CN109961313A - A kind of information-pushing method and information push-delivery apparatus - Google Patents
A kind of information-pushing method and information push-delivery apparatus Download PDFInfo
- Publication number
- CN109961313A CN109961313A CN201711437698.3A CN201711437698A CN109961313A CN 109961313 A CN109961313 A CN 109961313A CN 201711437698 A CN201711437698 A CN 201711437698A CN 109961313 A CN109961313 A CN 109961313A
- Authority
- CN
- China
- Prior art keywords
- information
- feature information
- chain
- behavior
- behavioral
- 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
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0255—Targeted advertisements based on user history
Abstract
The present invention provides a kind of information-pushing method and information push-delivery apparatus, wherein this method comprises: the behavioural information based on user, obtains fisrt feature information corresponding with the behavioural information；From preset behavioral chain library, behavioral chain corresponding with the fisrt feature information is obtained；Each behavioral chain includes multiple second feature information, and second feature information described in wherein at least one is identical as the fisrt feature information；Other second feature information of at least one of each second feature information and same behavior chain have behavior continuous relationship；The corresponding pushed information of the second feature information in the behavioral chain corresponding with the fisrt feature information is obtained, and pushes the pushed information.The embodiment of the present invention realizes the timely push of pushed information, avoids in the case where user demand has obtained satisfaction, corresponding pushed information is pushed to user, to avoid the lag of pushed information.
Description
Technical field
The present invention relates to computer application technologies, push in particular to a kind of information-pushing method and information
Device.
Background technique
User's portrait is also known as user role, delineates having for target user, connection user's demand and design direction as a kind of
Effect tool, user's portrait are used widely in each field.User's portrait is the virtual representations of real user, based on true people's
Behavior viewpoint obtains, and is analyzed by the historical behavior to user, adds spy corresponding with the historical behavior of user for user
Reference breath, to form user's portrait of some or certain a kind of user.
It when user's portrait is applied in information push field, first has to analyze user's history behavior, forms user
Portrait, is then based on user and draws a portrait to user's pushed information；For example, generate personage portrait when, can by user buy or
The historical information of browsing commodity can obtain the information such as the more interested type of merchandise of user, the price range of commodity, will be upper
State composition of the information obtained as user's characteristic information, as user's portrait；When to user's pushed information, it can be based on
The information such as the interested type of merchandise, the price range of commodity, acquisition and the interested type of merchandise, the price range of commodity etc.
Then pushed information is pushed to user as pushed information by the information of other commodity as info class.
It is this to be directly based upon characteristic information included in user's portrait and obtain pushed information corresponding with characteristic information, and
Pushed information is pushed to the mode of user, the excavation to user real demand is only more shallow hierarchy, is often had stagnant
Property afterwards can not bring preferable push effect；It is sold for example, user frequently accesses certain shopping website sport footwear within certain period
Webpage is sold, and the sale price section of sport footwear is between 300 yuan -500 yuan, then being included when generating user's portrait
Characteristic information may are as follows: sport footwear, price range, style；When obtaining pushed information, the two characteristic informations can be also based on,
Price is pushed in 300 yuan of -500 yuan of sport footwear advertisements to user；If user has purchased when accessing shopping website at this time
It buys, then no longer there is the demand of purchase sport footwear, push the advertisement of sport footwear to user again later, can not play preferable
Effect.
Therefore, method a kind of to obtain user demand in time, that information push is carried out based on the user demand obtained in time
As current urgent problem to be solved.
Summary of the invention
In view of this, the embodiment of the present invention is designed to provide a kind of information-pushing method and information push-delivery apparatus, energy
Enough behavioural informations according to user obtain user demand in time, and carry out information push based on the user demand obtained in time.
In a first aspect, the embodiment of the invention provides a kind of information-pushing methods, comprising:
Behavioural information based on user obtains fisrt feature information corresponding with the behavioural information；
From preset behavioral chain library, behavioral chain corresponding with the fisrt feature information is obtained；Each behavioral chain
Including multiple second feature information, and second feature information described in wherein at least one is identical as the fisrt feature information；
Other second feature information of at least one of each second feature information and same behavior chain have behavior continuous relationship；
Obtain the corresponding push of the second feature information in the behavioral chain corresponding with the fisrt feature information
Information, and push the pushed information.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, in which: institute
State behavioral chain further include: with the corresponding execution time interval of two of the behavior continuous relationship second feature information；
The second feature information obtained in the behavioral chain corresponding with the fisrt feature information is corresponding
Pushed information, and the pushed information is pushed, it specifically includes:
According to the fisrt feature information, identical with the fisrt feature information specific the is determined from the behavioral chain
Two characteristic informations；
The undertaking second feature information in behavioral chain with the specific second feature information with behavior continuous relationship is obtained,
And the downstream accepted second feature information and be located at the specific second feature information；
According to the specific corresponding behavior time of origin of second feature information and the specific second feature information with
The corresponding execution time interval of the undertaking second feature information calculates and accepts second feature information corresponding effective time；
Within the effective time, the pushed information corresponding with the undertaking second feature information is pushed.
With reference to first aspect, the embodiment of the invention provides second of possible embodiments of first aspect, in which: institute
It states from preset behavioral chain library, obtains behavioral chain corresponding with the fisrt feature information, specifically include:
From the acquisition goal behavior chain in the behavioral chain library；Have in the goal behavior chain and believes with the fisrt feature
Cease the identical second feature information；
Detect the end whether second feature information identical with the fisrt feature information is located at the goal behavior chain；
If it is not, then using the goal behavior chain as behavioral chain corresponding with the fisrt feature information.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, in which: institute
It states from preset behavioral chain library, before acquisition behavioral chain corresponding with the fisrt feature information, further includes:
Multiple behavioral chains are generated, and the behavioral chain library is established according to the behavioral chain generated.
With reference to first aspect, the embodiment of the invention provides the 4th kind of possible embodiments of first aspect, in which: institute
State and generate multiple behavioral chains, specifically include: the behavior sample information based on user obtains and the behavior sample information pair
The feature samples information answered；The feature samples information of each user has multiple；
The feature samples information is trained using the method for machine learning, obtains the institute with behavior continuous relationship
State feature samples information；
By according to the behavior continuous relationship between the second feature information, the behavioral chain is generated.
Second aspect, the embodiment of the present invention also provide a kind of information push-delivery apparatus, which includes:
First determining module obtains fisrt feature corresponding with the behavioural information for the behavioural information based on user
Information；
Second determining module, for obtaining behavior corresponding with the fisrt feature information from preset behavioral chain library
Chain；Each behavioral chain includes multiple second feature information, and second feature information described in wherein at least one with it is described
Fisrt feature information is identical；Other second feature information of at least one of each second feature information and same behavior chain
With behavior continuous relationship；
Pushing module, for obtaining the letter of the second feature in the behavioral chain corresponding with the fisrt feature information
Corresponding pushed information is ceased, and pushes the pushed information.
In conjunction with second aspect, the embodiment of the invention provides the first possible embodiments of second aspect, in which: institute
State behavioral chain further include: with the corresponding execution time interval of two of the behavior continuous relationship second feature information；
The pushing module, is specifically used for: determining with described the from the behavioral chain according to the fisrt feature information
The identical specific second feature information of one characteristic information；It obtains and is held with the specific second feature information with behavior in behavioral chain
The undertaking second feature information of relationship is connect, and the undertaking second feature information is located under the specific second feature information
Trip；According to the specific corresponding behavior time of origin of second feature information and the specific second feature information with it is described
The corresponding execution time interval of second feature information is accepted, calculates and accepts second feature information corresponding effective time；Described
In effective time, the pushed information corresponding with the undertaking second feature information is pushed.
In conjunction with second aspect, the embodiment of the invention provides second of possible embodiments of second aspect, in which: institute
The second determining module is stated, is specifically used for: from the acquisition goal behavior chain in the behavioral chain library；Have in the goal behavior chain
The second feature information identical with the fisrt feature information；Detect second feature identical with the fisrt feature information
Whether information is located at the end of the goal behavior chain；If it is not, then using the goal behavior chain as with the fisrt feature
The corresponding behavioral chain of information.
In conjunction with second aspect, the embodiment of the invention provides the third possible embodiments of second aspect, in which: also
Include: generation module, establishes the behavioral chain for generating multiple behavioral chains, and according to the behavioral chain generated
Library.
In conjunction with second aspect, the embodiment of the invention provides the 4th kind of possible embodiments of second aspect, in which: institute
State generation module to be specifically used for: the behavior sample information based on user obtains feature sample corresponding with the behavior sample information
This information；The feature samples information of each user has multiple；
The feature samples information is trained using the method for machine learning, obtains the institute with behavior continuous relationship
State feature samples information；
By according to the behavior continuous relationship between the second feature information, the behavioral chain is generated.
Information-pushing method and device provided by the embodiment of the present application will be first based on when carrying out information push
The behavioural information of user obtains corresponding fisrt feature information, and obtains in subordinate act chain library corresponding with fisrt feature information
Behavioral chain.It include multiple second feature information in behavioral chain, and in each second feature information and same behavior chain at least
One other second feature information has behavior continuous relationship, and then the second feature information in Behavior-based control chain carries out user
The push of pushed information.In this process, push is pushed based on the behavioral chain pre-established, behavioral chain itself
It is the set with the second feature information of behavior continuous relationship, is corresponded to when one of second feature information currently occurs for user
Behavior when, can know the subsequent issuable behavior of user, on this basis so as to generate in the subsequent behavior of user
The demand for just knowing user in advance before realizes the timely push of pushed information, to avoid the lag of pushed information, thus
It avoids that corresponding pushed information is just pushed to user in the case where user demand has obtained satisfaction.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows a kind of flow chart of information-pushing method provided by the embodiment of the present invention；
Fig. 2 shows the example schematic diagrams of behavioral chain provided by the embodiment of the present invention；
Fig. 3 shows the example schematic diagram of concrete behavior chain provided by the embodiment of the present invention；
Fig. 4 shows the example schematic diagram of another kind concrete behavior chain provided by the embodiment of the present invention；
Fig. 5 shows the flow chart of another kind information-pushing method provided by the embodiment of the present invention；
Fig. 6 shows the example schematic diagram of another kind concrete behavior chain provided by the embodiment of the present invention；
Fig. 7 shows the flow chart of another kind information-pushing method provided by the embodiment of the present invention；
Fig. 8 shows a kind of structural schematic diagram of information push-delivery apparatus provided by the embodiment of the present invention；
Fig. 9 shows the structural schematic diagram of another kind information push-delivery apparatus provided by the embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
Middle attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
It is a part of the embodiment of the present invention, instead of all the embodiments.The present invention being usually described and illustrated herein in the accompanying drawings is real
The component for applying example can be arranged and be designed with a variety of different configurations.Therefore, of the invention to what is provided in the accompanying drawings below
The detailed description of embodiment is not intended to limit the range of claimed invention, but is merely representative of selected reality of the invention
Apply example.Based on the embodiment of the present invention, those skilled in the art institute obtained without making creative work
There are other embodiments, shall fall within the protection scope of the present invention.
In information-pushing method currently based on user's portrait, generally comprise following several: first, to user's history behavior
Analysis forms user's portrait, and is directly based upon the characteristic information in user's portrait included, obtains push corresponding with characteristic information
Information, and pushed information is pushed to user；Second, being based on user's history behavioural analysis forms user's portrait, and finds and be somebody's turn to do
Target user as user class carries out information push to all target users searched out based on user portrait；Third, being based on
User's history row in certain period of time is analyzed to form user's portrait, and according to the preset time interval, draws a portrait and carry out to user
It updates, is drawn a portrait based on updated user and carry out the push of pushed information.In above-mentioned three kinds of information-pushing methods, the first letter
Method for pushing pushed information lag is ceased, user is just pushed in the case that often user demand has been met, it can not band
Preferably to push effect；Meanwhile the push of the pushed information can often continue one section in the case where user does not have demand
Long time, waste of resource；Although second of information-pushing method expands same demand user group to a certain extent,
It is unable to satisfy the demand that each user is different from other users in the demand user group；The third information-pushing method is due to meeting
Periodically user's portrait is updated, but this update also results in the certain periodic demands of user and is dropped, it is difficult
To be waken up in user demand.Based on this, a kind of information-pushing method, device and system provided by the present application can bases
The behavioural information of user obtains user demand immediately, and carries out information push based on the user demand obtained immediately；Meanwhile also
The personalized information push for each user may be implemented, meet the different demands of user；In addition it is possible to realize certain weeks
The push of phase property demand.
For convenient for understanding the present embodiment, first to a kind of information-pushing method disclosed in the embodiment of the present invention into
Row is discussed in detail.
Shown in Figure 1, information-pushing method provided by the embodiment of the present application includes:
S101: the behavioural information based on user obtains fisrt feature information corresponding with the behavioural information.
When specific implementation, the behavioural information of user can be grabbed from network by data packet grabber to be obtained,
Such as user can send access request to the management server of the webpage when accessing some webpage；Management server connects
After receiving access request, web data corresponding with access request can be returned to user；The web data carries management clothes
The address information of device of being engaged in and the address information of user, data packet grabber can address information based on management server and users
Address information, grab the data packet that management server is returned to user, and based on the data carried in the data packet, obtain
The behavioural information of user.Wherein, behavioural information can be text information included in the page that user is opened, picture letter
At least one of breath, audio-frequency information, video information.
Furthermore it is also possible to obtain the behavioural information of user by other means, such as directly record in the management server
The access request of certain user, and the access request based on user obtain the access request with user directly from management server
Corresponding behavioural information.
After the style of writing information for obtaining user, to the behavioural information of user into analysis, behavioural information corresponding first is obtained
Characteristic information.Wherein, the corresponding behavioural information of each user has multinomial；When carrying out semantic analysis to each behavioural information,
At least one fisrt feature information can be obtained.
If behavioural information includes text information, semantic analysis is carried out to text information, language is being carried out to text information
When justice analysis, usually there are several types of methods:
1, semantic analysis is carried out to text information using natural language processing method.Natural language processing (natural
Language processing, NLP) enable to server to obtain the meaning of one section of natural text, such as nature can be passed through
Language processing techniques obtain the digest etc. of certain section of text information, which is fisrt feature information.
2, semantic analysis is carried out to text information using the method for machine learning.Machine learning (Machine Learning,
ML it) can be automatically analyzed from text information and obtain rule, carry out natural language processing.
3, it is retrieved in text information according to preset keyword, when search result meets the first preset condition,
Using tag along sort corresponding to the keyword as fisrt feature information.When realizing, the keyword preset is corresponding with
Specific tag along sort retrieves preset keyword, if it is pre- to retrieve this when carrying out semantic analysis in text information
If keyword, and reached the first preset condition so just using specific tag along sort corresponding to the keyword as language
The result of justice analysis.
If behavioural information includes image information, fisrt feature included by image is obtained by the method for pattern analysis
Information, for example, when in the page that user is accessed include image when, base signature analysis obtain image included in article into
Row identification, to obtain the fisrt feature information for including in image information.
If behavioural information includes video information, similar, difference when processing mode with behavioural information includes image information
It is to carry out pattern analysis based on multiple images in video when behavioural information is video information, obtaining video information includes
Fisrt feature information.
It if behavioural information includes audio-frequency information, can first translate audio into as text information, then text information is carried out
Semantic analysis, to obtain fisrt feature information included in audio-frequency information.
S102: from preset behavioral chain library, behavioral chain corresponding with the fisrt feature information is obtained；Each row
It include multiple second feature information for chain, and wherein at least one second feature information is identical with the fisrt feature information
The second feature information；Each second feature information has at least one second feature information in same behavior chain
Behavior continuous relationship.
When specific implementation, multiple preset behavioral chains are saved in behavioral chain library, and each behavioral chain includes
Multiple second feature information；For example, see a behavioral chain shown in Fig. 2, is provided, the second feature for including in behavior chain is believed
Breath is respectively as follows: F1, F2, F3, F4, F5, F6, F7, F8 and F9, any one second feature information therein and the same behavioral chain
In at least one characteristic information have behavior continuous relationship；In behavior continuous relationship, including undertaking person and by undertaking person,
Wherein it is normally in behavioral chain by undertaking person by the upstream of undertaking person；Such as in the Fig. 2, there is undertaking to close by F1 and F2
System, and F1 is by undertaking person, F2 is undertaking person；Meanwhile F1 and F3 also have continuous relationship, and F1 is by undertaking person, F3 is to hold
The person of connecing；In addition, F2 and F4 also have continuous relationship, and F2 is by undertaking person, and F4 is undertaking person.
Before executing the S102, further includes: generate multiple behavioral chains, and establish behavioral chain according to behavioral chain generated
Library.
Specifically: behavioral chain includes showing in behavioral chain and two kinds of potential behavioral chain.
It shows and refers to that user after having carried out some behavior, can generally be executed more subsequent based on the behavior in behavioral chain
Behavior, the behavioral chain that these behaviors collectively form.For example, when user is to automobile to be bought, in the decision of purchase automobile, greatly
Groups of people can pay close attention to and utilization of a loan purchase；After buying automobile, whole users can protect for the automobile purchase vehicle bought
Danger；Meanwhile it also having groups of people and can select the purchase relative articles such as car decoration；In addition, also having the correlation that groups of people pay close attention to carwash
Service item；Then in the behavior of the purchase automobile of user, behavioral chain generated is shown in Figure 3, is wrapped in behavior chain
The second feature information included includes；Purchase vehicle decision, auto loan are handled, purchase vehicle, carwash, purchase vehicle insurance, purchase car decoration.Therefore,
When fisrt feature information obtained in the behavioural information from user is purchase vehicle decision, automobile, loan can be handled, be washed
The corresponding advertisement pushing of the second feature information such as vehicle, vehicle insurance, car decoration is to user.
Potential behavioral chain refers to the user with some common traits, can have more similar demand.These are common
Feature can be at least one of features such as occupation, gender, occupation, residence, nationality, hobby, such as certain engineering machinery produces
Product shopping website is analyzed by the behavioural information to user, and the buyer for obtaining engineering machinery is largely engineering construction phase
The boss of pass or high-rise procurement staff, these users usually prefer BMW, and behavioral chain generated is referring to fig. 4
Shown, the second feature information that behavior chain includes includes: engineering machinery purchase, BMW；The institute in the behavioural information from user
The fisrt feature information of acquisition is that when constituting mechanical purchase, advertisement corresponding with BMW can be moved back and be pushed to user.
Generate behavioral chain mode there are several types of:
1, it is generated based on common sense.This method is generally used for showing the generation in behavioral chain, is showing in each of behavioral chain
Between two characteristic informations and at least one other second feature information all have show behavior continuous relationship, such as Fig. 3 institute it is right
In the example answered, after buying automobile, a series of demands such as purchase insurance, carwash, decoration automobile, and then basis can be generated therewith
These have show behavior continuous relationship demand generate behavioral chain.
2, user such as occupation, age, love are obtained by modes such as questionnaire survey, data packet crawl, website data statistics
Good, behavioural information within a certain period of time etc., using these information as the behavior sample information of user, acquisition and behavior sample
Then the corresponding feature samples information of information is trained these feature samples information using the method for machine learning, obtain
Demand with potential behavior continuous relationship generates behavioral chain.Such as in the example corresponding to above-mentioned Fig. 4, engineering machinery purchase
It buys and has no present and direct behavior continuous relationship between the purchase of BMW car, but the user based on engineering machinery purchase
Possessed others behavioural information, establishes potential behavior continuous relationship.
Shown in Figure 5, the embodiment of the present application also provides one kind from preset behavioral chain library, obtains and first spy
Reference ceases the specific method of corresponding behavioral chain, this method comprises:
S501: from the acquisition goal behavior chain in the behavioral chain library；Have and described first in the goal behavior chain
The identical second feature information of characteristic information；
S502: whether detection second feature information identical with the fisrt feature information is located at the target where it
The end of behavioral chain；If it is not, then jumping to S203；If it is, the goal behavior chain is not corresponding with fisrt feature information
Behavioral chain.
S503: using the goal behavior chain as behavioral chain corresponding with the fisrt feature information.
When specific implementation, due to include at least one and first in behavioral chain corresponding with fisrt feature information
Characteristic information identical second feature information has second feature information identical with fisrt feature information in behavior chain, and
The second feature information can not be located at the least significant end of behavioral chain, i.e., its behavior between other second feature information, which is accepted, closes
It is as existing for undertaking person at least in a behavior continuous relationship in system.
Herein, it should be noted that fisrt feature information herein can have one or have multiple；Work as fisrt feature
When information has multiple, to obtain behavioral chain corresponding with each fisrt feature information respectively in subordinate act chain library.
S103: the second feature information obtained in the behavioral chain corresponding with the fisrt feature information is corresponding
Pushed information, and push the pushed information.
When specific implementation, after obtaining behavioral chain corresponding with fisrt feature information, it can obtain to the row
For all second feature information included in chain, it is then based on second feature information, is obtained corresponding with second feature information
Pushed information is pushed to user by pushed information.
Herein, it should be noted that can be by pushed information corresponding with second feature information all in behavior chain
It is pushed to user；Second feature information corresponding pushed information in part can also be pushed to user.By part second feature
It is first identical with fisrt feature information second special in the behavior chain when the corresponding pushed information of information is pushed to user
Reference breath is positioned, and the other second feature information only pushed in behavior chain positioned at second feature information downstream are corresponding
Pushed information achievees the purpose that accurately to push.
It, can be disposably by second feature information pair all in behavioral chain when pushing pushed information to user
The pushed information answered is pushed to user, can also push according to the possible time of origin of behavior to user.
Shown in Figure 7, the embodiment of the present application, which also provides, a kind of obtains corresponding with the fisrt feature information row
For the corresponding pushed information of the second feature information in chain, and the method for pushing the pushed information, this method are specifically wrapped
It includes:
S701: according to the fisrt feature information, spy identical with the fisrt feature information in the behavioral chain is determined
Determine second feature information；
S702: the undertaking second feature in behavioral chain with the specific second feature information with behavior continuous relationship is obtained
Information, and the downstream accepted second feature information and be located at the specific second feature information；
S703: according to the specific corresponding behavior time of origin of second feature information and the specific second feature
Information execution time interval corresponding with the undertaking second feature information, when calculating undertaking second feature information is corresponding effective
Between；
S704: within the effective time, the pushed information corresponding with the undertaking second feature information is pushed.
When specific implementation, specific second feature information is in behavioral chain corresponding with fisrt feature information, with the
The identical second feature information of one characteristic information；Accepting second feature information is to accept with specific second feature information with behavior
The second feature information of relationship, and in the same behavioral chain, it accepts second feature information and is located at specific second feature information
Downstream, i.e. undertaking second feature information are undertaking person, and specific second feature information is by undertaking person.
Such as: certain demands of user are that can have certain time of origin, such as in the corresponding behavioral chain of above-mentioned Fig. 6
In example, purchase vehicle, which is likely to occur in, to be constituted in 7-30 days after loan is handled；First purchase car insurance is likely to occur in purchase vehicle
Within 90 days afterwards；It 9th month to 13rd month after purchase vehicle insurance for the first time, needs to buy vehicle insurance again.In order to close
The corresponding pushed information of user demand is pushed to user by the suitable time, and the application is implemented further includes in provided behavioral chain
Execution time interval corresponding with having two second feature information of behavior continuous relationship in behavioral chain.Between the execution time
Every and being expert at when referring in behavior continuous relationship as being occurred by the corresponding user behavior of the second feature information of undertaking person
For the time interval between the corresponding user behavior time of origin of second feature information in continuous relationship as undertaking person；Such as
In the corresponding example of above-mentioned Fig. 6, " purchase vehicle insurance " and " purchase vehicle " is believed as two second feature with behavior continuous relationship
Breath, " purchase vehicle insurance " are by undertaking person, and " purchase vehicle " is undertaking person, the execution time interval of " purchase vehicle insurance " and " purchase vehicle " for 7 to
30 days；Then after the behavior that " purchase vehicle insurance " corresponding user buys vehicle insurance occurs, pushed away in the 7th to 30 day later to user
Send pushed information relevant to " purchase vehicle ", such as the advertising information of various automobiles.
It should be noted that obtain user behavioral chain after, can also saving to behavioral chain, and establish user with
Mapping relations between behavioral chain, while recording the corresponding behavior of second feature information in current behavior chain is had occurred and that
Two characteristic informations, and within the effective time of the nonevent second feature information of corresponding behavior, to user's push and the behavior
The corresponding pushed information of second feature information without generation.
Furthermore it is also possible to be completed within effective time using other modes to the process of user's pushed information: user's
Behavioural information can be updated according to a certain period.The period can be period of change, such as certain website can be according to user
The time for accessing website, the behavioural information of user is updated, i.e., when a user visits a web site, it is meant that have new behavior
Information generates, then after preset time, all behavioural informations that user in user's this period accesses the webpage is collected, as new
Behavioural information be updated；Also the period can also be that fixed cycle, such as certain website can collect a user every other week and exist
Generated behavioural information when access website in this time.Behavioural information, can be according to working as every time during update
Preceding behavioural information obtains fisrt feature information corresponding with current behavior information, and true based on current fisrt feature information
Determine behavioral chain；The behavioral chain of the new determination can be identical with identified behavioral chain before, can also be with predetermined behavior
Chain is different；If newly determining behavioral chain is identical with behavioral chain before, redefined based on current behavioural information and the
The corresponding second feature information of one characteristic information, and according to the second feature information of the new determination calculate its undertaking person it is effective when
Between.
In information-pushing method provided by the embodiment of the present application, when carrying out information push, it to be first based on user
Behavioural information, obtain corresponding fisrt feature information, and corresponding with fisrt feature information behavior is obtained in subordinate act chain library
Chain.It include multiple second feature information in behavioral chain, and at least one of each second feature information and same behavior chain
Other second feature information have behavior continuous relationship, and then the second feature information in Behavior-based control chain pushes user
The push of information.In this process, push is pushed based on the behavioral chain pre-established, and behavioral chain itself is tool
There is the set of the second feature information of behavior continuous relationship, it, can be basic herein when currently one of behavior occurs for user
On know the subsequent issuable behavior of user, so as to just know user in advance before the subsequent behavior of user generates
Demand, the timely push of pushed information is realized, to avoid the lag of pushed information, to avoid having obtained in user demand
In the case where satisfaction, corresponding pushed information is just pushed to user.
Based on the same inventive concept, information push dress corresponding with information-pushing method is additionally provided in the embodiment of the present invention
It sets, since the principle that the device in the embodiment of the present invention solves the problems, such as is similar to above- mentioned information method for pushing of the embodiment of the present invention,
Therefore the implementation of device may refer to the implementation of method, and overlaps will not be repeated.
Shown in Figure 8, information push-delivery apparatus provided by the embodiment of the present application includes:
It is special to obtain corresponding with the behavioural information first for the behavioural information based on user for first determining module 801
Reference breath；
Second determining module 802, for obtaining row corresponding with the fisrt feature information from preset behavioral chain library
For chain；Each behavioral chain includes multiple second feature information, and second feature information and institute described in wherein at least one
It is identical to state fisrt feature information；Other second feature of at least one of each second feature information and same behavior chain letter
Breath has behavior continuous relationship；
Pushing module 803, for obtaining second spy in the behavioral chain corresponding with the fisrt feature information
Reference ceases corresponding pushed information, and pushes the pushed information.
Optionally, the behavioral chain further include: two second feature information with behavior continuous relationship are corresponding
Execute time interval；
The pushing module 803, is specifically used for: according to the fisrt feature information, the determining and institute from the behavioral chain
State the identical specific second feature information of fisrt feature information；Obtaining in behavioral chain has row with the specific second feature information
For the undertaking second feature information of continuous relationship, and the undertaking second feature information is located at the specific second feature information
Downstream；According to the specific corresponding behavior time of origin of second feature information and the specific second feature information and institute
It states and accepts the corresponding execution time interval of second feature information, calculate and accept second feature information corresponding effective time；Institute
It states in effective time, pushes the pushed information corresponding with the undertaking second feature information.
Optionally, second determining module, is specifically used for: from the acquisition goal behavior chain in the behavioral chain library；Institute
Stating has the second feature information identical with the fisrt feature information in goal behavior chain；Detection and the fisrt feature
Whether the identical second feature information of information is located at the end of the goal behavior chain；If it is not, then by the goal behavior chain
As behavioral chain corresponding with the fisrt feature information.
Optionally, shown in Figure 9, further includes: generation module 901, for generating multiple behavioral chains, and according to institute
The behavioral chain generated establishes the behavioral chain library.
Optionally, the generation module is specifically used for: the behavior sample information based on user, obtains and the behavior sample
The corresponding feature samples information of information；The feature samples information of each user has multiple；
The feature samples information is trained using the method for machine learning, obtains the institute with behavior continuous relationship
State feature samples information；
By according to the behavior continuous relationship between the second feature information, the behavioral chain is generated.
The computer program product of information-pushing method and device provided by the embodiment of the present invention, including store program
The computer readable storage medium of code, the instruction that said program code includes can be used for executing described in previous methods embodiment
Method, specific implementation can be found in embodiment of the method, details are not described herein.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description
It with the specific work process of device, can refer to corresponding processes in the foregoing method embodiment, details are not described herein.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. a kind of information-pushing method characterized by comprising
Behavioural information based on user obtains fisrt feature information corresponding with the behavioural information；
From preset behavioral chain library, behavioral chain corresponding with the fisrt feature information is obtained；Each behavioral chain includes
Multiple second feature information, and second feature information described in wherein at least one is identical as the fisrt feature information；Each
Other second feature information of at least one of the second feature information and same behavior chain have behavior continuous relationship；
The corresponding pushed information of the second feature information in the behavioral chain corresponding with the fisrt feature information is obtained,
And push the pushed information.
2. the method according to claim 1, wherein the behavioral chain further include: with behavior continuous relationship
The corresponding execution time interval of two second feature information；
The corresponding push of the second feature information obtained in the behavioral chain corresponding with the fisrt feature information
Information, and the pushed information is pushed, it specifically includes:
According to the fisrt feature information, determine that identical with the fisrt feature information specific second is special from the behavioral chain
Reference breath；
Obtain the undertaking second feature information in behavioral chain with the specific second feature information with behavior continuous relationship, and institute
It states and accepts the downstream that second feature information is located at the specific second feature information；
According to the specific corresponding behavior time of origin of second feature information and the specific second feature information with it is described
The corresponding execution time interval of second feature information is accepted, calculates and accepts second feature information corresponding effective time；
Within the effective time, the pushed information corresponding with the undertaking second feature information is pushed.
3. method according to claim 1 or 2, which is characterized in that it is described from preset behavioral chain library, obtain with it is described
The corresponding behavioral chain of fisrt feature information, specifically includes:
From the acquisition goal behavior chain in the behavioral chain library；Have and the fisrt feature information phase in the goal behavior chain
The same second feature information；
Detect the end whether second feature information identical with the fisrt feature information is located at the goal behavior chain；
If it is not, then using the goal behavior chain as behavioral chain corresponding with the fisrt feature information.
4. method according to claim 1 or 2, which is characterized in that it is described from preset behavioral chain library, obtain with it is described
Before the corresponding behavioral chain of fisrt feature information, further includes:
Multiple behavioral chains are generated, and the behavioral chain library is established according to the behavioral chain generated.
5. according to the method described in claim 4, it is characterized in that, the multiple behavioral chains of generation, specifically include: being based on
The behavior sample information of user obtains feature samples information corresponding with the behavior sample information；The spy of each user
Sign sample information has multiple；
The feature samples information is trained using the method for machine learning, obtains the spy with behavior continuous relationship
Levy sample information；
By according to the behavior continuous relationship between the second feature information, the behavioral chain is generated.
6. a kind of information push-delivery apparatus, which is characterized in that the device includes:
First determining module obtains fisrt feature information corresponding with the behavioural information for the behavioural information based on user；
Second determining module, for obtaining behavioral chain corresponding with the fisrt feature information from preset behavioral chain library；Often
A behavioral chain includes multiple second feature information, and second feature information described in wherein at least one and described first special
Reference manner of breathing is same；Other second feature information of at least one of each second feature information and same behavior chain have row
For continuous relationship；
Pushing module, for obtaining the second feature information pair in the behavioral chain corresponding with the fisrt feature information
The pushed information answered, and push the pushed information.
7. device according to claim 6, which is characterized in that the behavioral chain further include: with behavior continuous relationship
The corresponding execution time interval of two second feature information；
The pushing module, is specifically used for: according to the fisrt feature information, determining and described first spy from the behavioral chain
Reference ceases identical specific second feature information；Obtain with the specific second feature information there is behavior, which to accept, in behavioral chain closes
The undertaking second feature information of system, and the downstream accepted second feature information and be located at the specific second feature information；Root
According to the specific corresponding behavior time of origin of second feature information and the specific second feature information and described accept the
The corresponding execution time interval of two characteristic informations calculates and accepts second feature information corresponding effective time；When described effective
In, push the pushed information corresponding with the undertaking second feature information.
8. device according to claim 6 or 7, which is characterized in that second determining module is specifically used for: from described
Acquisition goal behavior chain in behavioral chain library；Have identical with the fisrt feature information described the in the goal behavior chain
Two characteristic informations；Detect the end whether second feature information identical with the fisrt feature information is located at the goal behavior chain
End；If it is not, then using the goal behavior chain as behavioral chain corresponding with the fisrt feature information.
9. device according to claim 6 or 7, which is characterized in that further include: generation module, it is multiple described for generating
Behavioral chain, and the behavioral chain library is established according to the behavioral chain generated.
10. according to the method described in claim 9, it is characterized in that, the generation module is specifically used for: the behavior based on user
Sample information obtains feature samples information corresponding with the behavior sample information；The feature samples information of each user
Have multiple；
The feature samples information is trained using the method for machine learning, obtains the spy with behavior continuous relationship
Levy sample information；
By according to the behavior continuous relationship between the second feature information, the behavioral chain is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437698.3A CN109961313B (en) | 2017-12-26 | 2017-12-26 | Information pushing method and information pushing device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711437698.3A CN109961313B (en) | 2017-12-26 | 2017-12-26 | Information pushing method and information pushing device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109961313A true CN109961313A (en) | 2019-07-02 |
| CN109961313B CN109961313B (en) | 2020-12-01 |
Family
ID=67022671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711437698.3A Active CN109961313B (en) | 2017-12-26 | 2017-12-26 | Information pushing method and information pushing device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961313B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114329196A (en) * | 2021-12-27 | 2022-04-12 | 杭州金线连科技有限公司 | Information pushing method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103440199A (en) * | 2013-08-27 | 2013-12-11 | 百度在线网络技术（北京）有限公司 | Method and device for guiding test |
| US20140380210A1 (en) * | 2013-06-21 | 2014-12-25 | Oracle International Corporation | Method and system for reconfiguring a multidimensional interface using dimension tiles |
| CN105868207A (en) * | 2015-01-21 | 2016-08-17 | 方正宽带网络服务有限公司 | Network resource pushing method and apparatus |
| CN106779788A (en) * | 2015-11-24 | 2017-05-31 | 北京易车互联信息技术有限公司 | Bus service information method for pushing and device |
| CN107436897A (en) * | 2016-05-26 | 2017-12-05 | 冯小平 | A kind of determination method and device of user's current behavior |
-
2017
- 2017-12-26 CN CN201711437698.3A patent/CN109961313B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140380210A1 (en) * | 2013-06-21 | 2014-12-25 | Oracle International Corporation | Method and system for reconfiguring a multidimensional interface using dimension tiles |
| CN103440199A (en) * | 2013-08-27 | 2013-12-11 | 百度在线网络技术（北京）有限公司 | Method and device for guiding test |
| CN105868207A (en) * | 2015-01-21 | 2016-08-17 | 方正宽带网络服务有限公司 | Network resource pushing method and apparatus |
| CN106779788A (en) * | 2015-11-24 | 2017-05-31 | 北京易车互联信息技术有限公司 | Bus service information method for pushing and device |
| CN107436897A (en) * | 2016-05-26 | 2017-12-05 | 冯小平 | A kind of determination method and device of user's current behavior |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114329196A (en) * | 2021-12-27 | 2022-04-12 | 杭州金线连科技有限公司 | Information pushing method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109961313B (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106803190A (en) | A kind of ad personalization supplying system and method | |
| CN107871244B (en) | Method and device for detecting advertising effect | |
| US20080141110A1 (en) | Hot-linked images and methods and an apparatus for adapting existing images for the same | |
| US11132703B2 (en) | Platform for providing augmented reality based advertisements | |
| US20140236706A1 (en) | System and method for measuring advertising effectiveness | |
| US8527623B2 (en) | User vacillation detection and response | |
| CN102289761A (en) | Metadata-enabled dynamic updates of online advertisements | |
| Hill et al. | Measuring causal impact of online actions via natural experiments: Application to display advertising | |
| KR100903499B1 (en) | Method for offering advertisement according to search intention segmentation and system for executing the method | |
| CN113269232B (en) | Model training method, vectorization recall method, related equipment and storage medium | |
| CN109961313A (en) | A kind of information-pushing method and information push-delivery apparatus | |
| CN115775163A (en) | Programmed advertisement putting method, device and medium based on big data | |
| JP2022531410A (en) | Digital anthropology and ethnographic system | |
| KR102148563B1 (en) | Word of mouth sales system and its method of online real estate | |
| Neumann | The power of big data and algorithms for advertising and customer communication | |
| Li | The Impact of the Live Delivery of Goods on Consumers’ Purchasing Behaviour in Complex Situations Based on Artificial Intelligence Technology | |
| CN111582901A (en) | Method, device and system for quantizing display link effect and storage medium | |
| CN110069695A (en) | Label processing method and device | |
| Gupta et al. | Click-through rate estimation using CHAID classification tree model | |
| Goldstein et al. | Are we there yet? Analyzing progress in the conversion funnel using the diversity of searched products | |
| CN114429362A (en) | Advertisement product delivery method and device, electronic device and readable storage medium | |
| Ravi et al. | Predicting consumer level brand preferences using persistent mobility patterns | |
| CN105981064A (en) | Matching system, matching method, and non-transitory computer-readable medium on which program is stored | |
| Groth et al. | Granularity creates added value for every objective | |
| KR20090106198A (en) | Method and apparatus for advertisement campaign product assessment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |