under the event type of target basis event type dictionary carry out intelligent Matching, obtain with
Optimal keyword combines matching degree highest event type, the object event class using the event type as Present News event
Type.
The information-pushing method that the present embodiment provides, goes out Present News by the keyword intelligent Matching of Present News event
The event type that event is subordinate to, according to the mapping relations between event type and object, it is possible to determine and Present News thing
The related destination object of part.What the present embodiment was got is the objects such as the product directly related with Present News event or service,
Compared with prior art, ordinary individual can save from the mass data for exist complex relationship analyze related object expend when
Between.Further, due to ordinary individual and do not have professional experiences, the result accuracy obtained after analyzing mass data compared with
Difference, and the present invention directly provides the object such as Related product or service, improves the accuracy of analysis, can more meet ordinary individual
Demand, for ordinary individual formulate capital project significantly more efficient guiding or help are provided.
Example IV
As shown in figure 4, its schematic flow sheet for the information-pushing method of the embodiment of the present invention four.In above-described embodiment
On basis, the history media event according to corresponding to the object event type, determine the object event type to institute
The influence force parameter of destination object is stated, is comprised the following steps：
S401, traversal obtain the history media event included by object event type.
After the object event type of Present News event is determined, target thing can be got by way of traversal
History media event included by part type.
S402, it is object event type settling time sequence.
Specifically, after traversal gets all historical events of object event type, one is established for the object event type
Individual time series, wherein, it can embody what the history media event that object event type includes occurred in the time series
Time, and the duration after the appearance of history media event on being influenceed caused by destination object.
S403, the time series based on object event type carry out history to all history media events and return survey, obtain institute
There are influence data of the history media event to destination object.
After the time series of object event type is established, based on the time series can the object event type it is all
History media event carries out history and returns survey, to obtain influence data of each history media event to destination object.
S404, based on influence data determine influence force parameter.
According to object event type on influenceing data caused by destination object, the destination object in the time series is obtained
Situation of Profit or price trend, so as to form the influence force parameter after object event type occurs to destination object.Wherein, shadow
Ring force parameter can reflect media event under object event type occur after influence to the price trend of destination object with
And influence the duration.Wherein influenceing force parameter can be shown by an effect tendency figure, can be made user more
It is intuitive to see after the generation of object event type on influence caused by destination object, and then can is that the objective presentation investment of user be built
View.
Further, the influence force parameter after being occurred according to object event type to destination object, can predict current
Media event is to the influence power of destination object, i.e. Present News event is to effect caused by destination object, including to mesh
Mark the influence section of object and influence duration.For example, when destination object is an investment product, can predict current new
News event can predict price trend shadow of the Present News event to investment product to the price trend of the investment product
The loud duration.For example, user is after event type is received to the effect tendency figure of destination object, it can be determined that goes out current
After media event generation in 1-5 days, the destination object of influence has more than 5% excess earnings.
In order to more fully understand the present embodiment provide media event related object acquisition methods, be divided to two kinds of scenes illustrate into
Row explanation.
Scene one：When the basic event type of Present News type is socialization media event：
Present News event：National Development and Reform Committee was given notice December 16 title, and petrol and diesel oil highest was raised from December 17
Retail price.
Classification is identified to the type of Present News event, and obtains relative destination object, in this example,
Destination object is investment product.
Concrete processing procedure：
By the feature extraction to Present News event, following keyword is obtained：" National Development and Reform Committee, the Committee of Development and Reform, December 16
From day, December 17, raise, petrol and diesel oil, retail price ".
After keyword is got, keyword is combined, then using intelligent Matching algorithm, obtains matching degree most
High crucial phrase is combined into：From National Development and Reform Committee+December 17+and up-regulation+petrol and diesel oil+retail price, matching degree highest pass
The corresponding basic event type of keyword combination is socialization event.
Matching degree highest keyword is combined into the event type in traversal socialization event, can correspondingly be obtained with being somebody's turn to do
Keyword combines, matching degree highest object event type, and object event type is the petrochemical industry row in industry development planning
The adjusting and optimizing of industry, according to the mapping relations between event type and object, available destination object is and petrochemical industry
Business association highest destination object includes：Yong Da groups, CNPC, Air China's photoelectricity, vast new growth blend fund, middle oil
Strategic new industry blend fund, and the leading tactful blend fund of silver China.
Further, object event type is traveled through on influenceing data caused by destination object, and is the object event type
Establish a time series.It can embody what the history media event that object event type includes occurred in the time series
Event, and the duration after the appearance of history media event on being influenceed caused by destination object.According to object event type
Influence data to destination object, the situation of Profit or price trend of the destination object in the time series are obtained, so as to shape
To the effect tendency figure of destination object after occurring into object event type, and then it can be the objective presentation suggestion for investment of user.Its
In, effect tendency figure can reflect media event under object event event type after occurring to the price trend of destination object
Influence and influence the duration.
, can be by target pair after the influence force parameter of destination object and object event type to destination object is got
As being pushed to user with influence force parameter these information of object event type to destination object, and corresponding to user it is mobile eventually
It is shown on the display interface at end.As shown in figure 5, shown on display interface Present News event description and with it is current new
The destination object of the related recommendation of news event is investment product.
When user click on recommend an investment product after, can show in history occur same class event after to the throwing
The influence power of assets product is effect tendency figure.In this example, the investment product that user clicks on is Yong Da groups, when such as Fig. 6 institutes
Show, in figure can obtain with Present News event belong to object event type media event occur in history after to forever up to collection
The influence of this investment product of group, alternatively, can also include a historical review, can have in historical review and Present News
Event can be to Yong Da groups after belonging to the number that the media event of object event type occurred in history, and appearance
The simplified summary of the influence situation of income.To occurring belonging to object event with Present News event in effect tendency figure in Fig. 6
During the media event of type, it is possible to it is marked according in the time time series of appearance, can be with when user, which clicks on, to be marked
See specific influence of the event on Yong Da groups.
Scene two：When the basic event type of Present News event is that company announces class event：
Present News event：On December 25th, 2015, east China Telecom, big magnificent share and upper sea green kainogenesis senior executive/shareholder
Increasing is held, and shows confidence of the senior executive/shareholder for company's future development.
Classification is identified to the type of Present News event, and obtains relative destination object, in this example,
Destination object is investment product.
Concrete processing procedure：
By the feature extraction to Present News event, following keyword is obtained：" on December 25th, 2015, east China Telecom,
Big magnificent share, upper sea green new, senior executive/shareholder, increase and hold ".
After keyword is got, keyword is combined, then using intelligent Matching algorithm, obtains matching degree most
High crucial phrase is combined into：25 days+east China Telecom December in 2015+big magnificent share+goes up sea green new+senior executive/shareholder+increasing and held, should
Basic event type corresponding to the combination of matching degree highest keyword is that company announces class event.
Matching degree highest keyword is combined into the event type in traversal socialization event, can correspondingly be obtained with being somebody's turn to do
Keyword combines, matching degree highest object event type, and object event type is：Senior executive increase holds, according to event type with it is right
Mapping relations as between, available destination object include：East China Telecom, big magnificent share is upper sea green new, and Hua Taibai
Auspicious value increases blend fund.
Further, the historical data that traversal object event type has an impact to destination object, for the object event class
Type establishes a time series, and can embody the history media event that object event type includes in the time series occurs
Event, and on duration for being influenceed caused by destination object after the appearance of history media event.Further, according to mesh
The historical data that is had an impact to destination object of mark event type, obtain in the time series situation of Profit of destination object or
Person's price trend, so as to be formed to the effect tendency figure of destination object after object event type occurs, and then can be user visitor
See and suggestion for investment is presented.Wherein, effect tendency figure can reflect that object event type is walked after occurring to the price of destination object
The influence of gesture and influence duration.
After destination object and object event type is got to the influence power of destination object, can by destination object and
To the influence power of destination object, these information are pushed to user to object event type, and in the display of mobile terminal corresponding to user
It is shown on interface.As shown in fig. 7, on display interface show Present News event description and with Present News event phase
The destination object of the recommendation of pass is investment product.
When user click on recommend an investment product after, can show in history occur same class event after to the throwing
The influence power of assets product is effect tendency figure.
In this example, the investment product that user clicks on, when as shown in figure 8, can obtain in figure same with Present News event
Belong to object event type media event occur in history after influence to investment product, alternatively, one can also be included
Individual historical review, can have in historical review and belong to the media event of object event type in history with Present News event
Can be to the simplified summary of the influence situation of the income of investment product after the number occurred, and appearance.
The information-pushing method that the present embodiment provides, compared with prior art, ordinary individual can be saved complicated from existing
The time that related object expends is analyzed in the mass data of relation.Further, due to ordinary individual and do not have professional experiences,
The result accuracy obtained after analyzing mass data is poor, and the present invention directly provides the object such as Related product or service,
The accuracy of analysis is improved, can more meet the needs of ordinary individual, is provided for ordinary individual's formulation capital project more efficient
Guiding or help.
Embodiment five
As shown in figure 9, its structural representation for the information push-delivery apparatus of the embodiment of the present invention five.The information push-delivery apparatus
Including：Object acquisition module 11, determining module 12 and pushing module 13.
Wherein, object acquisition module 11, for the object event type according to belonging to Present News event, it is determined that described work as
Destination object corresponding to preceding media event.
In actual life, and not all media event can all affect for the tendency of industry, and be formed
The rule of ensured sustained development by data mining and history from a large amount of history media events, it is necessary to return the technological means such as survey to news thing
Part is sorted out.
The basic event type identified at present includes socialization media event and company's bulletin class media event.Example
Such as, magnitude exceeds certain series and the density of population of scene can be used as a socialization news more than certain level
Event, bulletin is held in senior executive's increase and decrease and increasing and decreasing the ratio held can be used as a company to announce class media event more than certain level.
In the present embodiment, object acquisition module 11 needs the content according to Present News event, determines Present News thing
The object event type that part is subordinate to.Specifically, object acquisition module 11 can get the keyword of Present News event, root
Matched according to keyword with the keyword of object event type, determine the object event type of the current media event.
Further, object acquisition module 11 can obtain difference according to influence of a large amount of history media events to industry
The industry that is influenceed of event type in specific product or service, wherein these products or service can be used as object,
Object corresponding with each event type can be got.After the object event type of Present News event is determined, object
Acquisition module 11 can obtain corresponding with the object event type according to the corresponding relation between above-mentioned event type and object
The related destination object of destination object, i.e. Present News event.
Determining module 12, for the history media event according to corresponding to the object event type, determine the target thing
Influence force parameter of the part type to the destination object.
After object event type is determined, determining module 12 can get included under the object event type go through
History media event.After all history media events are obtained, determining module 12 is obtained after history media event occurs to target
The influence data of object are right after the media event that can get object event type by analyzing these influence data occurs
The influence force parameter of destination object, wherein, influenceing force parameter includes being embodied in effect and the influence on destination object
Time on.For example, when destination object is an investment product, determining module 12 can be analyzed under object event type
Influence of the media event to the price trend of the investment product, and media event are held on what the price trend of investment product influenceed
The continuous time.Using object event type on the influence power of destination object as influence force parameter.
Pushing module 13, for the destination object and the influence force parameter to be pushed into user.
Influence power of the history media event included by object event type to destination object is determined in determining module 12
After parameter, destination object and history media event can be pushed to use by pushing module 13 to the affecting parameters of destination object
Family, in order to which user is according to the affecting parameters to destination object, formulate investment tactics.It is alternatively possible to getting target pair
As rear, destination object is first pushed into user, when user sounds out purpose object, the mesh can be chosen by operations such as clicks
Object is marked, influence force parameter of the history media event to destination object is now pushed into user again.
The information push-delivery apparatus that the present embodiment provides, by the object event type being subordinate to according to Present News event, really
The destination object with Present News event object is made, according to history media event corresponding to object event type, determines mesh
Influence force parameter of the event type to destination object is marked, destination object and influence force parameter are pushed to user.The present embodiment
What is got is the objects such as the product directly related with Present News event or service, and compared with prior art, ordinary individual can
To save the time that related object is analyzed from the mass data for exist complex relationship and is expended.Further, due to ordinary individual
And not having professional experiences, the result accuracy obtained after analyzing mass data is poor, and the present invention directly provides correlation
The object such as product or service, the accuracy of analysis is improved, can more meet the needs of ordinary individual, investment is formulated for ordinary individual
Scheme provides significantly more efficient guiding or help.
Embodiment six
As shown in Figure 10, its for the embodiment of the present invention six information push-delivery apparatus structural representation.The media event phase
Close object acquisition device：Object acquisition module 11, determining module 12, pushing module 13, keyword acquisition module 14 and classification mould
Block 15.
A kind of optional frame mode of object acquisition module 11 in the present embodiment, including：
Data mining unit 111, for being handled by data mining the classification of history media event.
Construction unit 112, for obtain basic event type, type of foundation dictionary corresponding to each basic event type,
Event type dictionary corresponding to event type and each event type included by each basic event type.
First establishes unit 113, for establishing the time series of each event type.
First history, which is returned, surveys unit 114, returns survey for carrying out history to the time series of each event type, obtains each
The related object of event type, builds the mapping relations between event type and object.
First acquisition unit 115, for according to the mapping relations and the object event type, obtaining the target
Object.
Keyword acquisition module 14, for obtaining the object event class belonging to Present News event in object acquisition module 11
Type, before determining destination object corresponding to the Present News event, to obtain the keyword of Present News event.
Classifying module 15, for carrying out event type classification to Present News event according to the keyword, to obtain
State object event type.
A kind of optional frame mode of determining module 12 in the present embodiment, including：
Traversal Unit 121, for traveling through the history media event included by the acquisition object event type.
Second establishes unit 122, for for the object event type settling time sequence.
Second history, which is returned, surveys unit 123, for the time series based on the object event type to all history news
Event carries out history and returns survey, obtains influence data of all history media events to the destination object.
Second acquisition unit 124, for determining the influence force parameter based on the influence data.
A kind of alternatively frame mode of classifying module 15 in the present embodiment, including：
Assembled unit 151, for all keywords of Present News event to be combined；
First sorts out unit 152, for being sorted out according to the combination of each keyword to Present News event base, to obtain when new
The target basis event type of news event；
Setting unit 153, for being optimal keyword by crucial phrase cooperation corresponding to the event type of the target basis
Combination；
Second sorts out unit 154, and event type classification is carried out to Present News event according to the optimal keyword combination,
To obtain the object event type.
Keyword acquisition module 14, specifically for carrying out feature information extraction to the body matter of Present News event, with
Obtain the keyword of Present News event.
First sorts out unit 152, specifically for each keyword is combined into type of foundation word corresponding with basic event type
Storehouse carries out intelligent Matching, using basic event type corresponding to the combination of matching degree highest keyword as target basis event class
Type.
Second sorts out unit 154, specifically for being combined using the optimal keyword to target basis event type
Event type dictionary corresponding to included all event types carries out intelligent Matching, obtains combining with the optimal keyword
With degree highest event type；
Setting unit 153, it is additionally operable to combine matching degree highest event type with the optimal keyword as current
The object event type of media event.
Compared with prior art, ordinary individual can save and be analyzed from the mass data for exist complex relationship the present embodiment
The time that related object expends.Further, due to ordinary individual and do not have professional experiences, obtained after analyzing mass data
Result accuracy it is poor, and the present invention directly offer object such as Related product or service improves the accuracy of analysis, more
Can meet the needs of ordinary individual, formulating capital project for ordinary individual provides significantly more efficient guiding or help.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above-mentioned each method embodiment can lead to
The related hardware of programmed instruction is crossed to complete.Foregoing program can be stored in a computer read/write memory medium.The journey
Sequence upon execution, execution the step of including above-mentioned each method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than its limitations；To the greatest extent
The present invention is described in detail with reference to foregoing embodiments for pipe, it will be understood by those within the art that：Its according to
The technical scheme described in foregoing embodiments can so be modified, either which part or all technical characteristic are entered
Row equivalent substitution；And these modifications or replacement, the essence of appropriate technical solution is departed from various embodiments of the present invention technology
The scope of scheme.
Claims (16)
- A kind of 1. information-pushing method, it is characterised in that including：Object event type according to belonging to Present News event, determine destination object corresponding to the Present News event；According to history media event corresponding to the object event type, determine the object event type to the destination object Influence force parameter；The destination object and the influence force parameter are pushed to user.
- 2. according to the method for claim 1, it is characterised in that the object event class according to belonging to Present News event Type, destination object corresponding to the Present News event is determined, including：The classification of history media event is handled by data mining；Obtain basic event type, each type of foundation dictionary, each basic event type are wrapped corresponding to basic event type Event type dictionary corresponding to the event type and each event type included；Establish the time series of each event type；History time survey is carried out to the time series of each event type, the related object of each event type is obtained, builds event Mapping relations between type and object；According to the mapping relations and the object event type, the destination object is obtained.
- 3. according to the method for claim 2, it is characterised in that the object event class according to belonging to Present News event Type, before determining destination object corresponding to the Present News event, in addition to：Obtain the keyword of Present News event；Event type classification is carried out to Present News event according to the keyword, described in obtaining and be subordinate to when media event Object event type.
- 4. according to the method described in claim any one of 1-3, it is characterised in that described corresponding according to the object event type History media event, determine influence force parameter of the object event type to the destination object, including：Traversal obtains the history media event included by the object event type；For the object event type settling time sequence；Time series based on the object event type carries out history to all history media events and returns survey, obtains all history Influence data of the media event to the destination object；The influence force parameter is determined based on the influence data.
- 5. according to the method for claim 3, it is characterised in that described that Present News event is carried out according to the keyword Event type is sorted out, to obtain the object event type being subordinate to when media event, including：All keywords of Present News event are combined；Present News event base is sorted out according to the combination of each keyword, to obtain the target basis event class when media event Type；Crucial phrase cooperation corresponding to the event type of the target basis is combined for optimal keyword；Event type classification is carried out to Present News event according to the optimal keyword combination, to obtain the object event class Type.
- 6. the method according to claim 3 or 5, it is characterised in that the keyword for obtaining Present News event, bag Include：Feature information extraction is carried out to the body matter of Present News event, to obtain the keyword of Present News event.
- 7. according to the method for claim 5, it is characterised in that described to be combined according to each keyword to Present News event base Plinth is sorted out, to obtain the target basis event type when media event, including：Each keyword is combined into type of foundation dictionary corresponding with basic event type and carries out intelligent Matching；Using basic event type corresponding to the combination of matching degree highest keyword as target basis event type.
- 8. according to the method for claim 5, it is characterised in that described to be combined according to the optimal keyword to Present News Event carries out event type classification, to obtain the object event type, including：Using the optimal keyword combination to thing corresponding to all event types included by the event type of the target basis Part type dictionary carries out intelligent Matching, obtains combining matching degree highest event type with the optimal keyword；The mesh being subordinate to matching degree highest event type is combined with the optimal keyword as Present News event Mark event type.
- A kind of 9. information push-delivery apparatus, it is characterised in that including：Object acquisition module, for the object event type according to belonging to Present News event, determine the Present News event Corresponding destination object；Determining module, for the history media event according to corresponding to the object event type, determine the object event type Influence force parameter to the destination object；Pushing module, for the destination object and the influence force parameter to be pushed into user.
- 10. device according to claim 9, it is characterised in that the object acquisition module, including：Data mining unit, for being handled by data mining the classification of history media event；Construction unit, for obtaining basic event type, each type of foundation dictionary, each basis corresponding to basic event type Event type dictionary corresponding to event type and each event type included by event type；First establishes unit, for establishing the time series of each event type；First history, which is returned, surveys unit, returns survey for carrying out history to the time series of each event type, obtains each event class The related object of type, builds the mapping relations between event type and object；First acquisition unit, for according to the mapping relations and the object event type, obtaining the destination object.
- 11. device according to claim 10, it is characterised in that also include：Keyword acquisition module, for obtaining the object event type belonging to Present News event in the object acquisition module, Before determining destination object corresponding to the Present News event, to obtain the keyword of Present News event；Classifying module, for carrying out event type classification to Present News event according to the keyword, to obtain the target Event type.
- 12. according to the device described in claim any one of 9-11, it is characterised in that the determining module, including：Traversal Unit, for traveling through the history media event included by the acquisition object event type；Second establishes unit, for for the object event type settling time sequence；Second history, which is returned, surveys unit, and all history media events are carried out for the time series based on the object event type History, which is returned, to be surveyed, and obtains influence data of all history media events to the destination object；Second acquisition unit, for determining the influence force parameter based on the influence data.
- 13. device according to claim 11, it is characterised in that the classifying module, including：Assembled unit, for all keywords of Present News event to be combined；First sorts out unit, for sorting out according to the combination of each keyword to Present News event base, works as media event to obtain Target basis event type；Setting unit, for crucial phrase cooperation corresponding to the event type of the target basis to be combined for optimal keyword；Second sorts out unit, event type classification is carried out to Present News event according to the optimal keyword combination, to obtain The object event type.
- 14. the device according to claim 11 or 13, it is characterised in that the keyword acquisition module, specifically for pair The body matter of Present News event carries out feature information extraction, to obtain the keyword of Present News event.
- 15. device according to claim 13, it is characterised in that described first sorts out unit, specifically for will each close Keyword combines type of foundation dictionary corresponding with basic event type and carries out intelligent Matching, by the combination pair of matching degree highest keyword The basic event type answered is as target basis event type.
- 16. device according to claim 13, it is characterised in that described second sorts out unit, specifically for described in utilization Optimal keyword combination is entered to event type dictionary corresponding to all event types included by the event type of the target basis Row intelligent Matching, obtain combining matching degree highest event type with the optimal keyword；The setting unit, it is additionally operable to combine matching degree highest event type as Present News using with the optimal keyword The object event type that event is subordinate to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610334269.2A CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610334269.2A CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107402925A true CN107402925A (en) | 2017-11-28 |
| CN107402925B CN107402925B (en) | 2020-11-20 |
Family
ID=60394014
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610334269.2A Active CN107402925B (en) | 2016-05-19 | 2016-05-19 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107402925B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110209930A (en) * | 2019-04-30 | 2019-09-06 | 平安科技（深圳）有限公司 | The investee's analysis method and device of knowledge based map |
| CN111209400A (en) * | 2020-01-03 | 2020-05-29 | 北京明略软件系统有限公司 | Data analysis method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050071217A1 (en) * | 2003-09-30 | 2005-03-31 | General Electric Company | Method, system and computer product for analyzing business risk using event information extracted from natural language sources |
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN104573054A (en) * | 2015-01-21 | 2015-04-29 | 杭州朗和科技有限公司 | Information pushing method and equipment |
| CN104902292A (en) * | 2015-05-20 | 2015-09-09 | 无锡天脉聚源传媒科技有限公司 | Television report-based public opinion analysis method and system |
-
2016
- 2016-05-19 CN CN201610334269.2A patent/CN107402925B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050071217A1 (en) * | 2003-09-30 | 2005-03-31 | General Electric Company | Method, system and computer product for analyzing business risk using event information extracted from natural language sources |
| CN103226554A (en) * | 2012-12-14 | 2013-07-31 | 西藏同信证券有限责任公司 | Automatic stock matching and classifying method and system based on news data |
| CN104573054A (en) * | 2015-01-21 | 2015-04-29 | 杭州朗和科技有限公司 | Information pushing method and equipment |
| CN104902292A (en) * | 2015-05-20 | 2015-09-09 | 无锡天脉聚源传媒科技有限公司 | Television report-based public opinion analysis method and system |
Non-Patent Citations (1)
| Title |
|---|
| 刘盼盼 等: "基于灰色关联分析的中文新闻事件关联性识别", 《计算机应用》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110209930A (en) * | 2019-04-30 | 2019-09-06 | 平安科技（深圳）有限公司 | The investee's analysis method and device of knowledge based map |
| CN111209400A (en) * | 2020-01-03 | 2020-05-29 | 北京明略软件系统有限公司 | Data analysis method and device |
| CN111209400B (en) * | 2020-01-03 | 2023-06-16 | 北京明略软件系统有限公司 | Data analysis method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107402925B (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Anton et al. | Enterprise risk management: A literature review and agenda for future research | |
| USRE46902E1 (en) | System and method for customized sentiment signal generation through machine learning based streaming text analytics | |
| Pradhan et al. | Digital marketing and SMES: An identification of research gap via archives of past research | |
| US8676691B2 (en) | System, report, and method for generating natural language news-based stories | |
| US20120179692A1 (en) | System and Method for Visualizing Sentiment Assessment from Content | |
| Zarutska et al. | Structural-functional analysis as innovation in public governance (case of of banking supervision) | |
| US20090319931A1 (en) | Visual Intelligence Systems | |
| US20190213194A1 (en) | System and method for information recommendation | |
| CN106651424A (en) | Electric power user figure establishment and analysis method based on big data technology | |
| EP2786335A1 (en) | Methods and systems for predicting market behavior based on news and sentiment analysis | |
| CN113535963B (en) | Long text event extraction method and device, computer equipment and storage medium | |
| CN112256762B (en) | Enterprise portrait method, system, equipment and medium based on industrial map | |
| US20150221038A1 (en) | Methods and system for financial instrument classification | |
| Chen et al. | From opinion mining to financial argument mining | |
| CN103258034A (en) | Economic and financial behavior analysis system model based on social media | |
| Alareeni | A review of auditors’ GCOs, statistical prediction models and artificial intelligence technology | |
| CN109344174A (en) | Financial analysis method and system | |
| Chi et al. | The use and usefulness of big data in finance: Evidence from financial analysts | |
| CN107402925A (en) | Information-pushing method and device | |
| CN108073678A (en) | Applied to document analyzing and processing method, system and the device in big data analysis | |
| Uyanik et al. | How Realistic Are Coal Phase-Out Timeline Targets for Turkey? | |
| Romano et al. | Towards a weighted voting system for Q&A sites | |
| Lee et al. | Exploiting linked data in financial engineering | |
| Linardos et al. | Using financial news articles with minimal linguistic resources to forecast stock behaviour | |
| AHMED et al. | REVIEW OF OPPORTUNITIES AND CHALLENGES REPRESENTED BY BIG DATA FOR FINTECH IN MALAYSIA |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 1247321 Country of ref document: HK |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20201014 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20201014 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |